Analyzing Tourists’ Scene Preferences through Digital Footprints in World Heritage Site: A Case Study of the Grand Canal’s Tongzhou Section

Analyzing Tourists’ Scene Preferences through Digital Footprints in World Heritage Site: A Case Study of the Grand Canal’s Tongzhou Section | 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 Article Analyzing Tourists’ Scene Preferences through Digital Footprints in World Heritage Site: A Case Study of the Grand Canal’s Tongzhou Section Xuanrui Gu, Zhongjun Wang, Jinxuan Qin This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6619412/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract World Cultural Heritage sites represent the historical value and spatial significance of human civilization, balancing cultural preservation with ecological protection. They serve not only as vessels of cultural continuity but also as key resources for tourism development. To enhance the understanding of tourist scene preferences and provide targeted strategies for future planning, this study focuses on the Tongzhou section of the Grand Canal—designated as a UNESCO World Cultural Heritage site—as a case study. By collecting content from mainstream short video platforms (primarily Douyin) and developing custom analysis tools using Python and Baidu AI Cloud APIs, the study extracts and interprets keyframes and associated textual data from videos to identify the types of scenes that capture visitors' attention, as well as variations across different heritage nodes. Results show that natural experiential scenes receive the highest level of visitor interest, while historical and cultural scenes—despite the area's rich heritage—remain underappreciated. Moreover, tourist scene preferences vary significantly between nodes, influenced by the specific characteristics of each site. Based on these findings, the study offers practical recommendations for site managers to better align interpretive and experiential offerings with visitor interests, while also addressing the needs of local residents. Business and commerce/Information systems and information technology Social science/Cultural and media studies Digital Footprints the Grand Canal World Cultural Heritage Site Tourism Short Videos Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 1 Introduction In June 2014, the Grand Canal of China was successfully inscribed on the World Heritage List. In July 2019, the "Construction Plan for the Great Wall, Grand Canal, and Long March National Cultural Parks" was approved, marking the official commencement of the construction of national cultural parks [ 12 ] . With the continuous development of China's economy and society, consumer upgrades and the evolution of demand have become important issues in the new era of social consumption. In the cultural tourism industry, traditional sightseeing tourism has begun to transform to meet the increasingly diversified needs of tourists. In the era of integrated development of culture and tourism, protecting and showcasing large-scale linear cultural heritage and cultural routes has become an increasingly hot topic for research. Scholars discuss the Grand Canal as linear cultural heritage, drawing comparisons with European cultural route theories and American heritage corridor theories, while also debating whether these places necessarily exhibit the form of linear cultural heritage [ 19 , 28 ] . Currently, academic research on linear cultural heritage and cultural routes including the Grand Canal mainly focuses on summarizing foreign experiences [ 18 , 32 ] , exploring the origins of the concept [ 7 , 20 , 23 ] , studying their functions and values [ 4 , 13 ] , as well as empirical research on specific park planning and construction [ 9 , 14 , 20 , 25 ] . However, there has been relatively little research on visitor behavior characteristics and preference within World Cultural Heritage sites, making it essential to study the scene preference of visitor in these heritages. Research on visitor’s scene preference in World Cultural Heritage sites based on digital footprints not only provides reference and support for the management and planning of heritage sites but also deepens understanding of visitor needs and behaviors, thereby facilitating the sustainable development of heritage. In the field of digital footprint research, Negroponte, N. proposed the concept of the "Slug trail" in the 1990s, which is generally considered as a precursor to the "digital footprint" [ 15 ] . In 2007, Stephen, D. first introduced the concept of the "digital footprint," defining it as the data traces left by individuals interacting with digital environments or media [ 23 ] . Since then, numerous scholars have studied this concept and proposed different definitions [ 5 , 6 , 16 , 21 , 22 ] . Regarding the application of digital footprints in tourism, current research both domestically and internationally often relies on data obtained from online texts such as travelogues and microblogs, as well as online photos [ 1 , 2 , 17 , 18 , 24 , 30 ] , while research on obtaining User-Generated Content (UGC) videos from emerging short video media platforms is scarce [ 3 ] . Therefore, this study incorporates short videos as a data source into digital footprint research to address the lack of academic research in this specific field. The theoretical framework of this study encompasses theories of tourist preferences and tourist experiential scenes. Tourist preference refers to the cognitive and affective components of tourists' inclinations towards a specific tourism product or destination, manifested by their emotional and intentional factors. It not only involves explicit preferences displayed during the travel process but also encompasses abstract and emotional tourism preferences [ 25 ] . Tourist preferences can be categorized into two aspects: explicit and stated preferences [ 29 ] . This study focuses on identifying tourists' preferences for attractions based on their geographic location information, falling under the category of explicit preferences research. Tourist experience scenes refer to the comprehensive perception of all tangible environments and intangible services that tourists encounter throughout their travel behaviors, a concept derived from Bitner's service scene [ 10 ] . Building upon this foundation, Zhang Hui et al. proposed the concept of tourist experience scenes, which encompasses four dimensions: physical, social, socio-symbolic, and natural dimensions [ 31 ] . Meanwhile, research on tourist scenes and visitor behavior characteristics has developed relatively mature theoretical frameworks, with rich research content and methods. However, these aspects of research lack the incorporation of emerging short video media as a data source. Therefore, this study will conduct analysis based on short video media. 