The contribution of accessible social media data to the work of tourism research communities: a bibliometric network analysis of studies using Flickr data

The contribution of accessible social media data to the work of tourism research communities: a bibliometric network analysis of studies using Flickr data | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article The contribution of accessible social media data to the work of tourism research communities: a bibliometric network analysis of studies using Flickr data Bálint Kádár, Márcio Ribeiro Martins This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7072228/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 11 You are reading this latest preprint version Abstract This paper explores how Flickr’s geotagged photos have contributed to the development of new research topics in tourism studies, particularly through the use of large-scale, freely available datasets. A systematic literature review and bibliometric network analysis were conducted using 333 Scopus-indexed papers that included “tourism” and “Flickr” in abstracts or keywords. An additional 519 papers citing this core set were identified. Network analysis using Gephi applied a gravity model and clustering algorithms to detect citation-based communities. Content analysis of highly cited papers helped define key research themes. Seven research clusters emerged, focusing on nature-based tourism, tourist activities by space-time behaviour, destination attractiveness, image development, travel route detection and recommendation, and machine learning for content analysis. These communities reflect the global academic interest enabled by Flickr’s open API, allowing reproducible and comparative analyses across destinations. This study highlights how a unified, open-access social media dataset has catalyzed the formation of global research communities in tourism. Unlike newer platforms with restricted access, Flickr’s openness fostered methodological innovation and deepened field-specific knowledge in tourism research. A critical analysis of existing work was provided, highlighting overlooked areas and synthesising insights to propose new directions for research. tourism studies bibliometric analysis Flickr open API network analysis Figures Figure 1 Figure 2 Figure 3 Figure 4 Introduction Much of tourism studies, like tourism geographies, rely on quantitative methods requiring travel-related data. The richness of data sources bought a wide variety of methods and a fragmentation of theory. Official statistical data on accommodation occupancy, air flights and border crossings are the globally available resources that give a unified basis for tourism researchers, but such data has low geographical and contextual accuracy. Main attractions have visitor data from ticket selling, but most tourists’ journeys remained hard to trace, if not with manual surveying methods, like questionnaires. Until the widespread use of social media, only large corporations had massive geographical data on the geographical consumption of travelling users; these are mainly payment card services and mobile phone operators. These corporations make commercial use of their datasets. Therefore, they rarely let independent researchers use their data, even though fruitful collaborations with privileged or well-funded scientific communities did result in some very relevant papers (Ahas et al., 2007 ; Sobolevsky et al., 2014 ). Other researchers had to rely on traditional surveys (Hwang & Fesenmaier, 2003 ), specialized digital apps collecting data from a very limited number of tourists (Dickinson et al., 2014 ), or data loggers distributed to tourists (Mckercher et al., 2012 ) – also a very resourceful data collection method. Researchers did explore every opportunity that new services or technologies offered (Shoval & Ahas, 2016 ), but such methods are hard to reproduce by other groups of scientific community. Therefore tourism studies, especially geographic and other quantitative studies, seem to be an endless production series of isolated case studies (Ashworth & Page, 2011 ). Social media, as the evolution of World Wide Web services, promised an open stream of data created by users, and shared by all other users of the internet. Suddenly, text, images, videos and geolocated content produced by individual users became available in unseen quantities, accessible to all and therefore, to researchers as well. The “Data Golden Age” in digital research lasted for a decade until the Cambridge Analytica (CA) scandal of 2018. Until then, most of the Application Programming Interfaces (APIs) of social media services provided easily downloadable data, meant to ease the communication between social media and external applications using it for commercial purposes, but available also for academic researchers. As CA used sensitive user data mined by scientists for unethical commercial use, Facebook and most large service providers shut down the openly accessible features of their APIs, causing a difficult situation for independent academic researchers mining such datasets (Perriam et al., 2020 ; Tromble, 2021 ). Tourism studies is one of the fields that benefitted from the open APIs of social media. Tracking tourists and analyzing destinations and tourist behaviour from travel-related images uploaded to Panoramio (launched in 2005, discontinued in 2016), Flickr (2004-) and Instagram (2010-) became established research methodology (Ashworth & Page, 2011 ; Shoval & Ahas, 2016 ). Today large Chinese social networks allow access to the metadata uploaded by their users, while among Western providers, only Flickr maintains its open-source API policy. This contributes to the growing imbalance between the West and the East regarding the access to and use of this type of information in tourism scientific research. Flickr’s popularity faded since it was the most used image-sharing platform in the 2010s. The new generation of social media is owned by corporations (Facebook, Google) even larger than the mobile phone operators or credit card companies, and these handle the immense datasets of their users as their own resources they capitalize on, excluding research communities from analytics. This leads to an unbalanced status in global knowledge, especially true for behavioural sciences and tourism studies, as large enterprises nowadays have a better understanding of travel patterns and user behaviour than academic science has. These datasets enabled researchers worldwide to map, track, and interpret tourist movements and preferences at unprecedented scales. However, with the advent of stricter data privacy regulations and the subsequent APIs restrictions, the access to such open datasets has diminished, creating an urgent need to understand the legacy and future possibilities of social media-derived data for tourism research. What do we miss in tourism studies by losing access to easily researchable social media data? It is more straightforward to analyze what the scientific field gained with large unified datasets available for tourism research, even if the new social media generation could have had even more massive data. Panoramio, Flickr, and, to some extent, Instagram - together with the large Chinese social media sites such as Weibo, providing data mostly from China – provided travel related data in unified formats from millions of users. The advantages of these resources for quantitative research lie not only in the immense amounts of georeferenced images from most destinations, but also in the unified format of such data, leading to interrelated research methodologies and more relevant cross-references between scholars from around the globe in the fields of tourism studies. This study addresses the following central research question: how has access to Flickr's open georeferenced photography data shaped the evolution and diversification of research communities in tourism studies, and what lessons can inform future research strategies in an era of restricted data access? In doing so, it critically evaluates the academic significance of open-source data in tourism research and proposes new directions to maintain research vitality in a post-open-data era. Aiming to prove the benefits of a unified global dataset of georeferenced travel photography for the advancement of scientific fields and theories, the central hypothesis is that a significant proportion of published papers in tourism studies, which utilise Flickr as their primary database, are interconnected. These studies build upon each other’s methodologies and findings, thereby fostering the development of new areas within tourism studies. It is argued that such progress would have occurred at a considerably slower pace without the widespread availability of user-generated content (UGC) on a global scale. Therefore, this systematic and bibliometric review intends to document and synthesize academic research published on tourism studies based on the Flickr social media database, identifying the research fields and results that could emerge in different fields because of the availability of UGC in the form of travel photography. Social media and Flick in tourism studies Technological advances and the progressive application of a varied set of tracking techniques in tourism research have made it possible to collect high-resolution georeferenced data about tourists' space-time behaviour, which can be worked on in a GIS (Geographic Information Systems) environment that explore and interpret increasingly large and complex databases (Martins & Costa, 2022 ). Social networks, such as Twitter, Instagram, Panoramio or Flickr, offer an enormous quantity of geotagged photos “through which a user provides spatial (latitudes and longitude coordinates) and temporal (date and time of day) information in the form of coordinates” (Giglio et al., 2019 ). Through the geotagged photos, it is possible to find the sequence of locations visited (Höpken et al., 2020 ) or to provide the best travel routes between popular travel destinations, minimizing the distance while including maximal tourism popularity (Sun et al., 2015 ). The photographs posted by visitors also suggest people’s interests and preferred activities while travelling, revealing where the users have been (Giglio et al., 2019 ). The main advantage of using social media with geotagged photos in tourism studies is the possibility of using a very large sample where the places can be related to tourist experiences (Caldeira & Kastenholz, 2020). It is not an intrusive method, and the large amount of data collected can be analyzed with less effort in data processing. It is also a low-cost technique, and the unlimited scope of study in terms of geographical scale (local, regional, national and global) makes it an interesting database for tourism research (Martins & Costa, 2022 ). Some disadvantages have also been identified, namely, the potential bias of the sample because only the platform users are tracked, and they also tend to share the most impressive photos and not those of more common environments (Caldeira & Kastenholz, 2020). As social media has become omnipresent among travellers (Curlin et al., 2019 ), tourism researchers have been using it to understand visitors’ spatiotemporal behaviour (Hruška & Pásková, 2018 ; Önder et al., 2016 ), also trying to understand the role of this type of social media in tourists’ travel behaviour patterns (Giglio et al., 2019 ; Paldino et al., 2015 ). Photo analysis in social media has been used in tourism research to estimate visitor trajectories (Mor & Dalyot, 2020 ), to find provenance and patterns of recreation (Sinclair et al., 2020 ) and tourist behaviour patterns (Yang et al., 2017 ), to propose or assess tourist routes (Kurashima et al., 2013 ), to identify social events such as parades, protests, sports, festivals (Yeran & Fan, 2014 ), to measure the tourist activities in cities (Kádár, 2014 ), natural areas (Wood et al., 2013 ), or larger regions (Girardin et al., 2007 dár & Gede, 2021; Önder et al., 2016 ), to identify areas with a concentration of tourists (Kádár & Gede, 2013 ) or to determine the attractiveness of various tourism sites (Giglio et al., 2019 ) among others. Most of these studies relied on the datasets of Flickr, as this service provider always kept its API open for external researchers. Flickr was developed by Ludicorp, a Vancouver, Canada-based company founded in 2004 by Stewart Butterfield and Caterina Fake. According to Broz ( 2022 ), it is one of the most popular sites worldwide, with around 25 million photos uploaded to Flickr every day, ranking top 400 globally, top 290 in the USA, and top 2 in its category – photography. In February 2017, the site hosted approximately 13 billion photos from 122 million users that come from 72 countries (USA, 31.03%, UK, 9.83% and Germany, 5.26%), getting up to 60 million visits per month. Even if male users and well-educated young and middle-aged people (60.73%) are slightly overrepresented, and older age groups are slightly underrepresented, it is still the platform with one of the most balanced social distributions of users (Kádár & Gede, 2013 ). Methodology This paper was based on the research of scientific documents available on the Scopus database concerning the research trends, patterns and the research gaps on the Flickr usage in tourism research, until 15th September 2023. The search and collection of information were carried out in the period 2010–2023 using the scheme code “tourism” or “tourist” and “flickr” or "geotagged photo" contained in titles, abstracts, and keywords. Scopus database returned 365 documents in the initial search and through a selection process all books and chapters (n = 11), editorials, notes, letters (n = 12) and all documents not written in English (n = 9) were excluded. Finally, 333 articles were selected for bibliometric and content analysis, and all the references cited in the papers have been analyzed. References have been downloaded as a continuous text field for each paper. Therefore, a simple algorithm had to be written in order to create single data entries for all papers referenced by the source papers. All papers have been given unique names following the APA citation style. 12986 citations have been analyzed, but 8702 connections have been discarded as these had only single occurrences, meaning that the paper cited did not appear to be cited by any other papers. 897 papers have been kept, with at least 2 citations in the pool, resulting in a total of 4284 connections between papers. Such methodology allowed for the analysis of a larger ecosystem of the research community working with the topic of social media photos and tourism, including all papers that influenced the work of those in the original query. After this first stage of database creation the 2nd stage of bibliometric analysis and the 3rd stage of content analysis followed (Fig. 1 ). Bibliometric methods were used through a co-occurrence analysis, allowing a systematic, transparent, and reproducible review process (Eck & Waltman, 2016). The performance evaluation combined with science mapping would allow the identification and visualization of research fields (van Eck & Waltman, 2017 ). While most similar studies used VOSviewer (Herrera-Franco et al., 2020 ), this bibliometric analysis needed more customization and control in building and analyzing the initial dataset, while in-depth content analysis of the papers needed more than just keyword occurrence analysis. Therefore Gephi 0.10.1 software was used to construct and visualize the bibliometric networks related with Flickr and tourism, aided by data processing in excel. A network consisting of 897 nodes (all papers downloaded from Scopus related to tourism and Flickr, plus all papers at least 2 of these have cited) and 4284 edges (all citations) have been created in Gephi 0.10.1. 564 nodes only have in-degree, as these were not part of the original pool, and their citations have not been analyzed, while at least 2 papers from the original 333 did cite them (max = 47). 223 nodes only have out-degree, meaning that these papers have not been cited by any other papers in the pool, but they have cited others. These were papers from the original source pool of Scopus containing the keywords, many of which are from recent years, making it impossible to cite. 110 papers have both in- and out-degree; these are supposed to be the most relevant papers in the network. In Gephi ForceAtlas 2, a network visualization showing a linear attraction gravity model between nodes has been used (Jacomy et al., 2011 ), showing which papers are interconnected between each other’s by citations. Figure 2. The bibliographic network of papers related to tourism and social media images (compiled in Gephi 0.10.1.) Among their position, the in-degree (number of citations from the network to the node representing a paper) and out-degree (number of citations from the paper represented by the node to other nodes in the graph) are the metrics most relevant describing the singular nodes. In Fig. 2 the size of the nodes reflects their in-degree, while in the darkness of their color correlates with higher out-degree. High out-degree papers are the ones that build most on the results of other works represented in the graph, therefore these papers were worth to be content analyzed in stage three. These papers should show most accurately the commons among topics and methodologies used in research communities represented by different sections of the graph. High in-degree papers are the most influential ones, defining the topics or methodologies used in many other papers of the research communities. Therefore, also these were worth to be analysed in the third stage. The connections – citations – between papers have been coloured to show which year the work cited had been published. Therefore, the temporal development of the network could be visualized. The earliest papers cited by more than one author are from the beginning of the ‘1960s (mathematical algorithms), and it is well visible that separate clusters in the network have different timelines in their activities. To define the different communities with similar topics or methodologies of research, a cluster analysis has been applied to the network using the community detection method of Blondel et al. ( 2008 ) integrated into Gephi. We assumed that papers focusing on different fields of study can be detected as different clusters of the network, having the most citations between each other (Fig. 3). As a third stage, the content analysis of the papers was made to show the specific scientific field developed in the revealed clusters. SciSummary, an AI-based tool developed to summarize scientific papers, was used to have an overview of all papers in the network. The assumption that papers cited by high out-degree papers or citing high in-degree papers inside a cluster follow the same topic as such high-degree papers has been largely verified; therefore, cluster detection and the manual analysis of high-degree nodes proved to be an effective way to define the topics present in a large bibliographic system. High in-degree papers are the most influential ones, delivering the bases of the methodologies or topics of the cluster, while high out-degree papers best summarize the overall topic of the cluster, as they gather in their references most of the influential works defining the cluster. Results Bibliometric Analysis to define tourism research communities The dataset of nodes consisting of 333 source papers and 564 papers cited by at least two source papers and of edges consisting of 4284 citations in total have been uploaded to Gephi 0.10.1, and analyzed. The resulting network visualized with a ForceAtlas 2 algorithm (Fig. 3) already showed strong clustering with large sub-graphs concentrated around some highly cited papers, distancing from the main network in structure but also in the average timeline of its publications. Figure 3. The 7 clusters found with Gephi in the bibliographic network (compiled in Gephi 0.10.1.) The cluster analysis (Blondel et al., 2008 ) was made with a resolution of 0.8, resulting in a modularity score of 0.512, with 7 distinct communities with more than 50 nodes (Fig. 4 ). It must be noted, that the clustering algorithm never gives two identical dispositions in this graph, but modifying the resolution above 1.1 and below 0.8 always resulted in modularity scores Q < 0.5, while inside this threshold this score was 0.5 < Q < 0.515 (Q = 0 is no separation at all, Q = 1 is total separation to another network). Selecting the lower resolution while still maintaining a high modularity leads to the detection of communities that have strong interrelations but are more hidden than the basic four that one can detect just from the primary form and structure of the graph. Figure 4 shows the weighted network representation of the 7 clusters, with the weighted in- and out-degrees (Kádár & Gede, 2021 ). References inside the clusters are represented as self-loops, and the 5 large clusters have significantly more self-references than in- or out-degrees. However, C4 has almost as many references from other clusters as inside references, while in the case of C6, both in- and out-degree are higher than inside references. In fact, these two clusters have not been detected with the resolution = 1, as their separation from the rest is not as obvious, but still the references to - and from - other clusters remain much lower one by one than the number of inside self-references. It can be seen that C1, as the central core cluster, is the most connected to others, especially to C2, while C7 is the most isolated from others. The 7 communities have been analyzed. Specifically, the papers with higher in-degree and out-degree values have been content analyzed, and therefore, seven different research topics or methodologies have been identified. Analysis and Discussion C1 The community measuring tourist activities by spatiotemporal analysis The graph is structured around a central cluster (C1; n = 197), defined by many strong nodes like Girardin, Dal Fiore, et al. ( 2008 ), Vu et al. ( 2015 ), García-Palomares et al. ( 2015 ), Zheng et al. ( 2012 )d dár) (2014). This is the most fluid cluster as nodes at its boundaries tend to be tied equally strongly to the neighbouring clusters; in fact most of the other clusters have their most influential nodes also connected to this central cluster. This fluid cluster is in the centre of the graph because the topics defined in it correlate most to the original topics searched in the Scopus database: tourism, geo-tagged photos and Flickr. The earliest influential work in this cluster is by Girardin and colleagues (Girardin, Calabrese, et al., 2008 ; Girardin, Dal Fiore, et al., 2008 ), who first presented a comprehensive methodology to define tourist movements in different territories and tourist typologies deriving from the spatiotemporal position of geo-tagged Flickr photographs. Most papers in this central community compare the behaviour of different user groups regarding the tourism performance of various locations (Table 1 ). Their themes are connected by methodology and the ambition to extract tourism-related statistics from geo-tagged photos, but not by a specific research focus, as seen in the other clusters. Table 1 Tourist activities by spatiotemporal analysis community Paper Aim In-/Out-degree Theme/topic (Girardin, Dal Fiore, et al., 2008 ; Girardin et al., 2009; ) Mining travel patterns from geotagged photos, differentiating between tourists and locals 41/0 10/0 Measuring tourist activity Differentiating user groups (Zheng et al., 2012 ) Mining travel patterns from geotagged photos 39/17 Measuring tourist activity Differentiating user groups (Vu et al., 2015 , 2018 ) Exploring the travel behaviours of inbound tourists to Hong Kong using Flickr geotagged photos (2015), Travel Diaries Analysis by Sequential Rule Mining (2018) 38/29 6/30 Urban tourism Measuring tourist activity Differentiating user groups (García-Palomares et al., 2015 ) Identification of tourist hot spots based on social networks: A comparative analysis of European metropolises using photo-sharing services and GIS 33/0 Urban tourism Measuring tourist activity Attractive places in cities (Kádár, 2014 ) Measuring tourist activities in cities using geotagged photographs 23/17 Urban tourism Measuring tourist activity Differentiating user groups (Hu et al., 2015 ) Extracting and understanding urban areas of interest using geotagged photos 23/0 Urban tourism Measuring tourist activity Attractive places in cities (Li et al., 2018 ) Analyzing and visualizing the spatial interactions between tourists and locals: A Flickr study in ten US cities 20/18 Urban tourism Measuring tourist activity Differentiating user groups (Yuan & Medel, 2016 ) Characterizing international travel behaviour from geotagged photos: A case study of Flickr 14/18 Measuring tourist activity Differentiating user groups (Önder et al., 2016 ) Tracing Tourists by Their Digital Footprints: The Case of Austria 12/18 Measuring tourist activity Differentiating user groups (Straumann et al., 2014 ) Towards (re)constructing narratives from georeferenced photographs through visual analytics 10/16 Measuring tourist activity Image content analysis The studies in C1 reveal that the temporal dynamics have been underexplored, i.e., most studies treat Flickr data as static rather than capturing evolving tourist flows. Therefore, some future research possibilities emerge, such as leveraging longitudinal Flickr datasets to analyse temporal evolution of tourist patterns (e.g., pre- and post-COVID impacts). While this cluster pioneered the large-scale mapping of tourist flows (e.g., Girardin et al., 2008 dár, 2014), it often treated tourists as homogenous agents. Using the profile data of Flickr users, future research should incorporate socio-demographic segmentation to better reflect the diversity of tourist experiences and motivations. Additionally, integrating sentiment analysis with spatiotemporal data could reveal not just movement patterns, but emotional geographies of tourism. C2 The travel route detection and clustering community The large community consisting of 195 papers can be considered one of the oldest as most work ranges from 1995 to 2019, with only 8 papers from the 2020’s. However, the main topic in the cluster – travel route recommendation – continues with newer works in the C3 community, much connected to this one (Table 2 ). The earliest works cited in this community (some from the ’1960s and ’70s) are related to clustering algorithms. Based on the work of Ester et al. ( 1996 ), many authors use the DBSCAN (Höpken et al., 2020 ; Miah et al., 2017 ; Zheng et al., 2012 ) or similar algorithms. The creation of travel route recommendation systems seemed to be the first practical outcome of researching geo-tagged social media photos, but large internet companies e.g. Google, could develop such services within their databases and closed company profiles; this is also a reason why this topic doesn’t have contemporary references. In fact, the travel route recommendation seems to be the most connecting topic, but in reality, the clustering of tagged photos, either by textual tags, geo-tags or image content, is the methodology that connects these works. Image content detection methods are also used mostly in this cluster, citing e.g. Lowe ( 2004 ) for methodology. The early works of Kennedy et al. ( 2007 ) do not mention travel route recommendation, only clustering methods to describe places, just as the very influential work of Kisilevich et al. ( 2010 ), also working with DBSCAN. Most papers cite Crandall et al. ( 2009 ), who employed a non-parametric clustering method named mean-shift to discover the significant landmarks/hotspots at a global level, providing a quick overview of the interesting places at travel destinations and demonstrating that visual and temporal features improved the ability to predict the photo location, compared to using textual features alone. The interplay between structure and content makes this paper (“Mapping the World’s Photos”) a very influential reference on tourist recommendation systems, travel route recommendation, or Points of Interest (POI’s) identification/recommendation. Table 2 Travel route detection and clustering community Paper Aim In-degree /Out-degree Theme/topic (Crandall et al., 2009 ) Personalized urban trip recommendations for tourists based on user interests, points of interest, visit durations, and visit recency. 