2 Research Design and Process The study aims to acquire visitor digital footprints of the Tongzhou section of the Grand Canal World Cultural Heritage through short video platforms, analyze visitors' preferences for Tourist Scenes, and attempt to construct the mechanism and logical origins of visitor scene preferences in heritage sights. This analysis seeks to understand the strengths and weaknesses of tourism space development in the Tongzhou section of the Grand Canal, with the goal of providing targeted opinions and strategies for the planning, enhancement, construction, and management of the heritage. 2.1 Overview of Research Destination The Grand Canal in China is a dual heritage of nature and culture, boasting distinct natural and cultural ecosystems [ 7 ] . China’s Grand Canal is both a natural and cultural World Heritage site, characterized by its distinct ecological and cultural systems. It functions not only as an ecological corridor linking northern and southern China but also as a cultural corridor that connects and integrates six major cultural regions: Beijing-Tianjin, Yan-Zhao, Qi-Lu, Central Plains, Huaiyang, and Wu-Yue. The canal embodies cultural diversity, reflects the essence of Chinese civilization, and carries profound national memory and cultural identity. Therefore, the cultural heritage of the Grand Canal is not intended to serve merely as a conventional cultural landmark. Rather, it is built upon the canal’s iconic tangible and intangible cultural heritage and aims to establish a symbolic representation of Chinese national culture in the new era. Guided by the strategic goal of promoting national spirit, the canal serves as a platform for public cultural expression. Through the effective preservation, inheritance, and utilization of its heritage, it fosters a unique system of cultural symbols and discourse rooted in contemporary Chinese context. According to a comprehensive survey conducted in 2008, the heritage structure of the Grand Canal consists of 1,154 heritage elements. These include not only the canal’s adjacent facilities such as docks and bridges but also industrial enterprises like factories and storage facilities that have developed directly or indirectly due to the canal. Additionally, it encompasses administrative and residential buildings historically associated with the canal’s development. The Tongzhou section of the Grand Canal (also known as the Tonghui River section) has played a significant role in the development of Beijing, witnessing the dramatic changes of the city and leaving behind rich historical remnants. It carries valuable cultural memories and is recognized as a segment of the World Heritage Site. Currently, the Tongzhou section of the Grand Canal includes three core attractions: the "Three Temples and One Pagoda," the Canal Olympic Park, and the Grand Canal Forest Park. Additionally, it extends to locations such as the City Green Heart Forest Park, Universal Studios Theme Park, Luxian Ancient City, Tongzhou Ancient City, Zhangjiawan Ancient Town, and Xihaizi Park. 2.2 Research Design In recent years, capturing and sharing short videos on social media platforms during travel has become a new trend, thanks to the rapid development of short video platforms. These short videos, posted on platforms such as Douyin (TikTok), not only document the scenery along the way but also provide important information about the visitors’ route and points of interest [ 11 ] . The rapid development and integration of online sharing platforms and geographic information systems have provided technical support for tracking the production and collection of data on tourists' spatial and temporal behavior distributions. Currently, mainstream short video platforms in China include Douyin, Kuaishou, WeChat Video Accounts, and Rednotes. Rednotes was excluded from this study due to its content often reflecting the perspective of content providers rather than visitors, which may differ from the analysis of "visitor scene preference." Douyin, being the most active and widely used short video platform in China, will be the primary focus of analysis in this study. On the Douyin platform, using keywords such as "Tongzhou", "Grand Canal", "Grand Canal Forest Park", and "Canal Park" for searching, dozens of topics such as #Grand Canal Forest Park, #Tongzhou Grand Canal, and #Tongzhou Canal were identified. Among them, there were 54 topics with over 10,000 views. The topic with the highest views was #Touring Tongzhou along the Grand Canal, with 2.622 million views; followed by the #Three Temples and One Pagoda topic, with 2.264 million views. However, compared to other popular Douyin topics in Tongzhou, such as #Wanda Tongzhou, #Tongzhou Marathon, and #Eating in Tongzhou, which have millions of views, the exposure of the Grand Canal topics is relatively low. Additionally, some topics, although not directly related to the Tongzhou Grand Canal, still contribute to its popularity, such as #Eating, Drinking, Having Fun, and Shopping in Tongzhou (with 16.88 million views). The research primarily employs the method that extracting keyframes from the videos and further extracting text information related to the destination's image from these keyframes, thereby analyzing the image characteristics of the cultural heritage in the short videos. Tourism videos often cover multiple scenes with frequent transitions and poor visual continuity, indicating significant differences between adjacent frames. Therefore, this study chooses an algorithm based on timed screenshots to extract the main scene images from the videos, converting dynamic videos into static images for analysis. 2.3 Research Process 2.3.1 Data Collection At the individual video level, sorting by the number of likes within the identified hashtags shows that, as of April 2025, the most popular single video had received 5,700 likes. In terms of video posting locations, videos published in the Tongzhou Grand Canal Cultural Tourism Area amassed a total of 9.558 million views, while those from the Grand Canal Forest Park received 7.56 million views. Both destinations ranked among the most popular scenic spots in Beijing according to viewer engagement on the platform.Douyin (TikTok) short video platform was selected as the source of material, with keywords such as "Grand Canal World Cultural Heritage" and "Tongzhou Grand Canal" used for searching. In the search results, the "video" category was chosen, and the results were sorted by the "most popular" tag. From January 2019 to December 2024, 200 videos relevant to the Tongzhou section of the Grand Canal were selected as video samples, while videos with less relevance to the Grand Canal, and those featuring only people or language descriptions, were excluded. Considering that some videos contain marketing content that may not objectively reflect the objects of visitors' attention, these videos were also