49/0 Travel route recommendation (Kisilevich et al., 2010 ) Clustering automatically popular places. 49/0 Travel route detection (Ester et al., 1996 ) Clustering algorithms in theory. (no Flickr at all). 34/0 Clustering algorithms (Kurashima et al., 2010 ) Travel recommendation itineraries based on present location and preferences. 23/18 Travel route recommendation (Majid et al., 2013 ; Memon et al., 2015 ) Travel recommendation system based on previous visits. 23/18 16/5 Travel route recommendation (Popescu et al., 2009 ) Urban trip extraction with inner stays (museum) and panoramic points: how long is a stay/walk? 23/6 Travel route detection (Sun et al., 2015 ) Road-based travel recommendation. 22/30 Travel route recommendation (Kurashima et al., 2013 ) Travel route recommendation using geotagged photos. 17/22 Travel route recommendation (Zhou et al., 2015 ) Detecting tourism destinations using scalable geospatial analysis based on cloud computing platform. 15/18 Travel route detection (Yang et al., 2017 ) Quantifying tourist behaviour patterns by travel motifs and geo-tagged photos from Flickr. 11/24 Travel route detection This cluster advanced route optimisation techniques using clustering algorithms like DBSCAN, and rarely engaged with the cultural implications of algorithmically proposed itineraries. A predominantly computational focus led to a technocentric view of travel, neglecting how tourists subjectively experience routes, serendipity, and intangible elements like safety perception or urban ambience. Therefore, the routes are optimised for efficiency, neglecting tourists’ desire for exploration and cultural immersion. Future research could combine Flickr photos with natural language processing (NLP) analysis of photo descriptions/titles to incorporate emotional and experiential dimensions into route recommendations. Regarding the dynamic route prediction, with the advances in computing power capacity, it would be interesting to use real-time or near-real-time Flickr uploads to adjust recommended routes based on emerging visitor flows. C3 The travel route recommendation and orienteering community As opposed to C2, all papers in C3 (n = 113) work on travel recommendation methods, with a methodological focus on the Orienteering Problem (Lim et al., 2018 ) using socially generated knowledge and focusing on user metadata where the problem was modelled as a single graph of POIs, their popularities and the transit times among them (Table 3 ). In fact, the most representative researcher is K. H. Lim, who co-authored 8 papers in this cluster. De Choudhury et al. ( 2010 ) authored the most cited and earliest paper, the one connecting this community most to C2: they studied and proposed a tour recommendation using the orienteering problem, which starts with a particular POI and ends with a different POI, while ensuring that the tour can be completed within a certain time. The orienteering problem is the methodological basis connecting all work in this community, all proposing effective tour recommendation algorithms. Table 3 Travel route recommendation and orienteering community Paper Aim In-/Out- degree Theme/topic (De Choudhury et al., 2010 ) Construct intra-city travel itineraries modelled as a single graph of POIs automatically using Flickr photos. 38/7 Personalized trip recommendation; Orienteering Problem (Lim et al., 2015, 2018 ) Personalized tour recommendation based on user interests, points of interest, visit durations and recency 7/44 7/22 Personalized trip recommendation (Sarkar et al., 2020 ) To recommend multiple urban itineraries based on the tourist’s interest, the popularity of itineraries and the cost associated with these itineraries 5/32 Travel itineraries recommendation (Mou et al., 2022 ) To propose tourist route recommendations using a personalized recurrent neural network (P-RecN) in Shanghai.. 0/37 Travel route recommendation (Korakakis et al., 2017 ) To automatically construct travel routes, i.e., ordered visits to various places-of-interest in Greek cities. 0/36 Travel route recommendation (Mor et al., 2021 ) To generate tourism walking routes in New York and evaluate them in terms of the visible space. 0/27 Evaluating the attractiveness of Tourism Routes (Lim, 2016 ) Developing algorithms for recommending personalized city tours to both individual travellers and groups of tourists based on their interest preferences (based on Flickr and Wikipedia) 0/24 Personalized tours recommendation (Mor & Dalyot, 2020 ) Using Flickr photos as a source for mining users’ trajectories to compute walking tourism routes effectively based on tourist activities in Tel-Aviv, Israel and Manhattan, USA. 0/24 Tour recommendation system Highly technical, C3 modeled the tourism experience as an optimization problem. However, the assumption that the ‘optimal’ itinerary is universally desirable ignores heterogeneous tourist profiles (family tourists, solo adventurers, elderly travellers, among others), and often neglects qualitative dimensions of travel, such as serendipity and personal meaning-making. Future advances should blend algorithmic efficiency with models that account for experiential richness and tourist agency, and develop segmented itinerary models reflecting diverse tourist personas, using Flickr data enriched by inferred user profiles (e.g., based on metadata about travel duration, locations visited). C4 The community measuring the attractivity of places C4 is the smallest community (n = 50) around various works of I. Bojic. The most influential paper of this researcher was co-authored with Paldino et al. ( 2015 ) entitled “Urban Magnetism through the Lens of Geo-Tagged Photography”. In fact, the attractiveness of places is the main topic of this cluster, often combining Flickr data with other more qualitative data sources to explore why some places are more attractive than others (Table 4 ). Table 4 Attractivity measure of places community Paper Aim In-/Out-degree Theme/topic (Paldino et al., 2015 ) Measuring the attractiveness of 10 different cities using Flickr Photos from residents and tourists, tracing the hotspots and mobility networks. 17/15 Attractivity of places in cities (Karayazi et al., 2021 ) Understanding the heritage attractiveness in Amsterdam by combining clusters of attractions from Flickr with heritage data and applying regression analyses to identify influencing factors. 3/14 Attractivity of heritage in a city (González et al., 2008 ) (a) Analysing the trajectory of 100,000 anonymized mobile phone users for a six-month period, showing simple patterns. 11/0 Tracking travel patterns through mobile phone data (Hawelka et al., 2014 ) (a) Analysing geo-located Twitter data for global mobility patterns 11/0 Tracking travel patterns through Twitter data (Valls & Roca, 2021 ) Measuring visitor hotspots in Barcelona, employing visualization methods on different scales 0/24 Attractivity of places in cities (Bojic et al., 2016 ) Investigating the attractiveness of countries' large-scale composite regions using Flickr data from foreign visitors and migration data from the UN. 0/16 Attractivity of regions (Jain & Singh, 2017 ) Investigating the attractiveness character of most checked-in countries and large-scale composite regions 0/16 Attractivity of regions (a) Articles where Flickr is not used but with relevant methodological contributions. This cluster revealed key insights into the magnetism of cities and regions. However, most studies were cross-sectional, capturing attractiveness at a single moment in time. Longitudinal analyses that trace how attractiveness evolves (especially in response to marketing, climate change, or socio-political shifts) remain a promising avenue for deeper understanding. Besides that, seasonal and event-based fluctuations in attractiveness are also under-researched. C5 The destination image research community This cluster (n = 95) is well separated from the central core and has the most relatively old papers from the ‘1990s and before (Fig. 3), organized around different influential works (Deng et al., 2019 ; Donaire et al., 2014 ; Stepchenkova & Zhan, 2013 ; K. Zhang et al., 2019 ). Most work in this cluster analyzes the image of destinations (Bhatt & Pickering, 2022 ; Deng et al., 2019 ; Lin et al., 2021 ; Mangachena et al., 2022 ; K. Zhang et al., 2020 ). One of the most central and influential papers by Stepchenkova & Zhan) (2013) compared images of Peru collected from a DMO’s website and Flickr, comparing how projected destination image(s) used in marketing match the tourist’s destination image. The qualitative research to collect tourists’ attitudes, opinions and emotions about the destinations using mainly quantitative databases of online photos is complex (Lin et al., 2021 ) because the meanings embedded in it is subjective to researchers’ interpretation. Therefore, this cluster has the most references to works related to the understanding of tourist photography as a phenomenon (Jenkins, 2003 ; Pan et al., 2014 ; Urry, 1990 ) and destination image formation (Garrod, 2009 ), explaining the high number of older references, all rooted in the qualitative analysis of destinations (Table 5 ). To analyse destination image (DI), content analysis of the photos is a must. K. Zhang et al. ( 2019 ) start it with the quantification of specific themes and attributes presented in the images, followed by the identification of the main focal items as well as their frequencies, co-occurrence, clustering, and other related issues to be recorded, as done also by Stepchenkova & Zhan ( 2013 ). In this diverse cluster, therefore, there are many older papers giving theoretical background, newer ones using content analysis and even others using machine learning techniques, are mixed. The Flickr related research topics can be retrieved from the content of the newer papers in the cluster. Table 5 Destination image research community Paper Aim In-/Out- degree Theme/topic (Donaire et al., 2014 ) Identifying the attributes of photos and determining the existence of four different photographer's behaviour 20/27 Destination image Content and cluster analysis of online photographs (Zhang et al., 2019 ) Using deep learning technology, the visual contents of photos were identified, and the perception and behavioural preferences of tourists from different countries were analyzed. 17/39 Visual content analysis Tourists' behaviours and perception (Stepchenkova & Zhan, 2013 ) Comparing images of Peru collected from a DMO’s website and Flickr, identifying statistical differences in several dimensions. 18/16 Content analysis of online photographs Destination image (Deng & Li, 2018 ) Using Flickr photos of New York City visitors to propose and implement a machine learning-based model to rank photos describing a specific theme from viewers’ perspective. 10/26 Destination image Content analysis of online photographs (Deng et al., 2019 ) Proposing a new method to compare destination image (DI) differences among inbound tourists in Shangai, China. 8/38 Destination image (K. Zhang et al., 2020 ) Using user-generated photos and artificial intelligence computer vision technologies to identify the differences in the perceived destination image and behavioural patterns between residents and tourists in Hong Kong. 7/32 Destination image Computer vision (Bhatt & Pickering, 2022 ) Exploring the uses of Flickr photos posted by Chitwan National Park, (Nepal) visitors, comparing the content of those photos (perceived image(s)) with those posted by tourism organizations online (projected destination image(s)). 0/43 Content analysis of online photographs Destination image (Lin et al., 2021 ) Using Flickr photos to investigate the unique merits and biases of social media analytics (SMA) and a traditional visitor intercept survey 0/29 Destination image (Content analysis, machine learning and text analysis) (Mangachena et al., 2022 ) Assessing how Africa is presented in images seen by tourists and how Africa is represented in destination image(s). 0/29 Destination image The use of Flickr to contrast perceived vs. projected destination images was a methodological innovation. However, studies often failed to theorize how digital images mediate memory and perception. Future research should build interdisciplinary bridges to visual studies and cognitive science to interpret photographic tourism narratives better. In addition, there is a risk of reifying stereotypes, i.e., repetitive photographic patterns (e.g., iconic landmarks) may reinforce narrow destination images, limiting tourists’ perceptions of diversity. Future research should deepen the use of sentiment analysis to photo captions and tags to complement visual analysis, uncovering affective dimensions of destination image, and explore how Flickr users depict less-promoted or alternative aspects of destinations, challenging the dominant projected images. C6 The community using deep learning methods C6 is a smaller cluster (n = 63) dispersed between C1, C2, and C5, found only by lower resolution settings in the clustering algorithm, represented somehow by the recent work of Cho and Kang (Cho et al., 2022 ; Kang et al., 2021 ). This is a community of researchers connected not by a specific research topic, but by methodology, where all work uses machine learning algorithms trained by and working with Flickr databases, but along different topics (Table 6 ). The nodes of this cluster are spread around the centre of the graph, connecting nearly to all other clusters in the topic, but interconnected inside the community by methodology – training machine learning algorithms with Flickr photos to detect and classify the content of photos. Some papers do exist outside of this cluster using machine learning technology, but those are more strictly connected to the topic of other clusters. Therefore, they do not appear in this cluster, where the methodological aspects of machine learning prevail over the thematic similarities. Table 6 Deep learning methods community Paper Aim In-degree /Out-degree Theme/topic (Hollenstein & Purves, 2010 ; Kisilevich et al., 2013 ; Rattenbury et al., 2009) Early methods to get the semantics of places from geotagged photography before using machine learning 17/0 15/0 10/0 Tags associated with Flickr images as a proxy for empirical data (Krizhevsky et al., 2012 ) ImageNet classification with deep convolutional neural networks 6/0 Deep convolutional neural network (Chen et al., 2019 ) Presenting several data processing methods for inbound tourist flow prediction in Beijing (China) validated through data correlation analysis and machine learning algorithm predictions 5/25 Machine learning algorithm Tourist flow prediction (Kim et al., 2020 ) Analyzing representative images and elements of sightseeing attractions using photos uploaded on Flickr by Seoul tourists,using convolutional neural networks. 5/12 Deep learning Photo’s characteristics analysis (Cho et al., 2022 ) Classifying Tourists’ Photos and Exploring Tourism Destination Image Using a Deep Learning Model 0/37 Convolutional Neural Network (Kang et al., 2021 ) Classifying a large volume of Flickr photos with transfer learning of a deep learning model for exploring tourists’ urban images. 0/29 Convolutional neural network Tourists’ photo classification (Zhu et al., 2019 ) Using fine-grained land use classification at the city scale using ground-level images to better understand what occurs in different parts of a city at fine spatial and activity class scales. 0/21 Convolutional neural networks (Lee et al., 2015 ) Predicting geo-informative attributes in large-scale image collections using convolutional neural networks 0/14 Deep convolutional neural networks (Novack et al., 2020 ) Detecting building facades with graffiti artwork based on street view images interpreted by a customized, convolutional neural network 0/11 Convolutional neural networks While machine learning enhanced the scalability of image content analysis, it also introduced a black-box problem, making it difficult to understand how classifications are made. Studies such as Zhang et al. ( 2019 ) highlight this limitation and suggest the need for greater transparency, interpretability, and bias auditing in the models used, especially given the sociocultural implications of automated analyses. The convolutional neural networks (CNNs) also deliver impressive classification results but rarely offer insight into how visual features map onto tourism concepts (e.g., what makes a photo depict ‘authenticity’?). C7 The nature-based tourism research community The large cluster (n = 184) that separates most clearly from the others consists of many quite recent works (Fig. 3) and is unequivocally dominated by the work of Wood et al. ( 2013 ). This influential paper defines the topic of this community in its title: “Using social media to quantify nature-based tourism and recreation“. All papers treat nature-based tourism and the methods to measure visitor activity in such destinations (Table 7 ). Wood et al. ( 2013 ) utilized user-generated content (UGC) to quantify nature-based tourism, approximating visitation rates at 836 recreational sites around the world using Flickr data. Spencer Wood is recognized by others in this cluster to be the most prolific author (Teles da Mota & Pickering, 2020 ), and this paper, in particular, is the most influential on Flickr research of nature-based tourism (H. Zhang et al., 2021 ). The other node that defines most of this cluster is the more recent work authored by Teles da Mota & Pickering ( 2020 ); their paper “Using social media to assess nature-based tourism: Current research and future trends” cites most of the papers in the cluster in order to theorize this topic. Studies in this community use similar methodologies to proxy tourism visitation data per country, comparing the number of images per day per user as done by an even more recent paper by Teles da Mota et al. ( 2022 ). National parks, coastal beach areas, and large natural areas are the focus of all of the studies. Table 7 Nature-based tourism research community Paper Aim In-/Out-degree Theme/topic (Wood et al., 2013 ) Using UGC to quantify nature-based tourism, approximating visitation rates at 836 recreational sites using Flickr data 47/3 Nature-based tourism (Hausmann et al., 2017 ; Tenkanen et al., 2017 ) Comparing the usability of social media data from Instagram, Flickr, and Twitter for visitor monitoring in protected areas, finding tourists’ preferences to see large mammals. 21/0 25/0 Nature-based tourism National parks (Hausmann et al., 2018 ) Understanding Tourists’preferences for nature-based experiences in Protected Areas (Flickr and Instagram) 18/15 Nature-based tourism (Gosal et al., 2019 ) Identifying different groups of recreation users with social media, machine learning, and natural language processing for better ecosystem services 7/24 Outdoor recreation Machine learning (Teles da Mota & Pickering, 2020 ) Literature review on the use of social media data to assess nature-base tourism (all social media) 5/60 Nature-based tourism (Pickering et al., 2020 ) Assessing seasonal preferences for tourism in a National Park (facilities and activities, landscapes, flora and fauna) comparing image content and text from Flickr 6/33 Nature-based tourism National parks (Runge et al., 2020 ) Analyzing localized tourism booms and quantify the spatial expansion of the Arctic tourism footprint 4/15 Nature-based tourism (Sinclair et al., 2020 ) Estimating the media visitors' home locations in German national parks (Flickr) 3/35 Nature-based tourism National parks (Huai et al., 2022 ) Using social media photos and computer vision to assess cultural ecosystem services and landscape features in urban parks 0/40 Urban Parks Computer vision (Zhang et al., 2021 ) To measure and map spatial patterns in visitation of public lands in Utah (Flickr and Panoramio) 0/38 Nature-based tourism This cluster demonstrated that social media can proxy visitor counts in natural parks. Nonetheless, it focused heavily on quantification, rarely examining qualitative experiences of nature or local community impacts. Therefore, future studies should triangulate social media data with participatory methods involving residents and tourists. There is also a predominantly quantitative focus risks reducing nature-based tourism to mere visitation numbers, overlooking issues like visitor impact, conservation challenges, and local community perspectives. Besides that, it can be highlighted that there are geographical biases because protected areas in less digitised regions may be underrepresented, skewing global analyses. Other future research possibilities include cross-validation of Flickr-based visitation proxies with ground-based conservation impact data (e.g., trail erosion, wildlife disturbance). Conclusions: The undeniable contribution of open-API social media to tourism research The first cited work using Flickr geotagged photography to understand tourist dynamics is from 2008 (Girardin, Dal Fiore, et al., 2008 ), and this study shows how 15 years later, at least 7 well-defined research topics or research methods have been developed by researchers from all continents. However, Flickr is not any more a popular platform to share UGC; the amount of user-generated content in Flickr social media has slowly decreased since 2016 (Solazzo et al., 2022 ). Other popular platforms closed their APIs for external data requests. Facebook suspended the export of friend data in 2015, bought Instagram in 2016 and suspended the possibility of downloading data such as GPS coordinates for photos and after the CA scandal (Tromble, 2021 ), and severely limited the API access to Facebook Groups and Events in April 2018. Instead of freely researchable datasets, web corporations often impose a commercial construct. Former Twitter (now X) made it possible to search only 7 days back in time, accessing only a sample of all tweets. However, researchers were able to pay for Twitter data access through their Historical API (Perriam et al., 2020 ). Researchers have more times publicly addressed Facebook and Twitter to reopen forms of public APIs with proposals that would have disclosed data abuse, but the corporations never even replied (Bruns, 2021 ). The only option for researchers remains web scraping, a much more resource-intensive way to gain data from social media, and one that has no possibilities to control the ethical issues Tromble ( 2021 ) proposing to reflect on. The scientific use of Flickr as an open API platform presented in this paper shows that the concerns of Tromble were not in every aspect relevant and large communities of researchers have used that open API social media without ethically questionable results and with great benefits for science. What were these? Tourism research must rely on visitor data, while tour operators, any business involved in the visitor economy and large corporations such as payment card services and mobile phone operators are commercially counteractive in giving out any statistical data to independent researchers. With openly accessible UGC data from all continents on photography – a large portion of all is travel photography – researchers could work with methodologies and on topics otherwise inaccessible. Much of Flickr photography is geo-tagged, meaning that the exact geographical location is accessible, together with the time when the photo was taken – associated with the user who uploaded it. Such data-enabled research on the spatiotemporal analysis of tourism (C1), even in large destination systems and natural parks (C7) – where other technologies tracing visitors’ movements does not exist. No other widely accessible data with such even geographic distribution across the globe exists that can exactly show where and when and how many visitors there were. Tracing tourist movements would have remained the privilege of corporations like Facebook, Google or mobile phone operators, which do use such data, but only for their own commercial interest. The comparison of different places being photographed, more or less differentiating also by visitor types, allowed the measurement of the attractivity of places (C4), resulting in a deeper understanding of why some attractions perform better than others – impossible to research without comparatively accessible data on many places of interest generated by users whose attributes are also possible to distinguish. The understanding of which places are of tourists’ interest depending on different preferences and how they select to visit these in a given timeframe is also a research question in need of large amounts of spatiotemporal data from which different clustering techniques and algorithms (like DBSCAN) can detect which are the geographical areas to be visited (C2). Many researchers could propose travel route recommendation techniques based on such datasets, which also helped to give a relevant quantity of data to be able to provide valid tourist itineraries in line with the Orienteering Problem (C3). Another application of Flickr data in tourism research was tied to image content analysis. Researchers could find no other data sources of systematically researchable tourism-related photography where, given the time and place of the photo, the relevant information deducted from their content could be analyzed. Image content analysis proved to be extremely useful in understanding the differences between the projected and perceived destination images of a destination (C5). Researchers could analyze the attraction preferences of different types of tourists, ranging from urban or natural landmarks to different food and beverages. Computer vision was much used for the more contemporary works with destination images. The immense quantity of geolocated photographs was given from Flickr API, therefore enough material was in the disposition of researchers to use machine learning to differentiate the content of photos. No other publicly