excluded to ensure the accuracy of the research results. Finally, 113 short videos were obtained as data sources. Proprietary software platform were employed to analyze 113 tourism videos. To ensure the accuracy of the research results, redundant frames such as black screens and author Douyin IDs were removed. A total of 1327 keyframes were extracted from the videos, serving as the direct data source for this study. Finally, information regarding the video's release date and location was recorded from the short video platform and saved as an XLS file. 2.3.2 Data Processing Considering the limitations of manual image coding, this study introduces "image recognition" technology to automatically identify and classify a large number of images, with each image being assigned different keywords. In previous studies, static images were typically manually identified and classified, with the advantage of high accuracy but the drawback of a large workload and the potential influence of individual subjective factors on classification. Therefore, a proprietary software platform was developed, which utilized the image recognition technology provided by Baidu AI Cloud to identify keyframes and extract the main visual elements from images. The study primarily used the "landmark recognition," "text recognition," and "general object and scene recognition" APIs to identify keyframes. Landmark recognition extracted destination landmark information from the images, while general object and scene recognition extracted other scene information, including scene categories and specific scenes. These pieces of information represent the specific attributes of the destination image portrayed by the video. This method can save researchers a significant amount of work and effectively avoid the influence of subjective factors on classification results, showing high development potential. The study utilized a self-developed analysis tool based on the Python language to analyze video content. The tool first read the videos downloaded from the short video platform in advance from the local path, regularly captured keyframes of the videos at intervals of 10 seconds, and saved them in a folder for subsequent data processing. Using the "landmark recognition" and "general object and scene recognition" API interfaces, the tool batch-analyzed keyframe images of the videos and outputted keywords and classification information. It finally outputted descriptions of image content and exported them in Excel format; it also utilized the "text recognition" API interface to recognize subtitles and other text in keyframe images and exported them in TXT format. Finally, the entire process was integrated into one program, enabling the analysis of video screenshots, extraction of text descriptions, and recognition of text content to be performed at once after inputting the video file. The data obtained from the capture were identified, resulting in 6638 keywords belonging to 199 different categories from the recognition of 1327 keyframes, and were exported for further processing. Due to the large volume of data and the presence of significant interference, further manual processing of the data is required. According to the recognition results, keyword entries are generally more detailed and specific, while category entries are more generalized. For example, when the main subject of an image is a plant, the keyword entry will provide the specific name of the plant, while the category entry will only provide "Plants - Forests/Trees/Shrubs/Flowers." In this study, the demand for precise identification of image content is relatively low, while the demand for batch classification of image content is higher. Comparatively, category entries are more aligned with the requirements of this study. Therefore, this study will primarily analyze the recognition results based on category entries. This study carried out manual corrections to the data in the following aspects:1. Removal of Irrelevant Categories: Categories clearly unrelated to the research focus on tourist behavior were excluded. Some of these categories appeared frequently but had no analytical relevance, thereby potentially distorting the results. For example, categories such as "Products–Wearables," "Products–Clothing," and "Products–Eyewear" appeared over a hundred times, mainly due to the presence of people in the videos. These entries overlapped significantly with the "People–Close-Up" category and were therefore removed to avoid redundancy and bias. 2.Merging Overly Detailed Categories: Some categories were excessively subdivided. These were merged into broader, more meaningful groups based on their suffixes. For instance, “Plants–Leafy Plants” and “Plants–Ericaceae” were categorized under the more general labels of “Plants–Flowers” or “Plants–Trees” to streamline analysis. 3.Correction of Low-Accuracy Classifications: Categories with low image recognition accuracy were manually revised and reclassified. For example, "Natural Scenery–Ocean" was reassigned to "Natural Scenery–Lake," and "Non-Natural Image–Design Rendering" was corrected to "Non-Natural Image–Map." These adjustments helped improve the consistency and accuracy of the dataset used in this study. Based on the location information annotated in the videos and the location information obtained from "landmark recognition," one to two nodes were selected as representatives for each section of the canal. Ultimately, nine nodes were selected, including the Grand Canal Forest Park, Grand Canal Forest Park Viewing Platform, Canal Olympic Park, Canal Cultural Square, Three Temples and One Pagoda Scenic Area, Canal Ecological Park, Canal Sightseeing Wharf, Xihaizi Park, and Wenyu River Forest Park. This study believes that there is a certain correlation between the number of video releases and the number of visitors. The quantity of short videos related to a particular node can to some extent reflect the level of attention from visitors to that node. The number of videos associated with each node was counted, with each node individually counted for videos that included multiple nodes. 3 Research Findings 3.1 temporal characteristics of tourist behaviour This study analyzes the temporal characteristics of short videos posted on short video platforms. The time-related metadata of the videos is categorized by year, month, and season. In addition, videos containing night scenes are specifically identified and tagged. By locating keyframes labeled as “Architecture – Night View,” the corresponding source videos are traced. From these, ten representative night tourism videos are selected for further analysis using the data processing methods described earlier. Relevant keywords, categories, and captions associated with night tourism are then extracted and examined. 3.1.1 Interannual Variation Characteristics By analyzing the publication dates of sample short videos, the study attempts to reveal the interannual variation in tourist activity. Using year as the statistical unit, the number of video samples collected each year is shown in Fig. 10. The data indicate that the number of videos has increased year by year; however, this trend is primarily influenced by the content distribution mechanisms of short video platforms. Platforms like Douyin tend to push newer videos more prominently to users. Additionally, due to limitations in the platform's display algorithms and the large volume of data, very few videos posted before 2020 were retrievable through chronological searches. This results in a noticeable increase in video quantity after 2020. Given the limited time span covered, no clear interannual trends can be established, and the year-to-year data offer limited reference value. 3.1.2 Intermonth Variation Characteristics From a seasonal perspective, tourist activity in the Tongzhou section of the Grand Canal is influenced by climatic conditions. The site lies within a temperate monsoon climate zone, characterized by hot, rainy summers and cold, dry winters, resulting in significant seasonality in tourism activity. As shown in Fig. 11, April, May, August, and October see the highest number of video postings, coinciding with major public holidays such as Qingming Festival, Labor Day, summer vacation, and National Day. There is a clear "holiday effect" on tourism activity. The highest number of videos is published in summer (June–August), accounting for approximately 33% of the annual total. Winter (December–February) has the lowest count, at around 17%, while spring and autumn show roughly equal levels. This indicates that tourist visits to the Grand Canal World Cultural Heritage site in Tongzhou are concentrated mainly in summer, followed by spring and autumn, with the lowest activity in winter. On one hand, the summer influx of tourists from outside Beijing creates a spillover effect that increases visitor numbers at the Tongzhou section. While the site is less attractive to non-local tourists compared to traditional hotspots like the Forbidden City, the Great Wall, or the Summer Palace—and captions rarely mention keywords related to tourists from outside Beijing—some video texts indicate that this influx impacts the leisure spaces of local residents, prompting them to seek less-crowded destinations like the Grand Canal site for recreation. As tourist numbers increase across Beijing, the Tongzhou section likewise sees a rise in visitation. On the other hand, local residents’ demand for nearby outings also shifts seasonally. Due to climate comfort, urban dwellers tend to engage in nearby leisure travel during the more pleasant spring, summer, and autumn seasons. Additionally, available free time significantly influences travel behavior: summer vacation is a peak season for student and family travel, while holidays like May Day, National Day, and Qingming Festival mark major peaks in local travel. This confirms that both climate and leisure time significantly impact visitor behavior at the Grand Canal heritage site. 3.1.3 Night Tourism Behavior Night tourism behavior is also commonly observed in videos related to the Tongzhou section of the Grand Canal. Video recognition results show that the category “Architecture – Night Scenes” appeared 197 times, ranking 9th among all identified categories—only behind categories such as “Portraits,” “Lakes and Rivers,” “Trees,” and “Modern/Traditional Architecture.” As a time-sensitive indicator of tourist activity, the frequent appearance of “night scenes” strongly suggests the popularity of nighttime tourism at this site. Further analysis of the night-themed videos reveals several frequently occurring categories, as illustrated in Fig. 13. Activities such as “Sports,” “Artistic Performances,” and “Ceremonial Events” ranked 9th, 11th, and 12th respectively under the “Human Activity” category, while the appearance of “Watercraft – Boats” was also notably high. The remaining high-frequency terms largely align with those in the overall dataset. Two key factors explain the prominence of night tourism at this location. First, parts of the Tongzhou section serve as evening leisure spaces for residents, hosting activities like square dancing, night jogging, and casual walks. Second, the site has benefited from promotional efforts such as the Beijing (International) Grand Canal Cultural Festival, which has played a significant role in encouraging nighttime visits. According to the video content, the festival features various events, including canal boat tours, immersive performances, archery and ceremonial sport experiences, intangible cultural heritage exhibitions, and academic forums. Among the 10 sampled night tourism videos, 6 were directly related to the festival—suggesting a strong correlation between festival events and nighttime visitor activity. In summary, interannual variation in tourist activity at the Grand Canal site in Tongzhou is difficult to assess due to limitations in the availability of earlier video data; thus, no reliable trend can be established. In contrast, intermonth variation shows clear seasonality, with video posting peaks aligning with major holidays and Beijing’s peak tourist seasons. These trends are driven both by the seasonal influx of non-local tourists and the changing recreational needs of local residents. Regarding night tourism, visitor engagement is notably high, particularly in cultural, ceremonial, and sports-related activities, influenced by both everyday leisure habits and organized festival events. 3.2 Determination of Scene Type Indicators Tourist scenes refer to the comprehensive tangible and intangible environmental elements that tourists perceive at tourist destinations. Before analyzing the characteristics of visitors' scene preferences, it is necessary to determine the dimensions of scenes to which the elements obtained by machine recognition of video keyframes belong. In this study, we refer to the classification research on dimensions of tourist experience scenes by Zhang Hui et al. [ 30 ] , and categorize tourist scenes into historical and cultural experiences, natural experiences, consumption experiences, sports experiences, urban views, festival activities, etc. According to this classification method, the "root" keywords obtained by machine recognition of video keyframes are reclassified. When classifying, irrelevant content related to landscape types is removed, and some content that is difficult to classify is manually processed. If multiple content units appear in one keyframe, they are counted separately. Table 1 scene type indicator system Scene type Corresponding content of machine recognition Historical and cultural experience scenes Constructions-traditional construction/cultural heritage…; goods- craft sculpture/painting… Consumption