available dataset was available for the training of computer vision; on the other hand, the qualitative analysis of such large datasets does need machine learning techniques to process all data. As such methodology is useful not only in destination image research but any tourism-related research which needed to analyze the content of Flickr photography, a cluster well dispersed between the other – much more separated - clusters of the bibliographical network was identified with quite recent work included (C6). The community using deep learning methods and the nature-based tourism research community provided the most recent works. The reason that C7 is the cluster with the most recent papers is because visitor statistics of locations inside natural parks and large natural areas or regions are impossible to get in any other ways, if not by expensive individual GPS trackers distributed. Many researchers in C7 directly reflect on the availability of Flickr data as a research assess. Tenkanen et al., ( 2017 ) state that data from Flickr is free and easy to download and the temporal patterns in social media data are often similar to results from park entry and trail counters in some natural areas; this statement has been directly quoted by others in C7 (Teles da Mota & Pickering, 2020 ). Hausmann et al. ( 2018 ) came to a similar conclusion for South African national parks, also mentioning that geodata from social media is a robust indicator for human presence and spatial variation of visitation in protected areas at regional, national and global scale. These observations prove that nature-based tourism research could evolve with the abovementioned relevant results because of the availability of the open API of Flickr, used by researchers worldwide. Flickr was never a flawless database. Some authors mention the existence of deviations and biases related to the popularity of the social media site, the demographic profile of the users, and the different usage of the social media website in terms of recurrence and selection of photos (Barros et al., 2019 ). In fact, there is an overrepresentation of certain demographics (e.g., affluent, tech-savvy tourists) and specific geographies (urban centres and iconic attractions). These limitations point to critical vulnerabilities in research that remain unaddressed in much of the existing literature. Still, there has never been a database so large, precise, and accurate for researching tourism - and with the current trend to use only closed internet ecosystems, there will never be one again. Without the open API UGC platform as Flickr, none of the 7 specific topics or methods researched by these large and independent global research communities could have advanced so much in their fields of tourism studies, providing so many highly ranked publications. Research fields of C4-5-6-7 would have developed much smaller without robust datasets to build on their observations, while C1-2-3 would never have even developed as research fields. While previous works have documented the use of social media in tourism (e.g., Shoval & Ahas, 2016 ; Giglio et al., 2019 ), they often remained descriptive, focusing merely on cataloguing applications rather than interrogating the structural transformations such datasets enabled in the field. Our analysis reveals that Flickr not only offered new methods for data collection but fundamentally altered the epistemological landscape of tourism studies, fostering collaborative methodologies, enhancing reproducibility, and encouraging global comparative studies. Moreover, while the literature extensively exploited geotagged photography for spatial analysis, conceptual integration between behavioural tourism studies and computational methods remains underdeveloped. And probably, this disconnect limits the theoretical maturation of findings derived from such datasets. Although Scopus offers greater coverage of scientific articles, the main limitation of this work is that it may not include all the documents on Flickr and tourism literature. The network could have been further expanded beyond the direct dataset with the given tags and the works referenced by these, including those referenced by this indirect dataset. The same clusters would have been found in a larger dataset where all relevant research with the expression “Flickr” in the keywords or abstracts would have been listed without any further limitations. It is an interesting question how much part of all research using Flickr datasets treats tourism. However, it is probable that a golden era of invisible and independent global research communities, giving at least seven new widely published research topics in a decade and a half within tourism, has come to an end. Declarations Author Contribution B. Kádár provided the theoretical parts, methodology and results and figures.M. Martins provided the extraction and management of bibliographic data, tables, and validated the results.All authors contributed to the main manuscript text and reviewed it. Data Availability The datasets generated by the Scopus bibliographic database analyzed during the current study are available in the Mendeley repository: Kadar, Balint; Martins, Márcio Ribeiro (2025), “ bibliometric network analysis of studies using Flickr data 2010-2023”, Mendeley Data, V2, doi: 10.17632/fkz7wv3f7y.2 References Ahas R, Aasa A, Mark Ü, Pae T, Kull A (2007) Seasonal tourism spaces in Estonia: Case study with mobile positioning data. Tour Manag 28:898–910. https://doi.org/10.1016/j.tourman.2006.05.010 Ashworth GJ, Page SJ (2011) Urban tourism research: Recent progress and current paradoxes. Tour Manag 32(1):1–15. https://doi.org/10.1016/j.tourman.2010.02.002 Barros C, Moya-Gómez B, García-Palomares JC (2019) Identifying temporal patterns of visitors to national parks through geotagged photographs. Sustain (Switzerland) 11(24). https://doi.org/10.3390/su11246983 Bhatt P, Pickering CM (2022) Destination image of Chitwan National Park, Nepal: Insights from a content analysis of online photographs. J Outdoor Recreation Tourism 37. https://doi.org/10.1016/j.jort.2022.100488 Blondel VD, Guillaume J-L, Lambiotte R, Lefebvre E (2008) Fast unfolding of communities in large networks. J Stat Mech: Theory Exp 2008(10):1–12 Bojic I, Belyi A, Ratti C, Sobolevsky S (2016) Scaling of foreign attractiveness for countries and states. Appl Geogr 73:47–52. https://doi.org/10.1016/j.apgeog.2016.06.006 Broz M (2022) Flickr Statistics, User Count, & Facts (September 2022). Photutorial.Com. https://photutorial.com/flickr-statistics/ Bruns A (2021) After the ‘APIcalypse’: social media platforms and their fight against critical scholarly research. Disinformation and Data Lockdown on Social Platforms, 14–36. https://doi.org/10.4324/9781003206972-2 Chen W, Xu Z, Zheng X, Luo Y (2019) Geo-tagged photo metadata processing method for Beijing inbound tourism flow. ISPRS Int J Geo-Information 8(12). https://doi.org/10.3390/ijgi8120556 Cho N, Kang Y, Yoon J, Park S, Kim J (2022) Classifying Tourists’ Photos and Exploring Tourism Destination Image Using a Deep Learning Model. J Qual Assur Hospitality Tourism 23(6):1480–1508. https://doi.org/10.1080/1528008X.2021.1995567 Crandall D, Backstrom L, Huttenlocher D, Kleinberg J (2009) Mapping the world’s photos. Proceedings of the 18th International Conference on World Wide Web., 761–770 Curlin T, Jaković B, Miloloža I (2019) Twitter usage in Tourism: Literature Review. Bus Syst Res 10(1):102–119. https://doi.org/10.2478/bsrj-2019-0008 De Choudhury M, Feldman M, Amer-Yahia S, Golbandi N, Lempel R, Yu C (2010) Automatic Construction of Travel Itineraries using Social Breadcrumbs. Proceedings of the 21st ACM Conference on Hypertext and Hypermedia, 35–44 Deng N, Li XR (2018) Feeling a destination through the right photos: A machine learning model for DMOs’ photo selection. Tour Manag 65:267–278. https://doi.org/10.1016/j.tourman.2017.09.010 Deng N, Liu J, Dai Y, Li H (2019) Different cultures, different photos: A comparison of Shanghai’s pictorial destination image between East and West. Tourism Manage Perspect 30:182–192. https://doi.org/10.1016/j.tmp.2019.02.016 Dickinson JE, Ghali K, Cherrett T, Speed C, Davies N, Norgate S (2014) Tourism and the smartphone app: capabilities, emerging practice and scope in the travel domain. Curr Issues Tourism 17(1):84–101. https://doi.org/10.1080/13683500.2012.718323 Donaire JA, Camprubí R, Galí N (2014) Tourist clusters from Flickr travel photography. Tourism Manage Perspect 11:26–33. https://doi.org/10.1016/j.tmp.2014.02.003 Eck NJ, Van, Waltman L (2016) Text Mining and Visualization. Text Mining and Visualization, 1–5. https://doi.org/10.1201/b19007 Ester M, Kriegel H-P, Sander J, Xu X (1996) A Density-Based Algorithm for Discovering Clusters in Large Spatial Databases with Noise. www.aaai.org García-Palomares JC, Gutiérrez J, Mínguez C (2015) Identification of tourist hot spots based on social networks: A comparative analysis of European metropolises using photo-sharing services and GIS. Appl Geogr 63:408–417. https://doi.org/10.1016/j.apgeog.2015.08.002 Garrod B (2009) Understanding the Relationship between Tourism Destination Imagery and Tourist Photography. J Travel Res 47(3):346–358 Giglio S, Bertacchini F, Bilotta E, Pantano P (2019) Using social media to identify tourism attractiveness in six Italian cities. Tour Manag 72:306–312. https://doi.org/10.1016/j.tourman.2018.12.007 Girardin F, Calabrese F, Fiore FD, Biderman A, Ratti C, Blat J (2008) Uncovering the presence and movements of tourists from user- generated content Girardin F, Dal Fiore F, Blat J, Ratti C (2007) Understanding of tourist dynamics from explicitly disclosed location information. 4th International Symposium on LBS and Telecartography , 58 Girardin F, Dal Fiore F, Ratti C, Blat J (2008) Leveraging explicitly disclosed location information to understand tourist dynamics: A case study. J Location Based Serv 2(1):41–56. https://doi.org/10.1080/17489720802261138 Girardin F, Fiore FD, Blat J, Ratti C, Fiore D, F (2008) Understanding of Tourist Dynamics from Explicitly Disclosed Location Information. J Location Based Serv, 2(1) González MC, Hidalgo CA, Barabási AL (2008) Understanding individual human mobility patterns. Nat 2008 453(7196):779–782. https://doi.org/10.1038/nature06958 . 453 Gosal AS, Geijzendorffer IR, Václavík T, Poulin B, Ziv G (2019) Using social media, machine learning and natural language processing to map multiple recreational beneficiaries. Ecosyst Serv 38. https://doi.org/10.1016/j.ecoser.2019.100958 Hausmann A, Toivonen T, Heikinheimo V, Tenkanen H, Slotow R, Minin EDI (2017) Social media reveal that charismatic species are not the main attractor of ecotourists to sub-Saharan protected areas. Sci Rep 2017 7(1):1–9. https://doi.org/10.1038/s41598-017-00858-6 Hausmann A, Toivonen T, Slotow R, Tenkanen H, Moilanen A, Heikinheimo V, Di Minin E (2018) Social Media Data Can Be Used to Understand Tourists’ Preferences for Nature-Based Experiences in Protected Areas. Conserv Lett 11(1):e12343. https://doi.org/10.1111/CONL.12343 Hawelka B, Sitko I, Beinat E, Sobolevsky S, Kazakopoulos P, Ratti C (2014) Geo-located Twitter as proxy for global mobility patterns. Cartography Geographic Inform Sci 41(3):260–271. https://doi.org/10.1080/15230406.2014.890072 Herrera-Franco G, Montalván-Burbano N, Carrión-Mero P, Apolo-Masache B, Jaya-Montalvo M (2020) Research Trends in Geotourism: A Bibliometric Analysis Using the Scopus Database. Geosci 2020 10(10):379. https://doi.org/10.3390/GEOSCIENCES10100379 . 10 Hollenstein L, Purves RS (2010) Exploring place through user-generated content: Using Flickr tags to describe city cores. J Spat Inform Sci 1(2010):21–48. https://doi.org/10.5311/JOSIS.2010.1.3 Höpken W, Müller M, Fuchs M, Lexhagen M (2020) Flickr data for analysing tourists’ spatial behaviour and movement patterns: A comparison of clustering techniques. J Hospitality Tourism Technol 11(1):69–82. https://doi.org/10.1108/JHTT-08-2017-0059 Hruška J, Pásková M (2018) How do the national tourism organizations use the social media? [Jak národní centrály cestovního ruchu využívají sociální média?]. E M: Ekonomie Manage 21(4):226–240. https://doi.org/10.15240/tul/001/2018-4-015 Hu Y, Gao S, Janowicz K, Yu B, Li W, Prasad S (2015) Extracting and understanding urban areas of interest using geotagged photos. Comput Environ Urban Syst 54:240–254. https://doi.org/10.1016/j.compenvurbsys.2015.09.001 Huai S, Chen F, Liu S, Canters F, Van de Voorde T (2022) Using social media photos and computer vision to assess cultural ecosystem services and landscape features in urban parks. Ecosyst Serv 57. https://doi.org/10.1016/j.ecoser.2022.101475 Hwang YH, Fesenmaier DR (2003) Multidestination pleasure travel patterns: Empirical evidence from the American Travel Survey. J Travel Res 42(2):166–171. https://doi.org/10.1177/0047287503253936 Jacomy M, Heymann S, Venturini T, Bastian M (2011) Force Atlas 2, a graph layout algorithm for handy network visualization. … Http://Www.Medialab. Sciences-Po. Fr/ …, 1–21. http://scholar.google.com/scholar?hl=en &btnG=Search&q=intitle:ForceAtlas2+,+A+Graph+Layout+Algorithm+for+Handy+Network+Visualization#0 Jain A, Singh SN (2017) Identifying concavity laws of attractiveness character for composite regions: Application through geo-tagged media check-ins. Proceeding - IEEE International Conference on Computing, Communication and Automation, ICCCA 2017, 2017-January, 1041–1045. https://doi.org/10.1109/CCAA.2017.8229994 Jenkins OH (2003) Photography and travel brochures: the circle of representation. Tourism Geographies 5(3):305–328. https://doi.org/10.1080/1461668032000098024 Kádár B (2014) Measuring tourist activities in cities using geotagged photography. Tourism Geographies 16(1):88–104. https://doi.org/10.1080/14616688.2013.868029 Kádár B, Gede M (2013) Where do tourists go? Visualizing and analysing the spatial distribution of geotagged photography. Cartographica 48(2):78–88. https://doi.org/10.3138/carto.48.2.1839 Kádár B, Gede M (2021) Tourism flows in large-scale destination systems. Annals Tourism Res 87. https://doi.org/10.1016/j.annals.2020.103113 Kang Y, Cho N, Yoon J, Park S, Kim J (2021) Transfer learning of a deep learning model for exploring tourists’ urban image using geotagged photos. ISPRS Int J Geo-Information 10(3). https://doi.org/10.3390/ijgi10030137 Karayazi SS, Dane G, de Vries B (2021) Utilizing urban geospatial data to understand heritage attractiveness in amsterdam. ISPRS Int J Geo-Information 10(4). https://doi.org/10.3390/ijgi10040198 Kennedy L, Naaman M, Ahern S, Nair R, Rattenbury T (2007) How Flickr Helps us Make Sense of the World: Context and Content in Community-Contributed Media Collections. Proceedings of the 15th ACM International Conference on Multimedia. http://flickr.com Kim D, Kang Y, Park Y, Kim N, Lee J (2020) Understanding tourists’ urban images with geotagged photos using convolutional neural networks. Spat Inform Res 28(2):241–255. https://doi.org/10.1007/s41324-019-00285-x Kisilevich S, Keim D, Andrienko N, Andrienko G (2013) Towards acquisition of semantics of places and events by multi-perspective analysis of geotagged photo collections. In: Moore A, Drecki I (eds) Geospatial Visualisation. Lecture Notes in Geoinformation and Cartography. Springer. http://www.flickr.com Kisilevich S, Mansmann F, Keim D (2010) P-DBSCAN: A density based clustering algorithm forexploration and analysis of attractive areas usingcollections of geo-tagged photos. COM. Geo 2010: 1st International Conference and Exhibition on Computing for Geospatial Research & Application, 274 Korakakis M, Spyrou E, Mylonas P, Perantonis SJ (2017) Exploiting social media information toward a context-aware recommendation system. Social Netw Anal Min 7(1). https://doi.org/10.1007/s13278-017-0459-9 Koutras A, Nikas IA, Panagopoulos A (2019) Towards Developing Smart Cities: Evidence from GIS Analysis on Tourists’ Behavior Using Social Network Data in the City of Athens. In: Katsoni V, Segarra-Oña M (eds) Smart Tourism as a Driver for Culture and Sustainability. Springer International Publishing, pp 407–418 Krizhevsky A, Sutskever I, Hinton GE (2012) ImageNet Classification with Deep Convolutional Neural Networks. Advances in Neural Information Processing Systems, 25. http://code.google.com/p/cuda-convnet/ Kurashima T, Iwata T, Irie G, Fujimura K (2010) Travel Route Recommendation Using Geotags in Photo Sharing Sites. Proceedings of the 19th ACM International Conference on Information and Knowledge Management , 579–588 Kurashima T, Iwata T, Irie G, Fujimura K (2013) Travel route recommendation using geotagged photos. Knowl Inf Syst 37(1):37–60. https://doi.org/10.1007/s10115-012-0580-z Lee S, Zhang H, Crandall DJ (2015) Predicting geo-informative attributes in large-scale image collections using convolutional neural networks. Proceedings – 2015 IEEE Winter Conference on Applications of Computer Vision, WACV 2015, 550–557. https://doi.org/10.1109/WACV.2015.79 Li D, Zhou X, Wang M (2018) Analyzing and visualizing the spatial interactions between tourists and locals: A Flickr study in ten US cities. Cities 74:249–258. https://doi.org/10.1016/j.cities.2017.12.012 Lim KH (2016) Recommending and planning trip itineraries for individual travellers and groups of tourists. Proceedings International Conference on Automated Planning and Scheduling, ICAPS, 2016-Janua(April) Lim KH, Chan J, Leckie C, Karunasekera S (2018) Personalized trip recommendation for tourists based on user interests, points of interest visit durations and visit recency. Knowl Inf Syst 54(2):375–406. https://doi.org/10.1007/s10115-017-1056-y Lin MS, Liang Y, Xue JX, Pan B, Schroeder A (2021) Destination image through social media analytics and survey method. Int J Contemp Hospitality Manage 33(6):2219–2238. https://doi.org/10.1108/IJCHM-08-2020-0861 Lowe DG (2004) Distinctive image features from scale-invariant keypoints. Int J Comput Vision 60(2):91–110. https://doi.org/10.1023/B:VISI.0000029664.99615.94/METRICS Majid A, Chen L, Chen G, Mirza HT, Hussain I, Woodward J (2013) A context-aware personalized travel recommendation system based on geotagged social media data mining. Int J Geogr Inf Sci 27(4):662–684. https://doi.org/10.1080/13658816.2012.696649 Mangachena JR, Chauvenet ALM, Pickering CM (2022) Destination Images of Africa Reflected in Social Media Photographs. https://doi.org/10.1080/08941920.2022.2100023 . Society and Natural Resources Martins M, Costa R (2022) Tracking Technologies in Tourism: A Bibliometric and Content Review. In: de Carvalho JV, Liberato P, Peña A (eds) Advances in Tourism, Technology and Systems. Springer Nature Singapore, pp 215–230 Mckercher B, Shoval N, Ng E, Birenboim A (2012) First and Repeat Visitor Behaviour: GPS Tracking and GIS Analysis in Hong Kong. Tourism Geographies 14(1):147–161. https://doi.org/10.1080/14616688.2011.598542 Memon I, Chen L, Majid A, Lv M, Hussain I, Chen G (2015) Travel Recommendation Using Geo-tagged Photos in Social Media for Tourist. Wireless Pers Commun 80(4):1347–1362. https://doi.org/10.1007/s11277-014-2082-7 Miah SJ, Vu HQ, Gammack J, McGrath M (2017) Inform Manage 54(6):771–785. https://doi.org/10.1016/j.im.2016.11.011 . A Big Data Analytics Method for Tourist Behaviour Analysis Mor M, Dalyot S (2020) Enriching walking routes with tourism attractions retrieved from crowdsourced user generated data. ISPRS Annals of the Photogrammetry. Remote Sens Spat Inform Sci 5(4):95–102. https://doi.org/10.5194/isprs-Annals-V-4-2020-95-2020 Mor M, Fisher-Gewirtzman D, Yosifof R, Dalyot S (2021) 3D visibility analysis for evaluating the attractiveness of tourism routes computed from social media photos. ISPRS Int J Geo-Information 10(5). https://doi.org/10.3390/ijgi10050275 Mou N, Jiang Q, Zhang L, Niu J, Zheng Y, Wang Y, Yang T (2022) Personalized tourist route recommendation model with a trajectory understanding via neural networks. Int J Digit Earth 15(1):1738–1759. https://doi.org/10.1080/17538947.2022.2130456 Nikas IA, Koutras A, Panagopoulos A, Vasileiadis A (2022) Exploring Tourists’ Food and Beverage Spots in an Urban Destination Using a Spatial–temporal Approach. In: Katsoni V, \cSerban AC (eds) Transcending Borders in Tourism Through Innovation and Cultural Heritage. Springer International Publishing, pp 977–989 Novack T, Vorbeck L, Lorei H, Zipf A (2020) Towards detecting building facades with graffiti artwork based on street view images. ISPRS Int J Geo-Information 9(2). https://doi.org/10.3390/ijgi9020098 Önder I, Koerbitz W, Hubmann-Haidvogel A (2016) Tracing Tourists by Their Digital Footprints: The Case of Austria. J Travel Res 55(5):566–573. https://doi.org/10.1177/0047287514563985 Paldino S, Bojic I, Sobolevsky S, Ratti C, González MC (2015) Urban magnetism through the lens of geo-tagged photography. EPJ Data Sci 4(1):1–17. https://doi.org/10.1140/epjds/s13688-015-0043-3 Pan S, Lee J, Tsai H (2014) Travel photos: Motivations, image dimensions, and affective qualities of places. Tour Manag 40:59–69. https://doi.org/10.1016/j.tourman.2013.05.007 Perriam J, Birkbak A, Freeman A (2020) Digital methods in a post-API environment. Int J Soc Res Methodol 23(3):277–290. https://doi.org/10.1080/13645579.2019.1682840 Pickering C, Walden-Schreiner C, Barros A, Rossi SD (2020) Using social media images and text to examine how tourists view and value the highest mountain in Australia. J Outdoor Recreation Tourism 29. https://doi.org/10.1016/j.jort.2019.100252 Popescu A, Grefenstette G, Moëllic PA (2009) Mining tourist information from user-supplied collections. International Conference on Information and Knowledge Management, Proceedings, 1713–1716. https://doi.org/10.1145/1645953.1646211 Rattenbury T (2009) (TWEB), M. N.-A. T. on the W., & undefined. (2009). Methods for extracting place semantics from Flickr tags. Dl.Acm.OrgT Rattenbury, M NaamanACM Transactions on the Web (TWEB), 2009•dl.Acm.Org, 3(1). https://doi.org/10.1145/1462148.1462149 Runge CA, Daigle RM, Hausner VH (2020) Quantifying tourism booms and the increasing footprint in the Arctic with social media data. PLoS ONE 15(1). https://doi.org/10.1371/journal.pone.0227189 Sarkar JL, Majumder A, Panigrahi CR, Roy S (2020) MULTITOUR: A multiple itinerary tourists recommendation engine. Electron Commer Res Appl 40. https://doi.org/10.1016/j.elerap.2020.100943 Shoval N, Ahas R (2016) The use of tracking technologies in tourism research: the first decade. Tourism Geographies 18(5):587–606. https://doi.org/10.1080/14616688.2016.1214977 Sinclair M, Mayer M, Woltering M, Ghermandi A (2020) Using social media to estimate visitor provenance and patterns of recreation in Germany’s national parks. J Environ Manage 263. https://doi.org/10.1016/j.jenvman.2020.110418 Sobolevsky S, Sitko I, Des Combes RT, Hawelka B, Arias JM, Ratti C (2014) Money on the move: Big data of bank card transactions as the new proxy for human mobility patterns and regional delineation. The case of residents and foreign visitors in Spain. Proceedings – 2014 IEEE International Congress on Big Data, BigData Congress 2014, 136–143. https://doi.org/10.1109/BigData.Congress.2014.28 Solazzo G, Maruccia Y, Lorenzo G, Ndou V, Del Vecchio P, Elia G (2022) Extracting insights from big social data for smarter tourism destination management. Measuring Bus Excellence 26(1):122–140. https://doi.org/10.1108/MBE-11-2020-0156 Stepchenkova S, Zhan F (2013) Visual destination images of Peru: Comparative content analysis of DMO and user-generated photography. Tour Manag 36:590–601. https://doi.org/10.1016/j.tourman.2012.08.006 Straumann RK, Çöltekin A, Andrienko G (2014) Towards (re)constructing narratives from georeferenced photographs through visual analytics. Cartographic J 51(2):152–165. https://doi.org/10.1179/1743277414Y.0000000079 Sun Y, Fan H, Bakillah M, Zipf A (2015) Road-based travel recommendation using geo-tagged images. Comput Environ Urban Syst 53:110–122. https://doi.org/10.1016/j.compenvurbsys.2013.07.006 Teles da Mota V, Pickering C (2020) Using social media to assess nature-based tourism: Current research and future trends. Journal of Outdoor Recreation and Tourism , 30 . https://doi.org/10.1016/j.jort.2020.100295 Teles da Mota V, Pickering C, Chauvenet A (2022) Popularity of Australian beaches: Insights from social media images for coastal management. Ocean Coast Manag 217. https://doi.org/10.1016/j.ocecoaman.2021.106018 Tenkanen H, Di Minin E, Heikinheimo V, Hausmann A, Herbst M, Kajala L, Toivonen T (2017) Instagram, Flickr, or Twitter: Assessing the usability of social media data for visitor monitoring in protected areas. Sci Rep 7(1):1–11. https://doi.org/10.1038/s41598-017-18007-4 Tromble R (2021) Social Media Soc 7(1). https://doi.org/10.1177/2056305121988929 . Where Have All the Data Gone? A Critical Reflection on Academic Digital Research in the Post-API Age Urry J (1990) The Tourist gaze - Leisure and Travel in Contemporary Societies. SAGE publications Ltd. Valls F, Roca J (2021) Visualizing digital traces for sustainable urban management: Mapping tourism activity on the virtual public space. Sustain (Switzerland) 13(6). https://doi.org/10.3390/su13063159 van Eck NJ, Waltman L (2017) Citation-based clustering of publications using CitNetExplorer and VOSviewer. Scientometrics 111(2):1053. https://doi.org/10.1007/S11192-017-2300-7 Vu HQ, Li G, Law R, Ye BH (2015) Exploring the travel behaviors of inbound tourists to Hong Kong using geotagged photos. Tour Manag 46:222–232. https://doi.org/10.1016/j.tourman.2014.07.003 Vu HQ, Li G, Law R, Zhang Y (2018) Travel Diaries Analysis by Sequential Rule Mining. J Travel Res 57(3):399–413. https://doi.org/10.1177/0047287517692446 Wood SA, Guerry AD, Silver JM, Lacayo M (2013) Using social media to quantify nature-based tourism and recreation. Sci Rep 3. https://doi.org/10.1038/srep02976 Yang L, Wu L, Liu Y, Kang C (2017) Quantifying tourist behavior patterns by travel motifs and geo-tagged photos from flickr. ISPRS Int J Geo-Information 6(11). https://doi.org/10.3390/ijgi6110345 Yeran S, Fan H (2014) Event Identification from Georeferenced Images. In J. Huerta, S. Schade, & C. Granell (Eds.), Connecting a digital Europe through location and place. Lecture Notes in Geoinformation and Cartography. (pp. 73–78). Springer International Publishing. https://doi.org/https://doi.org/10.1007/978-3-319-03611-3_5 Yuan Y, Medel M (2016) Characterizing international travel behavior from geotagged photos: A case study of Flickr. PLoS ONE 11(5). https://doi.org/10.1371/journal.pone.0154885 Zhang H, van Berkel D, Howe PD, Miller ZD, Smith JW (2021) Using social media to measure and map visitation to public lands in Utah. Appl Geogr 128. https://doi.org/10.1016/j.apgeog.2021.102389 Zhang K, Chen Y, Li C (2019) Discovering the tourists’ behaviors and perceptions in a tourism destination by analyzing photos’ visual content with a computer deep learning model: The case of Beijing. Tour Manag 75:595–608. https://doi.org/10.1016/j.tourman.2019.07.002 Zhang K, Chen Y, Lin Z (2020) Mapping destination images and behavioral patterns from user-generated photos: a computer vision approach. Asia Pac J Tourism Res 25(11):1199–1214. https://doi.org/10.1080/10941665.2020.1838586 Zheng Y-T, Zha Z-J, Chua T-S (2012) Mining travel patterns from geotagged photos. ACM Trans Intell Syst Technol 3(3). https://doi.org/10.1145/2168752.2168770 Zhou X, Xu C, Kimmons B (2015) Detecting tourism destinations using scalable geospatial analysis based on cloud computing platform. Comput Environ Urban Syst 54:144–153. https://doi.org/10.1016/j.compenvurbsys.2015.07.006 Zhu Y, Deng X, Newsam S (2019) Fine-Grained Land Use Classification at the City Scale Using Ground-Level Images. IEEE Trans Multimedia 21(7):1825–1838. https://doi.org/10.1109/TMM.2019.2891999 Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 06 Oct, 2025 Reviews received at journal 02 Oct, 2025 Reviews received at journal 30 Sep, 2025 Reviewers agreed at journal 24 Sep, 2025 Reviewers agreed at journal 22 Sep, 2025 Reviewers agreed at journal 25 Aug, 2025 Reviewers agreed at journal 18 Aug, 2025 Reviewers invited by journal 18 Aug, 2025 Editor assigned by journal 18 Aug, 2025 Submission checks completed at journal 09 Jul, 2025 First submitted to journal 08 Jul, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7072228","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":505400200,"identity":"be27fa13-73d5-41e0-95d5-014527a89ea9","order_by":0,"name":"Bálint Kádár","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAv0lEQVRIiWNgGAWjYBACCQYGAwbGBhsQiw0kkECsljTStRwmQYtk++GNnyt3nE/cPruB7dGNCoY8gwMEtEjzpBVLnj1zO3HOnQPsxjlnGIoJapFjyDGQbGy7nThDIoFNOreNIXEDQS38b4x/Nradg2r5R4QWaYkcM6AtB6BaGojQIjnjWZll45lk4xkyB9uNc45JJM4kpEXifPLmm4077GRnSDcfe5xTY5PYR0gLEmBsYADH0ygYBaNgFIwCygEANtVED5bdoqQAAAAASUVORK5CYII=","orcid":"","institution":"Budapest University of Technology and Economics","correspondingAuthor":true,"prefix":"","firstName":"Bálint","middleName":"","lastName":"Kádár","suffix":""},{"id":505400203,"identity":"61391c9b-141d-49f6-b14d-39a0dbf5c639","order_by":1,"name":"Márcio Ribeiro Martins","email":"","orcid":"","institution":"Polytechnic Institute of Bragança","correspondingAuthor":false,"prefix":"","firstName":"Márcio","middleName":"Ribeiro","lastName":"Martins","suffix":""}],"badges":[],"createdAt":"2025-07-08 08:08:18","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7072228/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7072228/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":89987337,"identity":"8e74a41c-5c99-46f6-99ba-5c881e20d453","added_by":"auto","created_at":"2025-08-27 06:58:43","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":2646265,"visible":true,"origin":"","legend":"\u003cp\u003eFlowchart of the methodology to process the bibliography\u003c/p\u003e","description":"","filename":"Figure1methods.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7072228/v1/2a04d6a5983b49330df658f2.jpg"},{"id":89987342,"identity":"5b573123-14e1-4e34-82b6-f53aefe63d0a","added_by":"auto","created_at":"2025-08-27 06:58:43","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":12994570,"visible":true,"origin":"","legend":"\u003cp\u003eThe bibliographic network of papers related to tourism and social media images (compiled in Gephi 0.10.1.)\u003c/p\u003e","description":"","filename":"Figure2years.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7072228/v1/25f80bfee0bca637b3ed3ff4.jpg"},{"id":89988849,"identity":"355b057c-bc6f-401f-a49c-ba579404ecf8","added_by":"auto","created_at":"2025-08-27 07:06:43","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":8924916,"visible":true,"origin":"","legend":"\u003cp\u003eThe 7 clusters found with Gephi in the bibliographic network (compiled in Gephi 0.10.1.)\u003c/p\u003e","description":"","filename":"Figure3clusters.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7072228/v1/78c6ce94515bc62b97fcbe63.jpg"},{"id":89987336,"identity":"560f0ace-f5a2-41c4-ab39-0316361a4c6b","added_by":"auto","created_at":"2025-08-27 06:58:43","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":1748308,"visible":true,"origin":"","legend":"\u003cp\u003eReferences inside and between the 7 clusters represented as a weighted network (compiled in Gephi 0.10.1.)\u003c/p\u003e","description":"","filename":"Figure4sevenclusters.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7072228/v1/8f14d7209d25687e477af5fa.jpg"},{"id":89990567,"identity":"ea4422a7-fa47-4966-8f08-235b9b728816","added_by":"auto","created_at":"2025-08-27 07:14:51","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":27391905,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7072228/v1/5f775dd0-52d9-469f-a0a6-a513c8578ad0.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"The contribution of accessible social media data to the work of tourism research communities: a bibliometric network analysis of studies using Flickr data","fulltext":[{"header":"Introduction","content":"\u003cp\u003eMuch of tourism studies, like tourism geographies, rely on quantitative methods requiring travel-related data. The richness of data sources bought a wide variety of methods and a fragmentation of theory. Official statistical data on accommodation occupancy, air flights and border crossings are the globally available resources that give a unified basis for tourism researchers, but such data has low geographical and contextual accuracy. Main attractions have visitor data from ticket selling, but most tourists’ journeys remained hard to trace, if not with manual surveying methods, like questionnaires. Until the widespread use of social media, only large corporations had massive geographical data on the geographical consumption of travelling users; these are mainly payment card services and mobile phone operators. These corporations make commercial use of their datasets. Therefore, they rarely let independent researchers use their data, even though fruitful collaborations with privileged or well-funded scientific communities did result in some very relevant papers (Ahas et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2007\u003c/span\u003e; Sobolevsky et al., \u003cspan citationid=\"CR84\" class=\"CitationRef\"\u003e2014\u003c/span\u003e). Other researchers had to rely on traditional surveys (Hwang \u0026amp; Fesenmaier, \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2003\u003c/span\u003e), specialized digital apps collecting data from a very limited number of tourists (Dickinson et al., \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2014\u003c/span\u003e), or data loggers distributed to tourists (Mckercher et al., \u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e2012\u003c/span\u003e) – also a very resourceful data collection method. Researchers did explore every opportunity that new services or technologies offered (Shoval \u0026amp; Ahas, \u003cspan citationid=\"CR82\" class=\"CitationRef\"\u003e2016\u003c/span\u003e), but such methods are hard to reproduce by other groups of scientific community. Therefore tourism studies, especially geographic and other quantitative studies, seem to be an endless production series of isolated case studies (Ashworth \u0026amp; Page, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2011\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eSocial media, as the evolution of World Wide Web services, promised an open stream of data created by users, and shared by all other users of the internet. Suddenly, text, images, videos and geolocated content produced by individual users became available in unseen quantities, accessible to all and therefore, to researchers as well. The “Data Golden Age” in digital research lasted for a decade until the Cambridge Analytica (CA) scandal of 2018. Until then, most of the Application Programming Interfaces (APIs) of social media services provided easily downloadable data, meant to ease the communication between social media and external applications using it for commercial purposes, but available also for academic researchers. As CA used sensitive user data mined by scientists for unethical commercial use, Facebook and most large service providers shut down the openly accessible features of their APIs, causing a difficult situation for independent academic researchers mining such datasets (Perriam et al., \u003cspan citationid=\"CR76\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Tromble, \u003cspan citationid=\"CR92\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eTourism studies is one of the fields that benefitted from the open APIs of social media. Tracking tourists and analyzing destinations and tourist behaviour from travel-related images uploaded to Panoramio (launched in 2005, discontinued in 2016), Flickr (2004-) and Instagram (2010-) became established research methodology (Ashworth \u0026amp; Page, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2011\u003c/span\u003e; Shoval \u0026amp; Ahas, \u003cspan citationid=\"CR82\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). Today large Chinese social networks allow access to the metadata uploaded by their users, while among Western providers, only Flickr maintains its open-source API policy. This contributes to the growing imbalance between the West and the East regarding the access to and use of this type of information in tourism scientific research. Flickr’s popularity faded since it was the most used image-sharing platform in the 2010s. The new generation of social media is owned by corporations (Facebook, Google) even larger than the mobile phone operators or credit card companies, and these handle the immense datasets of their users as their own resources they capitalize on, excluding research communities from analytics. This leads to an unbalanced status in global knowledge, especially true for behavioural sciences and tourism studies, as large enterprises nowadays have a better understanding of travel patterns and user behaviour than academic science has. These datasets enabled researchers worldwide to map, track, and interpret tourist movements and preferences at unprecedented scales. However, with the advent of stricter data privacy regulations and the subsequent APIs restrictions, the access to such open datasets has diminished, creating an urgent need to understand the legacy and future possibilities of social media-derived data for tourism research.\u003c/p\u003e\u003cp\u003eWhat do we miss in tourism studies by losing access to easily researchable social media data? It is more straightforward to analyze what the scientific field gained with large unified datasets available for tourism research, even if the new social media generation could have had even more massive data. Panoramio, Flickr, and, to some extent, Instagram - together with the large Chinese social media sites such as Weibo, providing data mostly from China – provided travel related data in unified formats from millions of users. The advantages of these resources for quantitative research lie not only in the immense amounts of georeferenced images from most destinations, but also in the unified format of such data, leading to interrelated research methodologies and more relevant cross-references between scholars from around the globe in the fields of tourism studies.\u003c/p\u003e\u003cp\u003eThis study addresses the following central research question: how has access to Flickr's open georeferenced photography data shaped the evolution and diversification of research communities in tourism studies, and what lessons can inform future research strategies in an era of restricted data access? In doing so, it critically evaluates the academic significance of open-source data in tourism research and proposes new directions to maintain research vitality in a post-open-data era. Aiming to prove the benefits of a unified global dataset of georeferenced travel photography for the advancement of scientific fields and theories, the central hypothesis is that a significant proportion of published papers in tourism studies, which utilise Flickr as their primary database, are interconnected. These studies build upon each other’s methodologies and findings, thereby fostering the development of new areas within tourism studies. It is argued that such progress would have occurred at a considerably slower pace without the widespread availability of user-generated content (UGC) on a global scale.\u003c/p\u003e\u003cp\u003eTherefore, this systematic and bibliometric review intends to document and synthesize academic research published on tourism studies based on the Flickr social media database, identifying the research fields and results that could emerge in different fields because of the availability of UGC in the form of travel photography.\u003c/p\u003e\u003cp\u003e\u003cb\u003eSocial media and Flick in tourism studies\u003c/b\u003e\u003c/p\u003e\u003cp\u003eTechnological advances and the progressive application of a varied set of tracking techniques in tourism research have made it possible to collect high-resolution georeferenced data about tourists' space-time behaviour, which can be worked on in a GIS (Geographic Information Systems) environment that explore and interpret increasingly large and complex databases (Martins \u0026amp; Costa, \u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Social networks, such as Twitter, Instagram, Panoramio or Flickr, offer an enormous quantity of geotagged photos “through which a user provides spatial (latitudes and longitude coordinates) and temporal (date and time of day) information in the form of coordinates” (Giglio et al., \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). Through the geotagged photos, it is possible to find the sequence of locations visited (Höpken et al., \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) or to provide the best travel routes between popular travel destinations, minimizing the distance while including maximal tourism popularity (Sun et al., \u003cspan citationid=\"CR88\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). The photographs posted by visitors also suggest people’s interests and preferred activities while travelling, revealing where the users have been (Giglio et al., \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2019\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThe main advantage of using social media with geotagged photos in tourism studies is the possibility of using a very large sample where the places can be related to tourist experiences (Caldeira \u0026amp; Kastenholz, 2020). It is not an intrusive method, and the large amount of data collected can be analyzed with less effort in data processing. It is also a low-cost technique, and the unlimited scope of study in terms of geographical scale (local, regional, national and global) makes it an interesting database for tourism research (Martins \u0026amp; Costa, \u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Some disadvantages have also been identified, namely, the potential bias of the sample because only the platform users are tracked, and they also tend to share the most impressive photos and not those of more common environments (Caldeira \u0026amp; Kastenholz, 2020).\u003c/p\u003e\u003cp\u003eAs social media has become omnipresent among travellers (Curlin et al., \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2019\u003c/span\u003e), tourism researchers have been using it to understand visitors’ spatiotemporal behaviour (Hruška \u0026amp; Pásková, \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Önder et al., \u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e2016\u003c/span\u003e), also trying to understand the role of this type of social media in tourists’ travel behaviour patterns (Giglio et al., \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Paldino et al., \u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). Photo analysis in social media has been used in tourism research to estimate visitor trajectories (Mor \u0026amp; Dalyot, \u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e2020\u003c/span\u003e), to find provenance and patterns of recreation (Sinclair et al., \u003cspan citationid=\"CR83\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) and tourist behaviour patterns (Yang et al., \u003cspan citationid=\"CR99\" class=\"CitationRef\"\u003e2017\u003c/span\u003e), to propose or assess tourist routes (Kurashima et al., \u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2013\u003c/span\u003e), to identify social events such as parades, protests, sports, festivals (Yeran \u0026amp; Fan, \u003cspan citationid=\"CR100\" class=\"CitationRef\"\u003e2014\u003c/span\u003e), to measure the tourist activities in cities (Kádár, \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2014\u003c/span\u003e), natural areas (Wood et al., \u003cspan citationid=\"CR98\" class=\"CitationRef\"\u003e2013\u003c/span\u003e), or larger regions (Girardin et al., \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2007\u003c/span\u003edár \u0026amp; Gede, 2021; Önder et al., \u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e2016\u003c/span\u003e), to identify areas with a concentration of tourists (Kádár \u0026amp; Gede, \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2013\u003c/span\u003e) or to determine the attractiveness of various tourism sites (Giglio et al., \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2019\u003c/span\u003e) among others. Most of these studies relied on the datasets of Flickr, as this service provider always kept its API open for external researchers.\u003c/p\u003e\u003cp\u003eFlickr was developed by Ludicorp, a Vancouver, Canada-based company founded in 2004 by Stewart Butterfield and Caterina Fake. According to Broz (\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2022\u003c/span\u003e), it is one of the most popular sites worldwide, with around 25\u0026nbsp;million photos uploaded to Flickr every day, ranking top 400 globally, top 290 in the USA, and top 2 in its category – photography. In February 2017, the site hosted approximately 13\u0026nbsp;billion photos from 122\u0026nbsp;million users that come from 72 countries (USA, 31.03%, UK, 9.83% and Germany, 5.26%), getting up to 60\u0026nbsp;million visits per month. Even if male users and well-educated young and middle-aged people (60.73%) are slightly overrepresented, and older age groups are slightly underrepresented, it is still the platform with one of the most balanced social distributions of users (Kádár \u0026amp; Gede, \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2013\u003c/span\u003e).\u003c/p\u003e"},{"header":"Methodology","content":"\u003cp\u003eThis paper was based on the research of scientific documents available on the Scopus database concerning the research trends, patterns and the research gaps on the Flickr usage in tourism research, until 15th September 2023. The search and collection of information were carried out in the period 2010–2023 using the scheme code “tourism” or “tourist” and “flickr” or \"geotagged photo\" contained in titles, abstracts, and keywords. Scopus database returned 365 documents in the initial search and through a selection process all books and chapters (n = 11), editorials, notes, letters (n = 12) and all documents not written in English (n = 9) were excluded. Finally, 333 articles were selected for bibliometric and content analysis, and all the references cited in the papers have been analyzed. References have been downloaded as a continuous text field for each paper. Therefore, a simple algorithm had to be written in order to create single data entries for all papers referenced by the source papers. All papers have been given unique names following the APA citation style. 12986 citations have been analyzed, but 8702 connections have been discarded as these had only single occurrences, meaning that the paper cited did not appear to be cited by any other papers. 897 papers have been kept, with at least 2 citations in the pool, resulting in a total of 4284 connections between papers. Such methodology allowed for the analysis of a larger ecosystem of the research community working with the topic of social media photos and tourism, including all papers that influenced the work of those in the original query. After this first stage of database creation the 2nd stage of bibliometric analysis and the 3rd stage of content analysis followed (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eBibliometric methods were used through a co-occurrence analysis, allowing a systematic, transparent, and reproducible review process (Eck \u0026amp; Waltman, 2016). The performance evaluation combined with science mapping would allow the identification and visualization of research fields (van Eck \u0026amp; Waltman, \u003cspan citationid=\"CR95\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). While most similar studies used VOSviewer (Herrera-Franco et al., \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2020\u003c/span\u003e), this bibliometric analysis needed more customization and control in building and analyzing the initial dataset, while in-depth content analysis of the papers needed more than just keyword occurrence analysis. Therefore Gephi 0.10.1 software was used to construct and visualize the bibliometric networks related with Flickr and tourism, aided by data processing in excel.\u003c/p\u003e\u003cp\u003eA network consisting of 897 nodes (all papers downloaded from Scopus related to tourism and Flickr, plus all papers at least 2 of these have cited) and 4284 edges (all citations) have been created in Gephi 0.10.1. 564 nodes only have in-degree, as these were not part of the original pool, and their citations have not been analyzed, while at least 2 papers from the original 333 did cite them (max = 47). 223 nodes only have out-degree, meaning that these papers have not been cited by any other papers in the pool, but they have cited others. These were papers from the original source pool of Scopus containing the keywords, many of which are from recent years, making it impossible to cite. 110 papers have both in- and out-degree; these are supposed to be the most relevant papers in the network.\u003c/p\u003e\u003cp\u003eIn Gephi ForceAtlas 2, a network visualization showing a linear attraction gravity model between nodes has been used (Jacomy et al., \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2011\u003c/span\u003e), showing which papers are interconnected between each other’s by citations.\u003c/p\u003e\u003cp\u003eFigure 2. The bibliographic network of papers related to tourism and social media images (compiled in Gephi 0.10.1.)\u003c/p\u003e\u003cp\u003eAmong their position, the in-degree (number of citations from the network to the node representing a paper) and out-degree (number of citations from the paper represented by the node to other nodes in the graph) are the metrics most relevant describing the singular nodes. In Fig.\u0026nbsp;2 the size of the nodes reflects their in-degree, while in the darkness of their color correlates with higher out-degree. High out-degree papers are the ones that build most on the results of other works represented in the graph, therefore these papers were worth to be content analyzed in stage three. These papers should show most accurately the commons among topics and methodologies used in research communities represented by different sections of the graph. High in-degree papers are the most influential ones, defining the topics or methodologies used in many other papers of the research communities. Therefore, also these were worth to be analysed in the third stage.\u003c/p\u003e\u003cp\u003eThe connections – citations – between papers have been coloured to show which year the work cited had been published. Therefore, the temporal development of the network could be visualized. The earliest papers cited by more than one author are from the beginning of the ‘1960s (mathematical algorithms), and it is well visible that separate clusters in the network have different timelines in their activities. To define the different communities with similar topics or methodologies of research, a cluster analysis has been applied to the network using the community detection method of Blondel et al. (\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2008\u003c/span\u003e) integrated into Gephi. We assumed that papers focusing on different fields of study can be detected as different clusters of the network, having the most citations between each other (Fig.\u0026nbsp;3).\u003c/p\u003e\u003cp\u003eAs a third stage, the content analysis of the papers was made to show the specific scientific field developed in the revealed clusters. SciSummary, an AI-based tool developed to summarize scientific papers, was used to have an overview of all papers in the network. The assumption that papers cited by high out-degree papers or citing high in-degree papers inside a cluster follow the same topic as such high-degree papers has been largely verified; therefore, cluster detection and the manual analysis of high-degree nodes proved to be an effective way to define the topics present in a large bibliographic system. High in-degree papers are the most influential ones, delivering the bases of the methodologies or topics of the cluster, while high out-degree papers best summarize the overall topic of the cluster, as they gather in their references most of the influential works defining the cluster.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cb\u003eBibliometric Analysis to define tourism research communities\u003c/b\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eThe dataset of nodes consisting of 333 source papers and 564 papers cited by at least two source papers and of edges consisting of 4284 citations in total have been uploaded to Gephi 0.10.1, and analyzed. The resulting network visualized with a ForceAtlas 2 algorithm (Fig.\u0026nbsp;3) already showed strong clustering with large sub-graphs concentrated around some highly cited papers, distancing from the main network in structure but also in the average timeline of its publications.\u003c/p\u003e\u003cp\u003eFigure 3. The 7 clusters found with Gephi in the bibliographic network (compiled in Gephi 0.10.1.)\u003c/p\u003e\u003cp\u003eThe cluster analysis (Blondel et al., \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2008\u003c/span\u003e) was made with a resolution of 0.8, resulting in a modularity score of 0.512, with 7 distinct communities with more than 50 nodes (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e4\u003c/span\u003e). It must be noted, that the clustering algorithm never gives two identical dispositions in this graph, but modifying the resolution above 1.1 and below 0.8 always resulted in modularity scores Q\u0026thinsp;\u0026lt;\u0026thinsp;0.5, while inside this threshold this score was 0.5\u0026thinsp;\u0026lt;\u0026thinsp;Q\u0026thinsp;\u0026lt;\u0026thinsp;0.515 (Q\u0026thinsp;=\u0026thinsp;0 is no separation at all, Q\u0026thinsp;=\u0026thinsp;1 is total separation to another network). Selecting the lower resolution while still maintaining a high modularity leads to the detection of communities that have strong interrelations but are more hidden than the basic four that one can detect just from the primary form and structure of the graph.