experience scenes Goods-food/toy/agricultural material/ sporting goods…; constructions-store and mall Natural experience scenes Natural landscapes-river/lake/waterfall/spring…; animals-fish/bird/dog/beetle…; plants-tree/shrub/flower/grass…; constructions-garden/natural park Sports experience scenes Human activities-sport activities/agricultural production…; transportation vehicles-bicycle/ships… Urban landscape scenes Constructions-modern constructions/architectural night scenery/landscape sketch… Festive events scenes Human activities- Literary and artistic activities/ceremonial activities 3.3 Overall Analysis of Scene Preferences This study completed the classification of scene categories contained in video keyframes through a combination of machine recognition and manual classification. The table below shows the number of scene types contained in keyframes. Table 2 Number of each scene type Scene type number Natural experience scenes 1735 Historical and cultural experience scenes 854 Urban landscape scenes 804 Consumption experience scenes 269 Sports experience scenes 108 Festive events scenes 89 Among all scene classifications, the "natural experience scenes" category accounts for the highest proportion, approximately 36% of the total. This category includes elements such as natural parks, rivers, vegetation, and weather. Upon further segmentation of this category, it is evident that vegetation elements dominate, including grasslands and trees along the canal, as well as most scenes in the Grand Canal Forest Park. Visitors often focus on vegetation when experiencing nature. The next most prominent element is rivers. As one of the core elements of the Grand Canal World Cultural Heritage, rivers play an important role in natural experience scenes. Even if visitors do not consider rivers the primary scene element, they still pay considerable attention to them while touring along the Grand Canal. While animals, weather, and climate are also important parts of natural experience scenes, their significance is relatively low in the Tongzhou section of the Grand Canal. "Historical and cultural experience scenes" appeared 854 times, with scene elements focusing on traditional architecture, handicraft sculptures, and paintings, among other material cultural heritage. The Tongzhou section of the Grand Canal boasts rich material and intangible cultural heritage. It is a UNESCO World Heritage site with abundant historical and cultural relics and significant potential for cultural activities. However, compared to natural experience scenes, historical and cultural experience scenes receive less attention, which does not match the rich heritage resources of the Tongzhou section of the Grand Canal. Many excellent heritage resources have not been fully developed and utilized, and the creation of historical and cultural experience scenes is still immature. Therefore, in future Tourist Scene creation, the cultural characteristics of national cultural parks should be fully leveraged, and the cultural resources along the canal should be effectively utilized to create historical and cultural experience scenes that can attract visitors. "Urban landscape scenes" encompass surrounding buildings, bridges, park facilities, and more. Upon further segmentation of the elements in this category, it is evident that modern architecture dominates, including buildings along the Grand Canal, bridges spanning the canal, sports venues, etc. Following closely are pathways and urban roads, which constitute basic infrastructure. During the process of shooting short videos, visitors often find it difficult to avoid including surrounding buildings in their shots. However, this also reflects the significant role of surrounding buildings in shaping the scene of the Grand Canal: due to the vague boundaries of the Grand Canal, visitors not only focus on the scenery within the park but also inevitably pay attention to the landscape along a certain distance of the canal. "Consumption experience scenes," "sports experience scenes," and "festive events scenes" appear relatively infrequently. The connection between consumption experience scenes and festive events scenes is close, but both are currently receiving low attention from visitors in the Tongzhou section of the Grand Canal. As for sports experience scenes, upon closer examination of specific videos, it is evident that due to the limited participation of out-of-town visitors in sports activities and the lack of prominent sports features in the Tongzhou section of the Grand Canal, only a few sports events attract special attention from out-of-town visitors. Therefore, most of the participants in sports activities are local residents, and the intensity of sports activities they engage in is relatively low, usually involving activities such as square dancing, cycling, walking, and jogging. These sports activities typically occur in everyday scenarios, and the desire of local residents to share these daily life activities is far less than that of out-of-town visitors to share their travel experiences, resulting in fewer such contents in short videos. In fact, in the process of creating tourist scenes, although community residents receive less attention on media platforms, their needs and interests must not be ignored. 3.4 Analysis of Scene Preferences at Different Nodes The collected short videos were divided according to nodes, with each video related to a specific node marked. Then, utilizing the data processing method described in the last chapter, the elements depicted in the videos were categorized into different scenes, thus obtaining the scene preference characteristics of different nodes. By categorizing the elements contained in the short videos corresponding to nodes such as the Grand Canal Forest Park, the Three Temples and One Pagoda, the Canal Olympic Park, and the Canal Cultural Square into the aforementioned scene types, the following results were obtained: Table 3 Number of each scene type in the Grand Canal Forest Park Scene type number Natural experience scenes 1328 Urban landscape scenes 397 Consumption experience scenes 96 Historical and cultural experience scenes 79 Sports experience scenes 64 Festive events scenes 34 Table 4 Number of each scene type in the Three Temples and One Pagoda Scene type number Historical and cultural experience scenes 612 Natural experience scenes 257 Urban landscape scenes 165 Consumption experience scenes 54 Festive events scenes 43 Sports experience scenes 21 Table 5 Number of each scene type in the Canal Olympic Park Scene type number Sports experience scenes 43 Urban landscape scenes 41 Natural experience scenes 21 Consumption experience scenes 13 Historical and cultural experience scenes 6 Festive events scenes 2 Table 6 Number of each scene type in the Canal Cultural Square Scene type number Urban landscape scenes 78 Festive