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eFigure \u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e4\u003c/span\u003e shows the weighted network representation of the 7 clusters, with the weighted in- and out-degrees (K\u0026aacute;d\u0026aacute;r \u0026amp; Gede, \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). References inside the clusters are represented as self-loops, and the 5 large clusters have significantly more self-references than in- or out-degrees. However, C4 has almost as many references from other clusters as inside references, while in the case of C6, both in- and out-degree are higher than inside references. In fact, these two clusters have not been detected with the resolution\u0026thinsp;=\u0026thinsp;1, as their separation from the rest is not as obvious, but still the references to - and from - other clusters remain much lower one by one than the number of inside self-references. It can be seen that C1, as the central core cluster, is the most connected to others, especially to C2, while C7 is the most isolated from others.\u003c/p\u003e\u003cp\u003eThe 7 communities have been analyzed. Specifically, the papers with higher in-degree and out-degree values have been content analyzed, and therefore, seven different research topics or methodologies have been identified.\u003c/p\u003e\u003cp\u003e\u003cb\u003eAnalysis and Discussion\u003c/b\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003eC1 The community measuring tourist activities by spatiotemporal analysis\u003c/b\u003e\u003c/p\u003e\u003cp\u003eThe graph is structured around a central cluster (C1; n\u0026thinsp;=\u0026thinsp;197), defined by many strong nodes like Girardin, Dal Fiore, et al. (\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2008\u003c/span\u003e), Vu et al. (\u003cspan citationid=\"CR96\" class=\"CitationRef\"\u003e2015\u003c/span\u003e), Garc\u0026iacute;a-Palomares et al. (\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2015\u003c/span\u003e), Zheng et al. (\u003cspan citationid=\"CR105\" class=\"CitationRef\"\u003e2012\u003c/span\u003e)d d\u0026aacute;r) (2014). This is the most fluid cluster as nodes at its boundaries tend to be tied equally strongly to the neighbouring clusters; in fact most of the other clusters have their most influential nodes also connected to this central cluster. This fluid cluster is in the centre of the graph because the topics defined in it correlate most to the original topics searched in the Scopus database: tourism, geo-tagged photos and Flickr. The earliest influential work in this cluster is by Girardin and colleagues (Girardin, Calabrese, et al., \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2008\u003c/span\u003e; Girardin, Dal Fiore, et al., \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2008\u003c/span\u003e), who first presented a comprehensive methodology to define tourist movements in different territories and tourist typologies deriving from the spatiotemporal position of geo-tagged Flickr photographs.\u003c/p\u003e\u003cp\u003eMost papers in this central community compare the behaviour of different user groups regarding the tourism performance of various locations (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Their themes are connected by methodology and the ambition to extract tourism-related statistics from geo-tagged photos, but not by a specific research focus, as seen in the other clusters.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eTourist activities by spatiotemporal analysis community\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"4\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePaper\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eAim\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eIn-/Out-degree\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eTheme/topic\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Girardin, Dal Fiore, et al., \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2008\u003c/span\u003e; Girardin et al., 2009; )\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eMining travel patterns from geotagged photos, differentiating between tourists and locals\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e41/0\u003c/p\u003e\u003cp\u003e10/0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eMeasuring tourist activity\u003c/p\u003e\u003cp\u003eDifferentiating user groups\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Zheng et al., \u003cspan citationid=\"CR105\" class=\"CitationRef\"\u003e2012\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eMining travel patterns from geotagged photos\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e39/17\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eMeasuring tourist activity\u003c/p\u003e\u003cp\u003eDifferentiating user groups\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Vu et al., \u003cspan citationid=\"CR96\" class=\"CitationRef\"\u003e2015\u003c/span\u003e, \u003cspan citationid=\"CR97\" class=\"CitationRef\"\u003e2018\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eExploring the travel behaviours of inbound tourists to Hong Kong using Flickr geotagged photos (2015), Travel Diaries Analysis by Sequential Rule Mining (2018)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e38/29\u003c/p\u003e\u003cp\u003e6/30\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eUrban tourism\u003c/p\u003e\u003cp\u003eMeasuring tourist activity\u003c/p\u003e\u003cp\u003eDifferentiating user groups\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Garc\u0026iacute;a-Palomares et al., \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2015\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eIdentification of tourist hot spots based on social networks: A comparative analysis of European metropolises using photo-sharing services and GIS\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e33/0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eUrban tourism\u003c/p\u003e\u003cp\u003eMeasuring tourist activity\u003c/p\u003e\u003cp\u003eAttractive places in cities\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(K\u0026aacute;d\u0026aacute;r, \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2014\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eMeasuring tourist activities in cities using geotagged photographs\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e23/17\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eUrban tourism\u003c/p\u003e\u003cp\u003eMeasuring tourist activity\u003c/p\u003e\u003cp\u003eDifferentiating user groups\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Hu et al., \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2015\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eExtracting and understanding urban areas of interest using geotagged photos\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e23/0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eUrban tourism\u003c/p\u003e\u003cp\u003eMeasuring tourist activity\u003c/p\u003e\u003cp\u003eAttractive places in cities\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Li et al., \u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e2018\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eAnalyzing and visualizing the spatial interactions between tourists and locals: A Flickr study in ten US cities\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e20/18\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eUrban tourism\u003c/p\u003e\u003cp\u003eMeasuring tourist activity\u003c/p\u003e\u003cp\u003eDifferentiating user groups\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Yuan \u0026amp; Medel, \u003cspan citationid=\"CR101\" class=\"CitationRef\"\u003e2016\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eCharacterizing international travel behaviour from geotagged photos: A case study of Flickr\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e14/18\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eMeasuring tourist activity\u003c/p\u003e\u003cp\u003eDifferentiating user groups\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(\u0026Ouml;nder et al., \u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e2016\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eTracing Tourists by Their Digital Footprints: The Case of Austria\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e12/18\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eMeasuring tourist activity\u003c/p\u003e\u003cp\u003eDifferentiating user groups\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Straumann et al., \u003cspan citationid=\"CR87\" class=\"CitationRef\"\u003e2014\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eTowards (re)constructing narratives from georeferenced photographs through visual analytics\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e10/16\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eMeasuring tourist activity\u003c/p\u003e\u003cp\u003eImage content analysis\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eThe studies in C1 reveal that the temporal dynamics have been underexplored, i.e., most studies treat Flickr data as static rather than capturing evolving tourist flows. Therefore, some future research possibilities emerge, such as leveraging longitudinal Flickr datasets to analyse temporal evolution of tourist patterns (e.g., pre- and post-COVID impacts). While this cluster pioneered the large-scale mapping of tourist flows (e.g., Girardin et al., \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2008\u003c/span\u003ed\u0026aacute;r, 2014), it often treated tourists as homogenous agents. Using the profile data of Flickr users, future research should incorporate socio-demographic segmentation to better reflect the diversity of tourist experiences and motivations. Additionally, integrating sentiment analysis with spatiotemporal data could reveal not just movement patterns, but emotional geographies of tourism.\u003c/p\u003e\u003cp\u003e\u003cb\u003eC2 The travel route detection and clustering community\u003c/b\u003e\u003c/p\u003e\u003cp\u003eThe large community consisting of 195 papers can be considered one of the oldest as most work ranges from 1995 to 2019, with only 8 papers from the 2020\u0026rsquo;s. However, the main topic in the cluster \u0026ndash; travel route recommendation \u0026ndash; continues with newer works in the C3 community, much connected to this one (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). The earliest works cited in this community (some from the \u0026rsquo;1960s and \u0026rsquo;70s) are related to clustering algorithms. Based on the work of Ester et al. (\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e1996\u003c/span\u003e), many authors use the DBSCAN (H\u0026ouml;pken et al., \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Miah et al., \u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Zheng et al., \u003cspan citationid=\"CR105\" class=\"CitationRef\"\u003e2012\u003c/span\u003e) or similar algorithms. The creation of travel route recommendation systems seemed to be the first practical outcome of researching geo-tagged social media photos, but large internet companies e.g. Google, could develop such services within their databases and closed company profiles; this is also a reason why this topic doesn\u0026rsquo;t have contemporary references. In fact, the travel route recommendation seems to be the most connecting topic, but in reality, the clustering of tagged photos, either by textual tags, geo-tags or image content, is the methodology that connects these works. Image content detection methods are also used mostly in this cluster, citing e.g. Lowe (\u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e2004\u003c/span\u003e) for methodology. The early works of Kennedy et al. (\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e2007\u003c/span\u003e) do not mention travel route recommendation, only clustering methods to describe places, just as the very influential work of Kisilevich et al. (\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e2010\u003c/span\u003e), also working with DBSCAN. Most papers cite Crandall et al. (\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2009\u003c/span\u003e), who employed a non-parametric clustering method named mean-shift to discover the significant landmarks/hotspots at a global level, providing a quick overview of the interesting places at travel destinations and demonstrating that visual and temporal features improved the ability to predict the photo location, compared to using textual features alone. The interplay between structure and content makes this paper (\u0026ldquo;Mapping the World\u0026rsquo;s Photos\u0026rdquo;) a very influential reference on tourist recommendation systems, travel route recommendation, or Points of Interest (POI\u0026rsquo;s) identification/recommendation.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eTravel route detection and clustering community\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"4\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePaper\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eAim\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eIn-degree /Out-degree\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eTheme/topic\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Crandall et al., \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2009\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003ePersonalized urban trip recommendations for tourists based on user interests, points of interest, visit durations, and visit recency.\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e49/0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eTravel route recommendation\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Kisilevich et al., \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e2010\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eClustering automatically popular places.\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e49/0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eTravel route detection\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Ester et al., \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e1996\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eClustering algorithms in theory. (no Flickr at all).\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e34/0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eClustering algorithms\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Kurashima et al., \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e2010\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eTravel recommendation itineraries based on present location and preferences.\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e23/18\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eTravel route recommendation\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Majid et al., \u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e2013\u003c/span\u003e; Memon et al., \u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e2015\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eTravel recommendation system based on previous visits.\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e23/18\u003c/p\u003e\u003cp\u003e16/5\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eTravel route recommendation\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Popescu et al., \u003cspan citationid=\"CR78\" class=\"CitationRef\"\u003e2009\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eUrban trip extraction with inner stays (museum) and panoramic points: how long is a stay/walk?\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e23/6\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eTravel route detection\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Sun et al., \u003cspan citationid=\"CR88\" class=\"CitationRef\"\u003e2015\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eRoad-based travel recommendation.\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e22/30\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eTravel route recommendation\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Kurashima et al., \u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2013\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eTravel route recommendation using geotagged photos.\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e17/22\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eTravel route recommendation\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Zhou et al., \u003cspan citationid=\"CR106\" class=\"CitationRef\"\u003e2015\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eDetecting tourism destinations using scalable geospatial analysis based on cloud computing platform.\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e15/18\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eTravel route detection\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Yang et al., \u003cspan citationid=\"CR99\" class=\"CitationRef\"\u003e2017\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eQuantifying tourist behaviour patterns by travel motifs and geo-tagged photos from Flickr.\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e11/24\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eTravel route detection\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eThis cluster advanced route optimisation techniques using clustering algorithms like DBSCAN, and rarely engaged with the cultural implications of algorithmically proposed itineraries. A predominantly computational focus led to a technocentric view of travel, neglecting how tourists subjectively experience routes, serendipity, and intangible elements like safety perception or urban ambience. Therefore, the routes are optimised for efficiency, neglecting tourists\u0026rsquo; desire for exploration and cultural immersion. Future research could combine Flickr photos with natural language processing (NLP) analysis of photo descriptions/titles to incorporate emotional and experiential dimensions into route recommendations. Regarding the dynamic route prediction, with the advances in computing power capacity, it would be interesting to use real-time or near-real-time Flickr uploads to adjust recommended routes based on emerging visitor flows.\u003c/p\u003e\u003cp\u003e\u003cb\u003eC3 The travel route recommendation and orienteering community\u003c/b\u003e\u003c/p\u003e\u003cp\u003eAs opposed to C2, all papers in C3 (n\u0026thinsp;=\u0026thinsp;113) work on travel recommendation methods, with a methodological focus on the Orienteering Problem (Lim et al., \u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e2018\u003c/span\u003e) using socially generated knowledge and focusing on user metadata where the problem was modelled as a single graph of POIs, their popularities and the transit times among them (Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). In fact, the most representative researcher is K. H. Lim, who co-authored 8 papers in this cluster. De Choudhury et al. (\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2010\u003c/span\u003e) authored the most cited and earliest paper, the one connecting this community most to C2: they studied and proposed a tour recommendation using the orienteering problem, which starts with a particular POI and ends with a different POI, while ensuring that the tour can be completed within a certain time. The orienteering problem is the methodological basis connecting all work in this community, all proposing effective tour recommendation algorithms.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eTravel route recommendation and orienteering community\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"4\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePaper\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eAim\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eIn-/Out- degree\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eTheme/topic\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(De Choudhury et al., \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2010\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eConstruct intra-city travel itineraries modelled as a single graph of POIs automatically using Flickr photos.\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e38/7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003ePersonalized trip recommendation; Orienteering Problem\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Lim et al., 2015, \u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e2018\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003ePersonalized tour recommendation based on user interests, points of interest, visit durations and recency\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e7/44\u003c/p\u003e\u003cp\u003e7/22\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003ePersonalized trip recommendation\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Sarkar et al., \u003cspan citationid=\"CR81\" class=\"CitationRef\"\u003e2020\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eTo recommend multiple urban itineraries based on the tourist\u0026rsquo;s interest, the popularity of itineraries and the cost associated with these itineraries\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e5/32\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eTravel itineraries recommendation\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Mou et al., \u003cspan citationid=\"CR70\" class=\"CitationRef\"\u003e2022\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eTo propose tourist route recommendations using a personalized recurrent neural network (P-RecN) in Shanghai..\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0/37\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eTravel route recommendation\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Korakakis et al., \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2017\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eTo automatically construct travel routes, i.e., ordered visits to various places-of-interest in Greek cities.\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0/36\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eTravel route recommendation\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Mor et al., \u003cspan citationid=\"CR69\" class=\"CitationRef\"\u003e2021\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eTo generate tourism walking routes in New York and evaluate them in terms of the visible space.\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0/27\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eEvaluating the attractiveness of Tourism Routes\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Lim, \u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e2016\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eDeveloping algorithms for recommending personalized city tours to both individual travellers and groups of tourists based on their interest preferences (based on Flickr and Wikipedia)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0/24\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003ePersonalized tours recommendation\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Mor \u0026amp; Dalyot, \u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e2020\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eUsing Flickr photos as\u0026nbsp;a\u0026nbsp;source\u0026nbsp;for\u0026nbsp;mining\u0026nbsp;users\u0026rsquo;\u0026nbsp;trajectories\u0026nbsp;to compute\u0026nbsp;walking\u0026nbsp;tourism\u0026nbsp;routes effectively based on tourist activities in Tel-Aviv, Israel and Manhattan, USA.\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0/24\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eTour recommendation system\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eHighly technical, C3 modeled the tourism experience as an optimization problem. However, the assumption that the \u0026lsquo;optimal\u0026rsquo; itinerary is universally desirable ignores heterogeneous tourist profiles (family tourists, solo adventurers, elderly travellers, among others), and often neglects qualitative dimensions of travel, such as serendipity and personal meaning-making. Future advances should blend algorithmic efficiency with models that account for experiential richness and tourist agency, and develop segmented itinerary models reflecting diverse tourist personas, using Flickr data enriched by inferred user profiles (e.g., based on metadata about travel duration, locations visited).\u003c/p\u003e\u003cp\u003e\u003cb\u003eC4 The community measuring the attractivity of places\u003c/b\u003e\u003c/p\u003e\u003cp\u003eC4 is the smallest community (n\u0026thinsp;=\u0026thinsp;50) around various works of I. Bojic. The most influential paper of this researcher was co-authored with Paldino et al. (\u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e2015\u003c/span\u003e) entitled \u0026ldquo;Urban Magnetism through the Lens of Geo-Tagged Photography\u0026rdquo;. In fact, the attractiveness of places is the main topic of this cluster, often combining Flickr data with other more qualitative data sources to explore why some places are more attractive than others (Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab4\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eAttractivity measure of places community\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"4\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePaper\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eAim\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eIn-/Out-degree\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eTheme/topic\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Paldino et al., \u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e2015\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eMeasuring the attractiveness of 10 different cities using Flickr Photos from residents and tourists, tracing the hotspots and mobility networks.\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e17/15\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eAttractivity of places in cities\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Karayazi et al., \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2021\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eUnderstanding the heritage attractiveness in Amsterdam by combining clusters of attractions from Flickr with heritage data and applying regression analyses to identify influencing factors.\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e3/14\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eAttractivity of heritage in a city\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Gonz\u0026aacute;lez et al., \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2008\u003c/span\u003e) (a)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eAnalysing the trajectory of 100,000 anonymized mobile phone users for a six-month period, showing simple patterns.