events scenes 57 Consumption experience scenes 44 Historical and cultural experience scenes 42 Sports experience scenes 19 Natural experience scenes 14 It can be observed that, due to the dense presence of natural elements in the Grand Canal Forest Park, natural experiential scenes receive the most attention at this node. This node stands out with a higher proportion of natural scene preferences compared to the overall Tongzhou section. The park features extensive green spaces, a variety of plant species, and water bodies that together form a serene and immersive environment, offering visitors a strong sense of natural immersion and relaxation. This appeal is particularly evident among urban residents seeking respite from the pressures of city life. In addition, the Forest Park Observation Deck serves as a strategic viewpoint that offers panoramic views of the surrounding urban skyline and iconic bridge nightscapes. As a result, interest in urban landscape scenes at this node is also notably high, as visitors are drawn to the contrast between natural tranquility and urban vitality. However, other types of scenes—particularly historical and cultural experiential scenes—receive relatively limited attention at this site. This is primarily because the area lacks substantial historical and cultural heritage elements, resulting in comparatively lower visitor engagement with such content. In contrast, the Three Temples and One Pagoda node demonstrates a pronounced increase in visitors’ preference for historical and cultural experiential scenes. This trend can be attributed to the concentration of well-preserved heritage assets in the area, such as ancient temples, religious structures, and traditional architectural elements that embody the region's historical significance. This node serves as a cultural core of the Tongzhou section, offering rich interpretive potential and educational value. Visitors to this node are more likely to engage with the site's cultural narratives, take part in heritage tours, and capture content reflecting traditional aesthetics. As interest in cultural experiences rises at this location, it naturally leads to a decline in attention to other scene categories such as natural, sports, or leisure-based scenes, indicating a focused rather than diversified visitor engagement pattern. At the Canal Olympic Park node, a different pattern emerges. This area is characterized by extensive sports infrastructure and the availability of water-based recreational activities such as kayaking and rowing. Consequently, visitor preferences here skew toward sports experiential scenes, as tourists and local residents alike participate in or observe athletic events and outdoor fitness activities. In addition, the park's open layout and integration with waterfront spaces provide scenic views of the canal and surrounding urban developments, leading to a heightened interest in urban landscape scenes as well. Nevertheless, other types of scenes—particularly those tied to culture or history—are underrepresented at this node, reflecting its modern and activity-oriented positioning within the broader heritage landscape. The Canal Cultural Square is another key node that displays a distinct pattern of visitor preference. Architecturally, the square is marked by tall, colorful archways, artistic landscape sculptures, and intricately carved granite pavement, all of which contribute to its visual identity and photogenic appeal. These features make the square a favored location for capturing urban landscape content, particularly during evening hours when the bridge lighting and city skyline create a dramatic backdrop. Beyond its architectural appeal, the square also functions as a major site for public events, community celebrations, and informal recreational activities such as square dancing, musical performances, and seasonal gatherings. These events contribute to a higher frequency of festival and event-related scenes in visitor-generated content. As such, the Canal Cultural Square reflects a dual preference among visitors—for urban aesthetics and socially driven experiences—making it one of the more dynamic and multifunctional spaces along the Tongzhou section. For other nodes, due to the limited mentions in the short videos and insufficient data volume, there is no clear preference relationship between scenes. It is difficult to draw effective conclusions based on existing data. In future research, data retrieval based on the names of other nodes can be conducted to enrich the results of this study. 4 Conclusion and Discussion 4.1 conclusion Through machine processing of visitors' related videos posted on the Douyin short video platform, an analysis of visitors' spatiotemporal behavior characteristics and landscape preferences leads to the following conclusions: From the perspective of temporal distribution, annual variations in tourist volume are not clearly observable due to limitations in short video platform data, making it difficult to identify consistent interannual trends. However, on a seasonal scale, tourist activity shows pronounced seasonality, largely influenced by climate conditions and the availability of leisure time. Nighttime tourism is also notably popular in the Tongzhou section of the Grand Canal World Cultural Heritage Site. This is driven both by the recreational habits of local residents and by the promotional impact of festivals and events organized by park authorities. In terms of visitors' scene preference characteristics, visitors' preferences for various types of scenes along the canal are as follows: natural experience scenes > historical and cultural experience scenes > urban landscape experience > consumer experience scenes > sports experience scenes > festival event scenes. Further analysis of specific elements reveals that the attention to historical and cultural experience scenes does not match the rich cultural heritage resources of the Tongzhou section of the Grand Canal. Many excellent heritage resources have not been fully developed and utilized. Regarding the shaping of sports experience scenes, although visitors do not show a high preference for sports experience scenes, in reality, most of the sports activities conducted here are by community residents. Their needs should not be overlooked when shaping the scenes. Visitors' scene preferences at different nodes vary according to the characteristics of the nodes themselves. At the Grand Canal Forest Park node, visitors show a strong preference for natural experiences; at the Three Temples and One Pagoda node, visitors have a higher preference for historical and cultural experience scenes; at the Canal Olympic Park node, visitors prefer urban landscape and sports experience scenes; at the Canal Cultural Square node, visitors prefer urban landscape and festival activity scenes. Meanwhile, due to the vague boundaries of the Grand Canal National Cultural Park, landscapes and cultural activities outside the scenic area have a significant impact on visitors' perceptions. 