\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e11/0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eTracking travel patterns through mobile phone data\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Hawelka et al., \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2014\u003c/span\u003e) (a)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eAnalysing geo-located Twitter data for global mobility patterns\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e11/0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eTracking travel patterns through Twitter data\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Valls \u0026amp; Roca, \u003cspan citationid=\"CR94\" class=\"CitationRef\"\u003e2021\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eMeasuring visitor hotspots in Barcelona, employing visualization methods on different scales\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0/24\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eAttractivity of places in cities\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Bojic et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2016\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eInvestigating the attractiveness of countries' large-scale composite regions using Flickr data from foreign visitors and migration data from the UN.\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0/16\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eAttractivity of regions\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Jain \u0026amp; Singh, \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2017\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eInvestigating the attractiveness character of most checked-in countries and large-scale composite regions\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0/16\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eAttractivity of regions\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e(a) Articles where Flickr is not used but with relevant methodological contributions.\u003c/p\u003e\u003cp\u003eThis cluster revealed key insights into the magnetism of cities and regions. However, most studies were cross-sectional, capturing attractiveness at a single moment in time. Longitudinal analyses that trace how attractiveness evolves (especially in response to marketing, climate change, or socio-political shifts) remain a promising avenue for deeper understanding. Besides that, seasonal and event-based fluctuations in attractiveness are also under-researched.\u003c/p\u003e\u003cp\u003e\u003cb\u003eC5 The destination image research community\u003c/b\u003e\u003c/p\u003e\u003cp\u003eThis cluster (n\u0026thinsp;=\u0026thinsp;95) is well separated from the central core and has the most relatively old papers from the \u0026lsquo;1990s and before (Fig.\u0026nbsp;3), organized around different influential works (Deng et al., \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Donaire et al., \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Stepchenkova \u0026amp; Zhan, \u003cspan citationid=\"CR86\" class=\"CitationRef\"\u003e2013\u003c/span\u003e; K. Zhang et al., \u003cspan citationid=\"CR103\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). Most work in this cluster analyzes the image of destinations (Bhatt \u0026amp; Pickering, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Deng et al., \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Lin et al., \u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Mangachena et al., \u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; K. Zhang et al., \u003cspan citationid=\"CR104\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). One of the most central and influential papers by Stepchenkova \u0026amp; Zhan) (2013) compared images of Peru collected from a DMO\u0026rsquo;s website and Flickr, comparing how projected destination image(s) used in marketing match the tourist\u0026rsquo;s destination image. The qualitative research to collect tourists\u0026rsquo; attitudes, opinions and emotions about the destinations using mainly quantitative databases of online photos is complex (Lin et al., \u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) because the meanings embedded in it is subjective to researchers\u0026rsquo; interpretation. Therefore, this cluster has the most references to works related to the understanding of tourist photography as a phenomenon (Jenkins, \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2003\u003c/span\u003e; Pan et al., \u003cspan citationid=\"CR75\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Urry, \u003cspan citationid=\"CR93\" class=\"CitationRef\"\u003e1990\u003c/span\u003e) and destination image formation (Garrod, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2009\u003c/span\u003e), explaining the high number of older references, all rooted in the qualitative analysis of destinations (Table\u0026nbsp;\u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e5\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eTo analyse destination image (DI), content analysis of the photos is a must. K. Zhang et al. (\u003cspan citationid=\"CR103\" class=\"CitationRef\"\u003e2019\u003c/span\u003e) start it with the quantification of specific themes and attributes presented in the images, followed by the identification of the main focal items as well as their frequencies, co-occurrence, clustering, and other related issues to be recorded, as done also by Stepchenkova \u0026amp; Zhan (\u003cspan citationid=\"CR86\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). In this diverse cluster, therefore, there are many older papers giving theoretical background, newer ones using content analysis and even others using machine learning techniques, are mixed. The Flickr related research topics can be retrieved from the content of the newer papers in the cluster.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab5\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 5\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eDestination image research community\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"4\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePaper\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eAim\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eIn-/Out- degree\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eTheme/topic\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Donaire et al., \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2014\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eIdentifying the attributes of photos and determining the existence of four different photographer's behaviour\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e20/27\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eDestination image\u003c/p\u003e\u003cp\u003eContent and cluster analysis of online photographs\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Zhang et al., \u003cspan citationid=\"CR103\" class=\"CitationRef\"\u003e2019\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eUsing deep learning technology, the visual contents of photos were identified, and the perception and behavioural preferences of tourists from different countries were analyzed.\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e17/39\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eVisual content analysis\u003c/p\u003e\u003cp\u003eTourists' behaviours and perception\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Stepchenkova \u0026amp; Zhan, \u003cspan citationid=\"CR86\" class=\"CitationRef\"\u003e2013\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eComparing images of Peru collected from a DMO\u0026rsquo;s website and Flickr, identifying statistical differences in several dimensions.\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e18/16\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eContent analysis of online photographs\u003c/p\u003e\u003cp\u003eDestination image\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Deng \u0026amp; Li, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2018\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eUsing Flickr photos of New York City visitors to propose and implement a machine learning-based model to rank photos describing a specific theme from viewers\u0026rsquo; perspective.\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e10/26\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eDestination image\u003c/p\u003e\u003cp\u003eContent analysis of online photographs\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Deng et al., \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2019\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eProposing a new method to compare destination image (DI) differences among inbound tourists in Shangai, China.\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e8/38\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eDestination image\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(K. Zhang et al., \u003cspan citationid=\"CR104\" class=\"CitationRef\"\u003e2020\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eUsing user-generated photos and artificial intelligence computer vision technologies to identify the differences in the perceived destination image and behavioural patterns between residents and tourists in Hong Kong.\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e7/32\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eDestination image\u003c/p\u003e\u003cp\u003eComputer vision\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Bhatt \u0026amp; Pickering, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2022\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eExploring the uses of Flickr photos posted by Chitwan National Park, (Nepal) visitors, comparing the content of those photos (perceived image(s)) with those posted by tourism organizations online (projected destination image(s)).\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0/43\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eContent analysis of online photographs\u003c/p\u003e\u003cp\u003eDestination image\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Lin et al., \u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e2021\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eUsing Flickr photos to investigate the unique merits and biases of social media analytics (SMA) and a traditional visitor intercept survey\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0/29\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eDestination image\u003c/p\u003e\u003cp\u003e(Content analysis, machine learning and text analysis)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Mangachena et al., \u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e2022\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eAssessing how Africa is presented in images seen by tourists and how Africa is represented in destination image(s).\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0/29\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eDestination image\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eThe use of Flickr to contrast perceived vs. projected destination images was a methodological innovation. However, studies often failed to theorize how digital images mediate memory and perception. Future research should build interdisciplinary bridges to visual studies and cognitive science to interpret photographic tourism narratives better. In addition, there is a risk of reifying stereotypes, i.e., repetitive photographic patterns (e.g., iconic landmarks) may reinforce narrow destination images, limiting tourists\u0026rsquo; perceptions of diversity. Future research should deepen the use of sentiment analysis to photo captions and tags to complement visual analysis, uncovering affective dimensions of destination image, and explore how Flickr users depict less-promoted or alternative aspects of destinations, challenging the dominant projected images.\u003c/p\u003e\u003cp\u003e\u003cb\u003eC6 The community using deep learning methods\u003c/b\u003e\u003c/p\u003e\u003cp\u003eC6 is a smaller cluster (n\u0026thinsp;=\u0026thinsp;63) dispersed between C1, C2, and C5, found only by lower resolution settings in the clustering algorithm, represented somehow by the recent work of Cho and Kang (Cho et al., \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Kang et al., \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). This is a community of researchers connected not by a specific research topic, but by methodology, where all work uses machine learning algorithms trained by and working with Flickr databases, but along different topics (Table\u0026nbsp;\u003cspan refid=\"Tab6\" class=\"InternalRef\"\u003e6\u003c/span\u003e). The nodes of this cluster are spread around the centre of the graph, connecting nearly to all other clusters in the topic, but interconnected inside the community by methodology \u0026ndash; training machine learning algorithms with Flickr photos to detect and classify the content of photos. Some papers do exist outside of this cluster using machine learning technology, but those are more strictly connected to the topic of other clusters. Therefore, they do not appear in this cluster, where the methodological aspects of machine learning prevail over the thematic similarities.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab6\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 6\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eDeep learning methods community\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"4\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePaper\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eAim\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eIn-degree /Out-degree\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eTheme/topic\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Hollenstein \u0026amp; Purves, \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2010\u003c/span\u003e; Kisilevich et al., \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e2013\u003c/span\u003e; Rattenbury et al., 2009)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eEarly methods to get the semantics of places from geotagged photography before using machine learning\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e17/0\u003c/p\u003e\u003cp\u003e15/0\u003c/p\u003e\u003cp\u003e10/0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eTags associated with Flickr images as a proxy for empirical data\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Krizhevsky et al., \u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e2012\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eImageNet classification with deep convolutional neural networks\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e6/0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eDeep convolutional neural network\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Chen et al., \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2019\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003ePresenting several data processing methods for inbound tourist flow prediction in Beijing (China) validated through data correlation analysis and machine learning algorithm predictions\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e5/25\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eMachine learning algorithm\u003c/p\u003e\u003cp\u003eTourist flow prediction\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Kim et al., \u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e2020\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eAnalyzing representative images and elements of sightseeing attractions using photos uploaded on Flickr by Seoul tourists,using convolutional neural networks.\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e5/12\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eDeep learning\u003c/p\u003e\u003cp\u003ePhoto\u0026rsquo;s characteristics analysis\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Cho et al., \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2022\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eClassifying Tourists\u0026rsquo; Photos and Exploring Tourism Destination Image Using a Deep Learning Model\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0/37\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eConvolutional Neural Network\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Kang et al., \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2021\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eClassifying a large volume of Flickr photos with transfer learning of a deep learning model for exploring tourists\u0026rsquo; urban images.\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0/29\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eConvolutional neural network\u003c/p\u003e\u003cp\u003eTourists\u0026rsquo; photo classification\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Zhu et al., \u003cspan citationid=\"CR107\" class=\"CitationRef\"\u003e2019\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eUsing fine-grained land use classification at the city scale using ground-level images to better understand what occurs in different parts of a city at fine spatial and activity class scales.\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0/21\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eConvolutional neural networks\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Lee et al., \u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e2015\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003ePredicting geo-informative attributes in large-scale image collections using convolutional neural networks\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0/14\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eDeep convolutional neural networks\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Novack et al., \u003cspan citationid=\"CR72\" class=\"CitationRef\"\u003e2020\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eDetecting building facades with graffiti artwork based on street view images interpreted by a customized, convolutional neural network\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0/11\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eConvolutional neural networks\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eWhile machine learning enhanced the scalability of image content analysis, it also introduced a black-box problem, making it difficult to understand how classifications are made. Studies such as Zhang et al. (\u003cspan citationid=\"CR103\" class=\"CitationRef\"\u003e2019\u003c/span\u003e) highlight this limitation and suggest the need for greater transparency, interpretability, and bias auditing in the models used, especially given the sociocultural implications of automated analyses. The convolutional neural networks (CNNs) also deliver impressive classification results but rarely offer insight into how visual features map onto tourism concepts (e.g., what makes a photo depict \u0026lsquo;authenticity\u0026rsquo;?).\u003c/p\u003e\u003cp\u003e\u003cb\u003eC7 The nature-based tourism research community\u003c/b\u003e\u003c/p\u003e\u003cp\u003eThe large cluster (n\u0026thinsp;=\u0026thinsp;184) that separates most clearly from the others consists of many quite recent works (Fig.\u0026nbsp;3) and is unequivocally dominated by the work of Wood et al. (\u003cspan citationid=\"CR98\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). This influential paper defines the topic of this community in its title: \u0026ldquo;Using social media to quantify nature-based tourism and recreation\u0026ldquo;. All papers treat nature-based tourism and the methods to measure visitor activity in such destinations (Table\u0026nbsp;\u003cspan refid=\"Tab7\" class=\"InternalRef\"\u003e7\u003c/span\u003e). Wood et al. (\u003cspan citationid=\"CR98\" class=\"CitationRef\"\u003e2013\u003c/span\u003e) utilized user-generated content (UGC) to quantify nature-based tourism, approximating visitation rates at 836 recreational sites around the world using Flickr data. Spencer Wood is recognized by others in this cluster to be the most prolific author (Teles da Mota \u0026amp; Pickering, \u003cspan citationid=\"CR89\" class=\"CitationRef\"\u003e2020\u003c/span\u003e), and this paper, in particular, is the most influential on Flickr research of nature-based tourism (H. Zhang et al., \u003cspan citationid=\"CR102\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). The other node that defines most of this cluster is the more recent work authored by Teles da Mota \u0026amp; Pickering (\u003cspan citationid=\"CR89\" class=\"CitationRef\"\u003e2020\u003c/span\u003e); their paper \u0026ldquo;Using social media to assess nature-based tourism: Current research and future trends\u0026rdquo; cites most of the papers in the cluster in order to theorize this topic. Studies in this community use similar methodologies to proxy tourism visitation data per country, comparing the number of images per day per user as done by an even more recent paper by Teles da Mota et al. (\u003cspan citationid=\"CR90\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). National parks, coastal beach areas, and large natural areas are the focus of all of the studies.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab7\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 7\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eNature-based tourism research community\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"4\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePaper\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eAim\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eIn-/Out-degree\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eTheme/topic\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Wood et al., \u003cspan citationid=\"CR98\" class=\"CitationRef\"\u003e2013\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eUsing UGC to quantify nature-based tourism, approximating visitation rates at 836 recreational sites using Flickr data\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e47/3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eNature-based tourism\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Hausmann et al., \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Tenkanen et al., \u003cspan citationid=\"CR91\" class=\"CitationRef\"\u003e2017\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eComparing the usability of social media data from Instagram, Flickr, and Twitter for visitor monitoring in protected areas, finding tourists\u0026rsquo; preferences to see large mammals.\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e21/0\u003c/p\u003e\u003cp\u003e25/0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eNature-based tourism\u003c/p\u003e\u003cp\u003eNational parks\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Hausmann et al., \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2018\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eUnderstanding Tourists\u0026rsquo;preferences for nature-based experiences in Protected Areas (Flickr and Instagram)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e18/15\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eNature-based tourism\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Gosal et al., \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2019\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eIdentifying different groups of recreation users with social media, machine learning, and natural language processing for better ecosystem services\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e7/24\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eOutdoor recreation\u003c/p\u003e\u003cp\u003eMachine learning\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Teles da Mota \u0026amp; Pickering, \u003cspan citationid=\"CR89\" class=\"CitationRef\"\u003e2020\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eLiterature review on the use of social media data to assess nature-base tourism (all social media)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e5/60\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eNature-based tourism\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Pickering et al., \u003cspan citationid=\"CR77\" class=\"CitationRef\"\u003e2020\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eAssessing seasonal preferences for tourism in a National Park (facilities and activities, landscapes, flora and fauna) comparing image content and text from Flickr\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e6/33\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eNature-based tourism\u003c/p\u003e\u003cp\u003eNational parks\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Runge et al., \u003cspan citationid=\"CR80\" class=\"CitationRef\"\u003e2020\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eAnalyzing localized tourism booms and quantify the spatial expansion of the Arctic tourism footprint\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e4/15\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eNature-based tourism\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Sinclair et al., \u003cspan citationid=\"CR83\" class=\"CitationRef\"\u003e2020\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eEstimating the media visitors' home locations in German national parks (Flickr)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e3/35\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eNature-based tourism\u003c/p\u003e\u003cp\u003eNational parks\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Huai et al., \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2022\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eUsing social media photos and computer vision to assess cultural ecosystem services and landscape features in urban parks\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0/40\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eUrban Parks\u003c/p\u003e\u003cp\u003eComputer vision\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Zhang et al., \u003cspan citationid=\"CR102\" class=\"CitationRef\"\u003e2021\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eTo measure and map spatial patterns in visitation of public lands in Utah (Flickr and Panoramio)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0/38\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eNature-based tourism\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eThis cluster demonstrated that social media can proxy visitor counts in natural parks. Nonetheless, it focused heavily on quantification, rarely examining qualitative experiences of nature or local community impacts. Therefore, future studies should triangulate social media data with participatory methods involving residents and tourists. There is also a predominantly quantitative focus risks reducing nature-based tourism to mere visitation numbers, overlooking issues like visitor impact, conservation challenges, and local community perspectives. Besides that, it can be highlighted that there are geographical biases because protected areas in less digitised regions may be underrepresented, skewing global analyses. Other future research possibilities include cross-validation of Flickr-based visitation proxies with ground-based conservation impact data (e.g., trail erosion, wildlife disturbance).