4.2 discussion To promote further development of the Tongzhou section of the Grand Canal, this study proposes the following recommendations for park managers and developers based on the research conclusions: 1.Focus on visitors' scene preferences. Addressing the low attention to historical and cultural experience scenes, efforts should be made to enhance the development and utilization of the rich historical and cultural heritage resources along the Tongzhou section of the Grand Canal. This can be achieved by integrating heritage resources with natural experiences, festival activities, and other scenes. Since visitors show a relatively high preference for urban landscape experiences, it is recommended to strengthen the integration of urban landscape elements throughout the entire Tongzhou section, including planning for urban landscapes and cultural activities beyond the boundaries of the scenic area, to enhance visitors' overall perception and experience. Tailor Tourist Scenes to the specific preferences of visitors and residents at different nodes, creating unique and attractive Tourist Scenes. 2. Pay attention to the actual needs of community residents. When developing scenes for cultural heritage, not only should the construction of scenic spots and supporting facilities along the route be considered, but also the cultural and natural environments within a certain range should be taken into account. In the planning and development process of the heritage, adopt a community-participation planning approach to understand the needs and expectations of local residents. Collect their opinions through forums, questionnaires, etc., to ensure that the park's design and services truly meet the practical needs of local residents. Understand the community residents' needs for sports and leisure, and consider adding appropriate sports facilities such as basketball courts, fitness areas, etc. Organize cultural activities such as community performances, handicraft markets, etc., to attract active participation from community residents. These activities not only enhance the social atmosphere of the park but also strengthen its role as a center for community cultural exchange. Declarations Data availability The author confirms that all data generated or analysed during this study are included in this published article and database. The datasets analysed during the current study are available in the Dataverse repository. Furthermore, primary and secondary sources and data supporting the findings of this study were all publicly available at the time of submission. Ethical Approval This study was conducted in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki and its later amendments. As the research involved only the analysis of publicly available user-generated content (UGC) on short video platforms, without any direct involvement or interaction with human subjects, biological material, or personal identifiers, formal ethical approval was not required. Informed Consent This study did not involve any direct interaction with human participants, and all data were collected from publicly available short video content. No identifiable personal data, private information, or images requiring consent were collected, analyzed, or published. Therefore, informed consent was not applicable. The authors affirm that all procedures respected the rights to privacy and anonymity of individuals, and any potentially sensitive content was excluded from analysis. References Becker R A, Caceres R, Hanson K, et al. 2011, A Tale of One City: Using Cellular Network Data for Urban Planning[J]. IEEE pervasive computing, 10(4):18-26. Birkin M, Malleson N. 2011, Microscopic simulations of complex metropolitan dynamics[J]. the Complex City workshop, Unpublished. Deng N, Qu L. 2022, Comparison of Destination Images Based on Video Analysis through Machine Learning——A Case Study on YouTube Videos of Beijing [J]. Tourism Tribune, 37(08):70-85. Dai J, Huang X, Sun J, et al. 2023, Research on the Formation Mechanism of Tourist Cultural Identity in National Cultural Park——Taking the Iconic Natural Landscapes of the Yellow River National Cultural Park as an Example [J]. Tourism Tribune, 38(01):31-41. Girardin F, Calabrese F, Fiore F D, et al. 2008, Digital Footprinting: Uncovering Tourists with User-Generated Content[J]. IEEE pervasive computing, 7(4):36-43. Girardin F, Fiore F D, Ratti C, et al. 2008, Leveraging explicitly disclosed location information to understand tourist dynamics: a case study[J]. Journal of location based services, 2(1):41-56. Gong L. 2019, The Grand Canal: From Cultural Landscape Heritage to National Cultural Park [J]. The Mass Weekly, 2019(24):17-18. Gong D. 2021, The Origin and Characteristics of the Concept of National Cultural Park [J]. Chinese Landscape Architecture, 37(06):38-42. Han Z, Ren H. 2022, Significance,Principle and Strategy of National Cultural Park Trail Construction——Take the Great Wall National Cultural Park as an Example [J]. Journal of Northwest Normal University(Social Sciences), 59(05):23-30. Li M, Ma Q, Zhao X. 2013, A Review of Researches on Servicescape and Future Prospects [J]. Foreign Economics & Management, 35(04):62-70. Li X. 2017, Research of Short Video Communication Strategy and effect——YiTiao TV as an example[D]. Anhui University. Li F, Zou C, 2021, National Culture Park: Logic, Origins and Implications [J]. Tourism Tribune, 36(01):14-26. Li S. 2021, Functions, Values, and Implementation Approaches of National Cultural Parks [J]. China Economic & Trade Herald, 2021(03):152-155. Liu L, Wang Q. 2021, Reflections on the Construction Path of the Long March National Cultural Park: A Case Study of the Sichuan Section of the Long March National Cultural Park [J]. Mao Zedong Thought Study, 38(01):108-113. Negroponte N. Being Digital[M]. 