\u003c/p\u003e\u003cp\u003e\u003cb\u003eConclusions: The undeniable contribution of open-API social media to tourism research\u003c/b\u003e\u003c/p\u003e\u003cp\u003eThe first cited work using Flickr geotagged photography to understand tourist dynamics is from 2008 (Girardin, Dal Fiore, et al., \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2008\u003c/span\u003e), and this study shows how 15 years later, at least 7 well-defined research topics or research methods have been developed by researchers from all continents. However, Flickr is not any more a popular platform to share UGC; the amount of user-generated content in Flickr social media has slowly decreased since 2016 (Solazzo et al., \u003cspan citationid=\"CR85\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Other popular platforms closed their APIs for external data requests. Facebook suspended the export of friend data in 2015, bought Instagram in 2016 and suspended the possibility of downloading data such as GPS coordinates for photos and after the CA scandal (Tromble, \u003cspan citationid=\"CR92\" class=\"CitationRef\"\u003e2021\u003c/span\u003e), and severely limited the API access to Facebook Groups and Events in April 2018. Instead of freely researchable datasets, web corporations often impose a commercial construct. Former Twitter (now X) made it possible to search only 7 days back in time, accessing only a sample of all tweets. However, researchers were able to pay for Twitter data access through their Historical API (Perriam et al., \u003cspan citationid=\"CR76\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Researchers have more times publicly addressed Facebook and Twitter to reopen forms of public APIs with proposals that would have disclosed data abuse, but the corporations never even replied (Bruns, \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThe only option for researchers remains web scraping, a much more resource-intensive way to gain data from social media, and one that has no possibilities to control the ethical issues Tromble (\u003cspan citationid=\"CR92\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) proposing to reflect on. The scientific use of Flickr as an open API platform presented in this paper shows that the concerns of Tromble were not in every aspect relevant and large communities of researchers have used that open API social media without ethically questionable results and with great benefits for science. What were these?\u003c/p\u003e\u003cp\u003eTourism research must rely on visitor data, while tour operators, any business involved in the visitor economy and large corporations such as payment card services and mobile phone operators are commercially counteractive in giving out any statistical data to independent researchers. With openly accessible UGC data from all continents on photography \u0026ndash; a large portion of all is travel photography \u0026ndash; researchers could work with methodologies and on topics otherwise inaccessible. Much of Flickr photography is geo-tagged, meaning that the exact geographical location is accessible, together with the time when the photo was taken \u0026ndash; associated with the user who uploaded it. Such data-enabled research on the spatiotemporal analysis of tourism (C1), even in large destination systems and natural parks (C7) \u0026ndash; where other technologies tracing visitors\u0026rsquo; movements does not exist. No other widely accessible data with such even geographic distribution across the globe exists that can exactly show where and when and how many visitors there were. Tracing tourist movements would have remained the privilege of corporations like Facebook, Google or mobile phone operators, which do use such data, but only for their own commercial interest.\u003c/p\u003e\u003cp\u003eThe comparison of different places being photographed, more or less differentiating also by visitor types, allowed the measurement of the attractivity of places (C4), resulting in a deeper understanding of why some attractions perform better than others \u0026ndash; impossible to research without comparatively accessible data on many places of interest generated by users whose attributes are also possible to distinguish. The understanding of which places are of tourists\u0026rsquo; interest depending on different preferences and how they select to visit these in a given timeframe is also a research question in need of large amounts of spatiotemporal data from which different clustering techniques and algorithms (like DBSCAN) can detect which are the geographical areas to be visited (C2). Many researchers could propose travel route recommendation techniques based on such datasets, which also helped to give a relevant quantity of data to be able to provide valid tourist itineraries in line with the Orienteering Problem (C3).\u003c/p\u003e\u003cp\u003eAnother application of Flickr data in tourism research was tied to image content analysis. Researchers could find no other data sources of systematically researchable tourism-related photography where, given the time and place of the photo, the relevant information deducted from their content could be analyzed. Image content analysis proved to be extremely useful in understanding the differences between the projected and perceived destination images of a destination (C5). Researchers could analyze the attraction preferences of different types of tourists, ranging from urban or natural landmarks to different food and beverages. Computer vision was much used for the more contemporary works with destination images. The immense quantity of geolocated photographs was given from Flickr API, therefore enough material was in the disposition of researchers to use machine learning to differentiate the content of photos. No other publicly available dataset was available for the training of computer vision; on the other hand, the qualitative analysis of such large datasets does need machine learning techniques to process all data. As such methodology is useful not only in destination image research but any tourism-related research which needed to analyze the content of Flickr photography, a cluster well dispersed between the other \u0026ndash; much more separated - clusters of the bibliographical network was identified with quite recent work included (C6).\u003c/p\u003e\u003cp\u003eThe community using deep learning methods and the nature-based tourism research community provided the most recent works. The reason that C7 is the cluster with the most recent papers is because visitor statistics of locations inside natural parks and large natural areas or regions are impossible to get in any other ways, if not by expensive individual GPS trackers distributed. Many researchers in C7 directly reflect on the availability of Flickr data as a research assess. Tenkanen et al., (\u003cspan citationid=\"CR91\" class=\"CitationRef\"\u003e2017\u003c/span\u003e) state that data from Flickr is free and easy to download and the temporal patterns in social media data are often similar to results from park entry and trail counters in some natural areas; this statement has been directly quoted by others in C7 (Teles da Mota \u0026amp; Pickering, \u003cspan citationid=\"CR89\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Hausmann et al. (\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2018\u003c/span\u003e) came to a similar conclusion for South African national parks, also mentioning that geodata from social media is a robust indicator for human presence and spatial variation of visitation in protected areas at regional, national and global scale. These observations prove that nature-based tourism research could evolve with the abovementioned relevant results because of the availability of the open API of Flickr, used by researchers worldwide. Flickr was never a flawless database. Some authors mention the existence of deviations and biases related to the popularity of the social media site, the demographic profile of the users, and the different usage of the social media website in terms of recurrence and selection of photos (Barros et al., \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). In fact, there is an overrepresentation of certain demographics (e.g., affluent, tech-savvy tourists) and specific geographies (urban centres and iconic attractions). These limitations point to critical vulnerabilities in research that remain unaddressed in much of the existing literature. Still, there has never been a database so large, precise, and accurate for researching tourism - and with the current trend to use only closed internet ecosystems, there will never be one again.\u003c/p\u003e\u003cp\u003eWithout the open API UGC platform as Flickr, none of the 7 specific topics or methods researched by these large and independent global research communities could have advanced so much in their fields of tourism studies, providing so many highly ranked publications. Research fields of C4-5-6-7 would have developed much smaller without robust datasets to build on their observations, while C1-2-3 would never have even developed as research fields. While previous works have documented the use of social media in tourism (e.g., Shoval \u0026amp; Ahas, \u003cspan citationid=\"CR82\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Giglio et al., \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2019\u003c/span\u003e), they often remained descriptive, focusing merely on cataloguing applications rather than interrogating the structural transformations such datasets enabled in the field. Our analysis reveals that Flickr not only offered new methods for data collection but fundamentally altered the epistemological landscape of tourism studies, fostering collaborative methodologies, enhancing reproducibility, and encouraging global comparative studies. Moreover, while the literature extensively exploited geotagged photography for spatial analysis, conceptual integration between behavioural tourism studies and computational methods remains underdeveloped. And probably, this disconnect limits the theoretical maturation of findings derived from such datasets.\u003c/p\u003e\u003cp\u003eAlthough Scopus offers greater coverage of scientific articles, the main limitation of this work is that it may not include all the documents on Flickr and tourism literature. The network could have been further expanded beyond the direct dataset with the given tags and the works referenced by these, including those referenced by this indirect dataset. The same clusters would have been found in a larger dataset where all relevant research with the expression \u0026ldquo;Flickr\u0026rdquo; in the keywords or abstracts would have been listed without any further limitations. It is an interesting question how much part of all research using Flickr datasets treats tourism. However, it is probable that a golden era of invisible and independent global research communities, giving at least seven new widely published research topics in a decade and a half within tourism, has come to an end.\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eB. K\u0026aacute;d\u0026aacute;r provided the theoretical parts, methodology and results and figures.M. Martins provided the extraction and management of bibliographic data, tables, and validated the results.All authors contributed to the main manuscript text and reviewed it.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eThe datasets generated by the Scopus bibliographic database analyzed during the current study are available in the Mendeley repository: Kadar, Balint; Martins, M\u0026aacute;rcio Ribeiro (2025), \u0026ldquo; bibliometric network analysis of studies using Flickr data 2010-2023\u0026rdquo;, Mendeley Data, V2, doi: 10.17632/fkz7wv3f7y.2\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eAhas R, Aasa A, Mark \u0026Uuml;, Pae T, Kull A (2007) Seasonal tourism spaces in Estonia: Case study with mobile positioning data. Tour Manag 28:898\u0026ndash;910. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.tourman.2006.05.010\u003c/span\u003e\u003cspan address=\"10.1016/j.tourman.2006.05.010\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAshworth GJ, Page SJ (2011) Urban tourism research: Recent progress and current paradoxes. Tour Manag 32(1):1\u0026ndash;15. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.tourman.2010.02.002\u003c/span\u003e\u003cspan address=\"10.1016/j.tourman.2010.02.002\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBarros C, Moya-G\u0026oacute;mez B, Garc\u0026iacute;a-Palomares JC (2019) Identifying temporal patterns of visitors to national parks through geotagged photographs. Sustain (Switzerland) 11(24). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/su11246983\u003c/span\u003e\u003cspan address=\"10.3390/su11246983\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBhatt P, Pickering CM (2022) Destination image of Chitwan National Park, Nepal: Insights from a content analysis of online photographs. J Outdoor Recreation Tourism 37. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jort.2022.100488\u003c/span\u003e\u003cspan address=\"10.1016/j.jort.2022.100488\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBlondel VD, Guillaume J-L, Lambiotte R, Lefebvre E (2008) Fast unfolding of communities in large networks. J Stat Mech: Theory Exp 2008(10):1\u0026ndash;12\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBojic I, Belyi A, Ratti C, Sobolevsky S (2016) Scaling of foreign attractiveness for countries and states. Appl Geogr 73:47\u0026ndash;52. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.apgeog.2016.06.006\u003c/span\u003e\u003cspan address=\"10.1016/j.apgeog.2016.06.006\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBroz M (2022) Flickr Statistics, User Count, \u0026amp; Facts (September 2022). Photutorial.Com. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://photutorial.com/flickr-statistics/\u003c/span\u003e\u003cspan address=\"https://photutorial.com/flickr-statistics/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBruns A (2021) After the \u0026lsquo;APIcalypse\u0026rsquo;: social media platforms and their fight against critical scholarly research. Disinformation and Data Lockdown on Social Platforms, 14\u0026ndash;36. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.4324/9781003206972-2\u003c/span\u003e\u003cspan address=\"10.4324/9781003206972-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eChen W, Xu Z, Zheng X, Luo Y (2019) Geo-tagged photo metadata processing method for Beijing inbound tourism flow. ISPRS Int J Geo-Information 8(12). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/ijgi8120556\u003c/span\u003e\u003cspan address=\"10.3390/ijgi8120556\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCho N, Kang Y, Yoon J, Park S, Kim J (2022) Classifying Tourists\u0026rsquo; Photos and Exploring Tourism Destination Image Using a Deep Learning Model. J Qual Assur Hospitality Tourism 23(6):1480\u0026ndash;1508. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/1528008X.2021.1995567\u003c/span\u003e\u003cspan address=\"10.1080/1528008X.2021.1995567\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCrandall D, Backstrom L, Huttenlocher D, Kleinberg J (2009) Mapping the world\u0026rsquo;s photos. Proceedings of the 18th International Conference on World Wide Web., 761\u0026ndash;770\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCurlin T, Jaković B, Miloloža I (2019) Twitter usage in Tourism: Literature Review. Bus Syst Res 10(1):102\u0026ndash;119. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2478/bsrj-2019-0008\u003c/span\u003e\u003cspan address=\"10.2478/bsrj-2019-0008\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDe Choudhury M, Feldman M, Amer-Yahia S, Golbandi N, Lempel R, Yu C (2010) Automatic Construction of Travel Itineraries using Social Breadcrumbs. Proceedings of the 21st ACM Conference on Hypertext and Hypermedia, 35\u0026ndash;44\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDeng N, Li XR (2018) Feeling a destination through the right photos: A machine learning model for DMOs\u0026rsquo; photo selection. Tour Manag 65:267\u0026ndash;278. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.tourman.2017.09.010\u003c/span\u003e\u003cspan address=\"10.1016/j.tourman.2017.09.010\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDeng N, Liu J, Dai Y, Li H (2019) Different cultures, different photos: A comparison of Shanghai\u0026rsquo;s pictorial destination image between East and West. Tourism Manage Perspect 30:182\u0026ndash;192. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.tmp.2019.02.016\u003c/span\u003e\u003cspan address=\"10.1016/j.tmp.2019.02.016\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDickinson JE, Ghali K, Cherrett T, Speed C, Davies N, Norgate S (2014) Tourism and the smartphone app: capabilities, emerging practice and scope in the travel domain. Curr Issues Tourism 17(1):84\u0026ndash;101. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/13683500.2012.718323\u003c/span\u003e\u003cspan address=\"10.1080/13683500.2012.718323\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDonaire JA, Camprub\u0026iacute; R, Gal\u0026iacute; N (2014) Tourist clusters from Flickr travel photography. Tourism Manage Perspect 11:26\u0026ndash;33. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.tmp.2014.02.003\u003c/span\u003e\u003cspan address=\"10.1016/j.tmp.2014.02.003\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eEck NJ, Van, Waltman L (2016) Text Mining and Visualization. Text Mining and Visualization, 1\u0026ndash;5. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1201/b19007\u003c/span\u003e\u003cspan address=\"10.1201/b19007\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eEster M, Kriegel H-P, Sander J, Xu X (1996) A Density-Based Algorithm for Discovering Clusters in Large Spatial Databases with Noise. www.aaai.org\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGarc\u0026iacute;a-Palomares JC, Guti\u0026eacute;rrez J, M\u0026iacute;nguez C (2015) Identification of tourist hot spots based on social networks: A comparative analysis of European metropolises using photo-sharing services and GIS. Appl Geogr 63:408\u0026ndash;417. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.apgeog.2015.08.002\u003c/span\u003e\u003cspan address=\"10.1016/j.apgeog.2015.08.002\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGarrod B (2009) Understanding the Relationship between Tourism Destination Imagery and Tourist Photography. J Travel Res 47(3):346\u0026ndash;358\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGiglio S, Bertacchini F, Bilotta E, Pantano P (2019) Using social media to identify tourism attractiveness in six Italian cities. Tour Manag 72:306\u0026ndash;312. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.tourman.2018.12.007\u003c/span\u003e\u003cspan address=\"10.1016/j.tourman.2018.12.007\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGirardin F, Calabrese F, Fiore FD, Biderman A, Ratti C, Blat J (2008) Uncovering the presence and movements of tourists from user- generated content\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGirardin F, Dal Fiore F, Blat J, Ratti C (2007) Understanding of tourist dynamics from explicitly disclosed location information. \u003cem\u003e4th International Symposium on LBS and Telecartography\u003c/em\u003e, \u003cem\u003e58\u003c/em\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGirardin F, Dal Fiore F, Ratti C, Blat J (2008) Leveraging explicitly disclosed location information to understand tourist dynamics: A case study. J Location Based Serv 2(1):41\u0026ndash;56. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/17489720802261138\u003c/span\u003e\u003cspan address=\"10.1080/17489720802261138\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGirardin F, Fiore FD, Blat J, Ratti C, Fiore D, F (2008) Understanding of Tourist Dynamics from Explicitly Disclosed Location Information. J Location Based Serv, 2(1)\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGonz\u0026aacute;lez MC, Hidalgo CA, Barab\u0026aacute;si AL (2008) Understanding individual human mobility patterns. Nat 2008 453(7196):779\u0026ndash;782. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/nature06958\u003c/span\u003e\u003cspan address=\"10.1038/nature06958\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. \u003cem\u003e453\u003c/em\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGosal AS, Geijzendorffer IR, V\u0026aacute;clav\u0026iacute;k T, Poulin B, Ziv G (2019) Using social media, machine learning and natural language processing to map multiple recreational beneficiaries. Ecosyst Serv 38. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.ecoser.2019.100958\u003c/span\u003e\u003cspan address=\"10.1016/j.ecoser.2019.100958\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHausmann A, Toivonen T, Heikinheimo V, Tenkanen H, Slotow R, Minin EDI (2017) Social media reveal that charismatic species are not the main attractor of ecotourists to sub-Saharan protected areas. Sci Rep 2017 7(1):1\u0026ndash;9. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41598-017-00858-6\u003c/span\u003e\u003cspan address=\"10.1038/s41598-017-00858-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHausmann A, Toivonen T, Slotow R, Tenkanen H, Moilanen A, Heikinheimo V, Di Minin E (2018) Social Media Data Can Be Used to Understand Tourists\u0026rsquo; Preferences for Nature-Based Experiences in Protected Areas. Conserv Lett 11(1):e12343. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/CONL.12343\u003c/span\u003e\u003cspan address=\"10.1111/CONL.12343\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHawelka B, Sitko I, Beinat E, Sobolevsky S, Kazakopoulos P, Ratti C (2014) Geo-located Twitter as proxy for global mobility patterns. Cartography Geographic Inform Sci 41(3):260\u0026ndash;271. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/15230406.2014.890072\u003c/span\u003e\u003cspan address=\"10.1080/15230406.2014.890072\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHerrera-Franco G, Montalv\u0026aacute;n-Burbano N, Carri\u0026oacute;n-Mero P, Apolo-Masache B, Jaya-Montalvo M (2020) Research Trends in Geotourism: A Bibliometric Analysis Using the Scopus Database. Geosci 2020 10(10):379. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/GEOSCIENCES10100379\u003c/span\u003e\u003cspan address=\"10.3390/GEOSCIENCES10100379\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. \u003cem\u003e10\u003c/em\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHollenstein L, Purves RS (2010) Exploring place through user-generated content: Using Flickr tags to describe city cores. J Spat Inform Sci 1(2010):21\u0026ndash;48. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.5311/JOSIS.2010.1.3\u003c/span\u003e\u003cspan address=\"10.5311/JOSIS.2010.1.3\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eH\u0026ouml;pken W, M\u0026uuml;ller M, Fuchs M, Lexhagen M (2020) Flickr data for analysing tourists\u0026rsquo; spatial behaviour and movement patterns: A comparison of clustering techniques. J Hospitality Tourism Technol 11(1):69\u0026ndash;82. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1108/JHTT-08-2017-0059\u003c/span\u003e\u003cspan address=\"10.1108/JHTT-08-2017-0059\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHruška J, P\u0026aacute;skov\u0026aacute; M (2018) How do the national tourism organizations use the social media? [Jak n\u0026aacute;rodn\u0026iacute; centr\u0026aacute;ly cestovn\u0026iacute;ho ruchu využ\u0026iacute;vaj\u0026iacute; soci\u0026aacute;ln\u0026iacute; m\u0026eacute;dia?]. E M: Ekonomie Manage 21(4):226\u0026ndash;240. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.15240/tul/001/2018-4-015\u003c/span\u003e\u003cspan address=\"10.15240/tul/001/2018-4-015\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHu Y, Gao S, Janowicz K, Yu B, Li W, Prasad S (2015) Extracting and understanding urban areas of interest using geotagged photos. Comput Environ Urban Syst 54:240\u0026ndash;254. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.compenvurbsys.2015.09.001\u003c/span\u003e\u003cspan address=\"10.1016/j.compenvurbsys.2015.09.001\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHuai S, Chen F, Liu S, Canters F, Van de Voorde T (2022) Using social media photos and computer vision to assess cultural ecosystem services and landscape features in urban parks. Ecosyst Serv 57. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.ecoser.2022.101475\u003c/span\u003e\u003cspan address=\"10.1016/j.ecoser.2022.101475\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHwang YH, Fesenmaier DR (2003) Multidestination pleasure travel patterns: Empirical evidence from the American Travel Survey. J Travel Res 42(2):166\u0026ndash;171. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1177/0047287503253936\u003c/span\u003e\u003cspan address=\"10.1177/0047287503253936\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJacomy M, Heymann S, Venturini T, Bastian M (2011) Force Atlas 2, a graph layout algorithm for handy network visualization. \u0026hellip; Http://Www.Medialab. Sciences-Po. Fr/ \u0026hellip;, 1\u0026ndash;21. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://scholar.google.com/scholar?hl=en\u003c/span\u003e\u003cspan address=\"http://scholar.google.com/scholar?hl=en\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u0026amp;btnG=Search\u0026amp;q=intitle:ForceAtlas2+,+A+Graph+Layout+Algorithm+for+Handy+Network+Visualization#0\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJain A, Singh SN (2017) Identifying concavity laws of attractiveness character for composite regions: Application through geo-tagged media check-ins. Proceeding - IEEE International Conference on Computing, Communication and Automation, ICCCA 2017, 2017-January, 1041\u0026ndash;1045. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1109/CCAA.2017.8229994\u003c/span\u003e\u003cspan address=\"10.1109/CCAA.2017.8229994\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJenkins OH (2003) Photography and travel brochures: the circle of representation. Tourism Geographies 5(3):305\u0026ndash;328. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/1461668032000098024\u003c/span\u003e\u003cspan address=\"10.1080/1461668032000098024\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eK\u0026aacute;d\u0026aacute;r B (2014) Measuring tourist activities in cities using geotagged photography. Tourism Geographies 16(1):88\u0026ndash;104. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/14616688.2013.868029\u003c/span\u003e\u003cspan address=\"10.1080/14616688.2013.868029\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eK\u0026aacute;d\u0026aacute;r B, Gede M (2013) Where do tourists go? Visualizing and analysing the spatial distribution of geotagged photography. Cartographica 48(2):78\u0026ndash;88. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3138/carto.48.2.1839\u003c/span\u003e\u003cspan address=\"10.3138/carto.48.2.1839\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eK\u0026aacute;d\u0026aacute;r B, Gede M (2021) Tourism flows in large-scale destination systems. Annals Tourism Res 87. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.annals.2020.103113\u003c/span\u003e\u003cspan address=\"10.1016/j.annals.2020.103113\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKang Y, Cho N, Yoon J, Park S, Kim J (2021) Transfer learning of a deep learning model for exploring tourists\u0026rsquo; urban image using geotagged photos. ISPRS Int J Geo-Information 10(3). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/ijgi10030137\u003c/span\u003e\u003cspan address=\"10.3390/ijgi10030137\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKarayazi SS, Dane G, de Vries B (2021) Utilizing urban geospatial data to understand heritage attractiveness in amsterdam. ISPRS Int J Geo-Information 10(4). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/ijgi10040198\u003c/span\u003e\u003cspan address=\"10.3390/ijgi10040198\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKennedy L, Naaman M, Ahern S, Nair R, Rattenbury T (2007) How Flickr Helps us Make Sense of the World: Context and Content in Community-Contributed Media Collections. Proceedings of the 15th ACM International Conference on Multimedia. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://flickr.com\u003c/span\u003e\u003cspan address=\"http://flickr.com\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKim D, Kang Y, Park Y, Kim N, Lee J (2020) Understanding tourists\u0026rsquo; urban images with geotagged photos using convolutional neural networks. Spat Inform Res 28(2):241\u0026ndash;255. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s41324-019-00285-x\u003c/span\u003e\u003cspan address=\"10.1007/s41324-019-00285-x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKisilevich S, Keim D, Andrienko N, Andrienko G (2013) Towards acquisition of semantics of places and events by multi-perspective analysis of geotagged photo collections. In: Moore A, Drecki I (eds) Geospatial Visualisation. Lecture Notes in Geoinformation and Cartography. Springer. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://www.flickr.com\u003c/span\u003e\u003cspan address=\"http://www.flickr.com\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKisilevich S, Mansmann F, Keim D (2010) P-DBSCAN: A density based clustering algorithm forexploration and analysis of attractive areas usingcollections of geo-tagged photos. COM. Geo 2010: 1st International Conference and Exhibition on Computing for Geospatial Research \u0026amp; Application, 274\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKorakakis M, Spyrou E, Mylonas P, Perantonis SJ (2017) Exploiting social media information toward a context-aware recommendation system. Social Netw Anal Min 7(1). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s13278-017-0459-9\u003c/span\u003e\u003cspan address=\"10.1007/s13278-017-0459-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKoutras A, Nikas IA, Panagopoulos A (2019) Towards Developing Smart Cities: Evidence from GIS Analysis on Tourists\u0026rsquo; Behavior Using Social Network Data in the City of Athens. In: Katsoni V, Segarra-O\u0026ntilde;a M (eds) Smart Tourism as a Driver for Culture and Sustainability. Springer International Publishing, pp 407\u0026ndash;418\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKrizhevsky A, Sutskever I, Hinton GE (2012) ImageNet Classification with Deep Convolutional Neural Networks. Advances in Neural Information Processing Systems, 25. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://code.google.com/p/cuda-convnet/\u003c/span\u003e\u003cspan address=\"http://code.google.com/p/cuda-convnet/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKurashima T, Iwata T, Irie G, Fujimura K (2010) Travel Route Recommendation Using Geotags in Photo Sharing Sites. \u003cem\u003eProceedings of the 19th ACM International Conference on Information and Knowledge Management\u003c/em\u003e, 579\u0026ndash;588\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKurashima T, Iwata T, Irie G, Fujimura K (2013) Travel route recommendation using geotagged photos. Knowl Inf Syst 37(1):37\u0026ndash;60. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s10115-012-0580-z\u003c/span\u003e\u003cspan address=\"10.1007/s10115-012-0580-z\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLee S, Zhang H, Crandall DJ (2015) Predicting geo-informative attributes in large-scale image collections using convolutional neural networks. Proceedings \u0026ndash;\u0026thinsp;2015 IEEE Winter Conference on Applications of Computer Vision, WACV 2015, 550\u0026ndash;557. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1109/WACV.2015.79\u003c/span\u003e\u003cspan address=\"10.1109/WACV.2015.79\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLi D, Zhou X, Wang M (2018) Analyzing and visualizing the spatial interactions between tourists and locals: A Flickr study in ten US cities. Cities 74:249\u0026ndash;258. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.cities.2017.12.012\u003c/span\u003e\u003cspan address=\"10.1016/j.cities.2017.12.012\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLim KH (2016) Recommending and planning trip itineraries for individual travellers and groups of tourists. Proceedings International Conference on Automated Planning and Scheduling, ICAPS, 2016-Janua(April)\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLim KH, Chan J, Leckie C, Karunasekera S (2018) Personalized trip recommendation for tourists based on user interests, points of interest visit durations and visit recency. Knowl Inf Syst 54(2):375\u0026ndash;406. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s10115-017-1056-y\u003c/span\u003e\u003cspan address=\"10.1007/s10115-017-1056-y\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLin MS, Liang Y, Xue JX, Pan B, Schroeder A (2021) Destination image through social media analytics and survey method. Int J Contemp Hospitality Manage 33(6):2219\u0026ndash;2238. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1108/IJCHM-08-2020-0861\u003c/span\u003e\u003cspan address=\"10.1108/IJCHM-08-2020-0861\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLowe DG (2004) Distinctive image features from scale-invariant keypoints. Int J Comput Vision 60(2):91\u0026ndash;110. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1023/B:VISI.0000029664.99615.94/METRICS\u003c/span\u003e\u003cspan address=\"10.1023/B:VISI.0000029664.99615.94/METRICS\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMajid A, Chen L, Chen G, Mirza HT, Hussain I, Woodward J (2013) A context-aware personalized travel recommendation system based on geotagged social media data mining. Int J Geogr Inf Sci 27(4):662\u0026ndash;684. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/13658816.2012.696649\u003c/span\u003e\u003cspan address=\"10.1080/13658816.2012.696649\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMangachena JR, Chauvenet ALM, Pickering CM (2022) Destination Images of Africa Reflected in Social Media Photographs. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/08941920.2022.2100023\u003c/span\u003e\u003cspan address=\"10.1080/08941920.2022.2100023\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Society and Natural Resources\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMartins M, Costa R (2022) Tracking Technologies in Tourism: A Bibliometric and Content Review. In: de Carvalho JV, Liberato P, Pe\u0026ntilde;a A (eds) Advances in Tourism, Technology and Systems. Springer Nature Singapore, pp 215\u0026ndash;230\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMckercher B, Shoval N, Ng E, Birenboim A (2012) First and Repeat Visitor Behaviour: GPS Tracking and GIS Analysis in Hong Kong. Tourism Geographies 14(1):147\u0026ndash;161. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/14616688.2011.598542\u003c/span\u003e\u003cspan address=\"10.1080/14616688.2011.598542\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMemon I, Chen L, Majid A, Lv M, Hussain I, Chen G (2015) Travel Recommendation Using Geo-tagged Photos in Social Media for Tourist. Wireless Pers Commun 80(4):1347\u0026ndash;1362. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s11277-014-2082-7\u003c/span\u003e\u003cspan address=\"10.1007/s11277-014-2082-7\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMiah SJ, Vu HQ, Gammack J, McGrath M (2017) Inform Manage 54(6):771\u0026ndash;785. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.im.2016.11.011\u003c/span\u003e\u003cspan address=\"10.1016/j.im.2016.11.011\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. A Big Data Analytics Method for Tourist Behaviour Analysis\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMor M, Dalyot S (2020) Enriching walking routes with tourism attractions retrieved from crowdsourced user generated data. ISPRS Annals of the Photogrammetry. Remote Sens Spat Inform Sci 5(4):95\u0026ndash;102. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.5194/isprs-Annals-V-4-2020-95-2020\u003c/span\u003e\u003cspan address=\"10.5194/isprs-Annals-V-4-2020-95-2020\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMor M, Fisher-Gewirtzman D, Yosifof R, Dalyot S (2021) 3D visibility analysis for evaluating the attractiveness of tourism routes computed from social media photos. ISPRS Int J Geo-Information 10(5). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/ijgi10050275\u003c/span\u003e\u003cspan address=\"10.3390/ijgi10050275\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMou N, Jiang Q, Zhang L, Niu J, Zheng Y, Wang Y, Yang T (2022) Personalized tourist route recommendation model with a trajectory understanding via neural networks. Int J Digit Earth 15(1):1738\u0026ndash;1759. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/17538947.2022.2130456\u003c/span\u003e\u003cspan address=\"10.1080/17538947.2022.2130456\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNikas IA, Koutras A, Panagopoulos A, Vasileiadis A (2022) Exploring Tourists\u0026rsquo; Food and Beverage Spots in an Urban Destination Using a Spatial\u0026ndash;temporal Approach. In: Katsoni V, \\cSerban AC (eds) Transcending Borders in Tourism Through Innovation and Cultural Heritage. Springer International Publishing, pp 977\u0026ndash;989\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNovack T, Vorbeck L, Lorei H, Zipf A (2020) Towards detecting building facades with graffiti artwork based on street view images. ISPRS Int J Geo-Information 9(2). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/ijgi9020098\u003c/span\u003e\u003cspan address=\"10.3390/ijgi9020098\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003e\u0026Ouml;nder I, Koerbitz W, Hubmann-Haidvogel A (2016) Tracing Tourists by Their Digital Footprints: The Case of Austria. J Travel Res 55(5):566\u0026ndash;573. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1177/0047287514563985\u003c/span\u003e\u003cspan address=\"10.1177/0047287514563985\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePaldino S, Bojic I, Sobolevsky S, Ratti C, Gonz\u0026aacute;lez MC (2015) Urban magnetism through the lens of geo-tagged photography. EPJ Data Sci 4(1):1\u0026ndash;17. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1140/epjds/s13688-015-0043-3\u003c/span\u003e\u003cspan address=\"10.1140/epjds/s13688-015-0043-3\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePan S, Lee J, Tsai H (2014) Travel photos: Motivations, image dimensions, and affective qualities of places. Tour Manag 40:59\u0026ndash;69. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.tourman.2013.05.007\u003c/span\u003e\u003cspan address=\"10.1016/j.tourman.2013.05.007\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePerriam J, Birkbak A, Freeman A (2020) Digital methods in a post-API environment. Int J Soc Res Methodol 23(3):277\u0026ndash;290. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/13645579.2019.1682840\u003c/span\u003e\u003cspan address=\"10.1080/13645579.2019.1682840\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePickering C, Walden-Schreiner C, Barros A, Rossi SD (2020) Using social media images and text to examine how tourists view and value the highest mountain in Australia. J Outdoor Recreation Tourism 29. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jort.2019.100252\u003c/span\u003e\u003cspan address=\"10.1016/j.jort.2019.100252\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePopescu A, Grefenstette G, Mo\u0026euml;llic PA (2009) Mining tourist information from user-supplied collections. International Conference on Information and Knowledge Management, Proceedings, 1713\u0026ndash;1716. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1145/1645953.1646211\u003c/span\u003e\u003cspan address=\"10.1145/1645953.1646211\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRattenbury T (2009) (TWEB), M. N.-A. T. on the W., \u0026amp; undefined. (2009). Methods for extracting place semantics from Flickr tags. Dl.Acm.OrgT Rattenbury, M NaamanACM Transactions on the Web (TWEB), 2009\u0026bull;dl.Acm.Org, 3(1). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1145/1462148.1462149\u003c/span\u003e\u003cspan address=\"10.1145/1462148.1462149\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRunge CA, Daigle RM, Hausner VH (2020) Quantifying tourism booms and the increasing footprint in the Arctic with social media data. PLoS ONE 15(1). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1371/journal.pone.0227189\u003c/span\u003e\u003cspan address=\"10.1371/journal.pone.0227189\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSarkar JL, Majumder A, Panigrahi CR, Roy S (2020) MULTITOUR: A multiple itinerary tourists recommendation engine. Electron Commer Res Appl 40. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.elerap.2020.100943\u003c/span\u003e\u003cspan address=\"10.1016/j.elerap.2020.100943\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eShoval N, Ahas R (2016) The use of tracking technologies in tourism research: the first decade. Tourism Geographies 18(5):587\u0026ndash;606. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/14616688.2016.1214977\u003c/span\u003e\u003cspan address=\"10.1080/14616688.2016.1214977\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSinclair M, Mayer M, Woltering M, Ghermandi A (2020) Using social media to estimate visitor provenance and patterns of recreation in Germany\u0026rsquo;s national parks. J Environ Manage 263. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jenvman.2020.110418\u003c/span\u003e\u003cspan address=\"10.1016/j.jenvman.2020.110418\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSobolevsky S, Sitko I, Des Combes RT, Hawelka B, Arias JM, Ratti C (2014) Money on the move: Big data of bank card transactions as the new proxy for human mobility patterns and regional delineation. The case of residents and foreign visitors in Spain. Proceedings \u0026ndash;\u0026thinsp;2014 IEEE International Congress on Big Data, BigData Congress 2014, 136\u0026ndash;143. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1109/BigData.Congress.2014.28\u003c/span\u003e\u003cspan address=\"10.1109/BigData.Congress.2014.28\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSolazzo G, Maruccia Y, Lorenzo G, Ndou V, Del Vecchio P, Elia G (2022) Extracting insights from big social data for smarter tourism destination management. Measuring Bus Excellence 26(1):122\u0026ndash;140. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1108/MBE-11-2020-0156\u003c/span\u003e\u003cspan address=\"10.1108/MBE-11-2020-0156\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eStepchenkova S, Zhan F (2013) Visual destination images of Peru: Comparative content analysis of DMO and user-generated photography. Tour Manag 36:590\u0026ndash;601. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.tourman.2012.08.006\u003c/span\u003e\u003cspan address=\"10.1016/j.tourman.2012.08.006\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eStraumann RK, \u0026Ccedil;\u0026ouml;ltekin A, Andrienko G (2014) Towards (re)constructing narratives from georeferenced photographs through visual analytics. Cartographic J 51(2):152\u0026ndash;165. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1179/1743277414Y.0000000079\u003c/span\u003e\u003cspan address=\"10.1179/1743277414Y.0000000079\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSun Y, Fan H, Bakillah M, Zipf A (2015) Road-based travel recommendation using geo-tagged images. Comput Environ Urban Syst 53:110\u0026ndash;122. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.compenvurbsys.2013.07.006\u003c/span\u003e\u003cspan address=\"10.1016/j.compenvurbsys.2013.07.006\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTeles da Mota V, Pickering C (2020) Using social media to assess nature-based tourism: Current research and future trends. \u003cem\u003eJournal of Outdoor Recreation and Tourism\u003c/em\u003e, \u003cem\u003e30\u003c/em\u003e. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jort.2020.100295\u003c/span\u003e\u003cspan address=\"10.1016/j.jort.2020.100295\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTeles da Mota V, Pickering C, Chauvenet A (2022) Popularity of Australian beaches: Insights from social media images for coastal management. Ocean Coast Manag 217. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.ocecoaman.2021.106018\u003c/span\u003e\u003cspan address=\"10.1016/j.ocecoaman.2021.106018\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTenkanen H, Di Minin E, Heikinheimo V, Hausmann A, Herbst M, Kajala L, Toivonen T (2017) Instagram, Flickr, or Twitter: Assessing the usability of social media data for visitor monitoring in protected areas. Sci Rep 7(1):1\u0026ndash;11. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41598-017-18007-4\u003c/span\u003e\u003cspan address=\"10.1038/s41598-017-18007-4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTromble R (2021) Social Media Soc 7(1). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1177/2056305121988929\u003c/span\u003e\u003cspan address=\"10.1177/2056305121988929\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Where Have All the Data Gone? A Critical Reflection on Academic Digital Research in the Post-API Age\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eUrry J (1990) The Tourist gaze - Leisure and Travel in Contemporary Societies. SAGE publications Ltd.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eValls F, Roca J (2021) Visualizing digital traces for sustainable urban management: Mapping tourism activity on the virtual public space. Sustain (Switzerland) 13(6). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/su13063159\u003c/span\u003e\u003cspan address=\"10.3390/su13063159\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003evan Eck NJ, Waltman L (2017) Citation-based clustering of publications using CitNetExplorer and VOSviewer. Scientometrics 111(2):1053. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/S11192-017-2300-7\u003c/span\u003e\u003cspan address=\"10.1007/S11192-017-2300-7\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eVu HQ, Li G, Law R, Ye BH (2015) Exploring the travel behaviors of inbound tourists to Hong Kong using geotagged photos. Tour Manag 46:222\u0026ndash;232. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.tourman.2014.07.003\u003c/span\u003e\u003cspan address=\"10.1016/j.tourman.2014.07.003\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eVu HQ, Li G, Law R, Zhang Y (2018) Travel Diaries Analysis by Sequential Rule Mining. J Travel Res 57(3):399\u0026ndash;413. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1177/0047287517692446\u003c/span\u003e\u003cspan address=\"10.1177/0047287517692446\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWood SA, Guerry AD, Silver JM, Lacayo M (2013) Using social media to quantify nature-based tourism and recreation. Sci Rep 3. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/srep02976\u003c/span\u003e\u003cspan address=\"10.1038/srep02976\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYang L, Wu L, Liu Y, Kang C (2017) Quantifying tourist behavior patterns by travel motifs and geo-tagged photos from flickr. ISPRS Int J Geo-Information 6(11). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/ijgi6110345\u003c/span\u003e\u003cspan address=\"10.3390/ijgi6110345\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYeran S, Fan H (2014) Event Identification from Georeferenced Images. In J. Huerta, S. Schade, \u0026amp; C. Granell (Eds.), Connecting a digital Europe through location and place. Lecture Notes in Geoinformation and Cartography. (pp. 73\u0026ndash;78). Springer International Publishing. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/https://doi.org/10.1007/978-3-319-03611-3_5\u003c/span\u003e\u003cspan address=\"10.1007/978-3-319-03611-3_5\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYuan Y, Medel M (2016) Characterizing international travel behavior from geotagged photos: A case study of Flickr. PLoS ONE 11(5). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1371/journal.pone.0154885\u003c/span\u003e\u003cspan address=\"10.1371/journal.pone.0154885\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZhang H, van Berkel D, Howe PD, Miller ZD, Smith JW (2021) Using social media to measure and map visitation to public lands in Utah. Appl Geogr 128. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.apgeog.2021.102389\u003c/span\u003e\u003cspan address=\"10.1016/j.apgeog.2021.102389\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZhang K, Chen Y, Li C (2019) Discovering the tourists\u0026rsquo; behaviors and perceptions in a tourism destination by analyzing photos\u0026rsquo; visual content with a computer deep learning model: The case of Beijing. Tour Manag 75:595\u0026ndash;608. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.tourman.2019.07.002\u003c/span\u003e\u003cspan address=\"10.1016/j.tourman.2019.07.002\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZhang K, Chen Y, Lin Z (2020) Mapping destination images and behavioral patterns from user-generated photos: a computer vision approach. Asia Pac J Tourism Res 25(11):1199\u0026ndash;1214. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/10941665.2020.1838586\u003c/span\u003e\u003cspan address=\"10.1080/10941665.2020.1838586\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZheng Y-T, Zha Z-J, Chua T-S (2012) Mining travel patterns from geotagged photos. ACM Trans Intell Syst Technol 3(3). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1145/2168752.2168770\u003c/span\u003e\u003cspan address=\"10.1145/2168752.2168770\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZhou X, Xu C, Kimmons B (2015) Detecting tourism destinations using scalable geospatial analysis based on cloud computing platform. Comput Environ Urban Syst 54:144\u0026ndash;153. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.compenvurbsys.2015.07.006\u003c/span\u003e\u003cspan address=\"10.1016/j.compenvurbsys.2015.07.006\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZhu Y, Deng X, Newsam S (2019) Fine-Grained Land Use Classification at the City Scale Using Ground-Level Images. IEEE Trans Multimedia 21(7):1825\u0026ndash;1838. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1109/TMM.2019.2891999\u003c/span\u003e\u003cspan address=\"10.1109/TMM.2019.2891999\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"information-technology-and-tourism","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"jitt","sideBox":"Learn more about [Information Technology \u0026 Tourism](https://link.springer.com/journal/40558)","snPcode":"40558","submissionUrl":"https://submission.springernature.com/new-submission/40558/3","title":"Information Technology \u0026 Tourism","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"tourism studies, bibliometric analysis, Flickr, open API, network analysis","lastPublishedDoi":"10.21203/rs.3.rs-7072228/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7072228/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"This paper explores how Flickr’s geotagged photos have contributed to the development of new research topics in tourism studies, particularly through the use of large-scale, freely available datasets. A systematic literature review and bibliometric network analysis were conducted using 333 Scopus-indexed papers that included “tourism” and “Flickr” in abstracts or keywords. An additional 519 papers citing this core set were identified. Network analysis using Gephi applied a gravity model and clustering algorithms to detect citation-based communities. Content analysis of highly cited papers helped define key research themes. Seven research clusters emerged, focusing on nature-based tourism, tourist activities by space-time behaviour, destination attractiveness, image development, travel route detection and recommendation, and machine learning for content analysis. These communities reflect the global academic interest enabled by Flickr’s open API, allowing reproducible and comparative analyses across destinations. This study highlights how a unified, open-access social media dataset has catalyzed the formation of global research communities in tourism. Unlike newer platforms with restricted access, Flickr’s openness fostered methodological innovation and deepened field-specific knowledge in tourism research. A critical analysis of existing work was provided, highlighting overlooked areas and synthesising insights to propose new directions for research.","manuscriptTitle":"The contribution of accessible social media data to the work of tourism research communities: a bibliometric network analysis of studies using Flickr data","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-08-27 06:58:38","doi":"10.21203/rs.3.rs-7072228/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-10-06T17:13:03+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-10-02T20:58:30+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-09-30T17:11:57+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"35964558590660306535847854690351767895","date":"2025-09-24T17:22:05+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"255854398933252275158637454937214295412","date":"2025-09-22T14:21:29+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"11488424921035073152321667067552753370","date":"2025-08-25T17:20:37+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"288265262696148470755789036424139268829","date":"2025-08-18T14:45:04+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-08-18T14:14:10+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-08-18T09:18:51+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-07-09T10:24:55+00:00","index":"","fulltext":""},{"type":"submitted","content":"Information Technology \u0026 Tourism","date":"2025-07-08T07:57:59+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"information-technology-and-tourism","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"jitt","sideBox":"Learn more about [Information Technology \u0026 Tourism](https://link.springer.com/journal/40558)","snPcode":"40558","submissionUrl":"https://submission.springernature.com/new-submission/40558/3","title":"Information Technology \u0026 Tourism","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"e8db2a5e-97a7-4782-bba3-eca29e5cb86a","owner":[],"postedDate":"August 27th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-05-12T11:17:37+00:00","versionOfRecord":[],"versionCreatedAt":"2025-08-27 06:58:38","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7072228","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7072228","identity":"rs-7072228","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}