1996. Nerute K S. 2012, Digital Footprints in the Context of Professional Ethics[J]. Informatics in education, 11(1):65-79. Pan Z, Liang B. 2015, Tourist Perception of Tourism Destination Image Analysis Based on Digital Footprints:A Case Study of World Natural Heritage Tianshan Mountain [J]. Journal of Beijing International Studies University, 37(05):40-48. Shan J, 2006, Preliminary Discussion on the Protection of Large Linear Cultural Heritage: Breakthrough and Pressure [J]. Cultural Relics in Southern China, 2006(03):2-5. Sun H. 2021, Preliminary Study on National Cultural Parks: Concepts, Types, Characteristics, and Construction [J]. China Cultural Heritage, 2021(05):4-14. Song Y, Zhang Y. 2022, On the design and construction of environmental aesthetic experience in Hangzhou Section of Grand Canal National Cultural Park [J]. Journal of Zhejiang University of Technology(Social Sciences), 21(03):308-313. Thatcher J. 2014, Living on Fumes: Digital Footprints, Data Fumes, and the Limitations of Spatial Big Data[J]. International Journal of Communication. Vaccari A, Liu L, Biderman A, 2009, et al. A holistic framework for the study of urban traces and the profiling of urban processes and dynamics, [C]. IEEE, 2009. Weaver S D, Gahegan M. 2007, Constructing, Visualizing, and Analyzing A Digital Footprint[J]. Geographical review, 97(3):324-350. Wober K. 2007, Similarities in Information Search of City Break Travelers - A Web Usage Mining Exercise[J]. Institute for Tourism and Leisure Studies, DBLP. Wang X. 2014, The Research Review on Tourism Preference at Home and Abroad [J]. Modern Urban Research, 2014(01):110-115. Wang K, 2021, Theoretical Exploration and Practical Review of National Cultural Park [J]. enterprise.economy, 40(04):5-12. Wang X, Bai L. 2021, Logical Following and Path Exploration of the Construction of Grand Canal National Cultural Park: Based on the Dual Theoretical Perspectives of Cultural Memory and Spatial Production [J]. Zhejiang Social Sciences, 2021(10):72-80. Wu D, Liu H, Wang B. 2021, Realistic Misunderstanding and Improved Suggestions during the Construction of National Cultural Park [J]. Research on Development, 2021(03):1-7. Yu C, Lu L. 2008, A Comment on Overseas Studies of Tourists' Preferences for Tourism in Recent Years [J]. Journal of Anhui Normal University(Natural Science), 31(06):590-595. Zhang W, Tao Z, Qin L, et al. 2016, Space Response of Tourists' Digital Footprint in Suzhou Ancient Garden Based on Online Travel Note [J]. Resource Development & Market, 32(07):886-891. Zhang H, Xu H. 2019, A structural model of liminal experience in tourism[J]. Tourism Management, 71:84-98. Zou C, Chang M, Lai M, 2019, International Experience and Reference on the Management Model of National Cultural Parks [N]. China Tourism News, 11(05), 003 Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted 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-6619412","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":494556317,"identity":"55b000e5-069e-420a-a0dd-55a6b0516787","order_by":0,"name":"Xuanrui Gu","email":"","orcid":"","institution":"Beijing Forestry University","correspondingAuthor":false,"prefix":"","firstName":"Xuanrui","middleName":"","lastName":"Gu","suffix":""},{"id":494556318,"identity":"910f7de4-603e-4c7f-889d-ba1b4fd5c74a","order_by":1,"name":"Zhongjun Wang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA+ElEQVRIie3PMWsCMRTA8RcO7PLqfIdiv8KTg8Ph4L5KSuGmIE7SrUJBl4DriV+iHyFyUJdY17jpKCgoLnZrrHvuxkLzX5JAfiQPwOf7s2GKbCKBXm8HVY/k7UBqIF2fQJk+FAKgFqHlqtxf2zk+zo5qsBhDp2k4uwxcRPfzXmhnieZ9TpbEkeFBq3CQRImECHPszgXF5zE8fxjeCNBF1oeEOJaYbTTdXnmrJkbEW2UJK/CXcKoimTkkbGQ/xqSdRX2F3ZnevbdcJJqK+Pwt04xNVp+khulTc/myuLiIrREyed8QQGhXNnIDgOAE1/tmW3XV5/P5/mc/kNNQKZHsHbQAAAAASUVORK5CYII=","orcid":"","institution":"Beijing Forestry University","correspondingAuthor":true,"prefix":"","firstName":"Zhongjun","middleName":"","lastName":"Wang","suffix":""},{"id":494556319,"identity":"9f0140aa-3bcb-4703-9136-5ff27667c46e","order_by":2,"name":"Jinxuan Qin","email":"","orcid":"","institution":"Beijing Forestry University","correspondingAuthor":false,"prefix":"","firstName":"Jinxuan","middleName":"","lastName":"Qin","suffix":""}],"badges":[],"createdAt":"2025-05-08 10:08:09","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6619412/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6619412/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":88310240,"identity":"c1b4ed81-175b-41c9-a7c5-bb111b19b84b","added_by":"auto","created_at":"2025-08-05 06:50:23","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":459366,"visible":true,"origin":"","legend":"\u003cp\u003eLocation Map of the Grand Canal National Park in Tongzhou Section\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-6619412/v1/e45aec258a761c9c4525e6d7.png"},{"id":88310242,"identity":"4fc4714e-c40f-4f75-9055-3f79842528b0","added_by":"auto","created_at":"2025-08-05 06:50:23","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":1167044,"visible":true,"origin":"","legend":"\u003cp\u003eDivision of River Segments into Nodes\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-6619412/v1/726d8539e30da36ecc31f670.png"},{"id":88310238,"identity":"a06a2506-18ed-42ba-a7e5-62ca85ef166c","added_by":"auto","created_at":"2025-08-05 06:50:23","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":59944,"visible":true,"origin":"","legend":"\u003cp\u003eInterannual variation characteristics of the number of short video samples\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-6619412/v1/2193eee9cf3e21dfd295f71a.png"},{"id":88311619,"identity":"28f2b765-339c-4577-925b-0fdbaa713b35","added_by":"auto","created_at":"2025-08-05 06:58:23","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":91748,"visible":true,"origin":"","legend":"\u003cp\u003eIntermonth variation characteristics of the number of short video samples\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-6619412/v1/fb5f62d9025a9cc4b56ca84e.png"},{"id":88311859,"identity":"4d5a2fc6-38f9-4792-9d34-41150485e9b5","added_by":"auto","created_at":"2025-08-05 07:06:23","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":98658,"visible":true,"origin":"","legend":"\u003cp\u003eRanking of the number of keywords appearing in the night view video\u003c/p\u003e","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-6619412/v1/8b250af4244867f42face8c3.png"},{"id":92664879,"identity":"496f710f-0c1e-4972-b4f2-3c9b785538ba","added_by":"auto","created_at":"2025-10-02 16:01:51","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2553666,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6619412/v1/64086bb6-0e2f-4e60-aa36-5a72d8740e3a.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Analyzing Tourists’ Scene Preferences through Digital Footprints in World Heritage Site: A Case Study of the Grand Canal’s Tongzhou Section","fulltext":[{"header":"1 Introduction","content":"\u003cp\u003eIn June 2014, the Grand Canal of China was successfully inscribed on the World Heritage List. In July 2019, the \"Construction Plan for the Great Wall, Grand Canal, and Long March National Cultural Parks\" was approved, marking the official commencement of the construction of national cultural parks\u003csup\u003e[\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]\u003c/sup\u003e. With the continuous development of China's economy and society, consumer upgrades and the evolution of demand have become important issues in the new era of social consumption. In the cultural tourism industry, traditional sightseeing tourism has begun to transform to meet the increasingly diversified needs of tourists.\u003c/p\u003e\u003cp\u003eIn the era of integrated development of culture and tourism, protecting and showcasing large-scale linear cultural heritage and cultural routes has become an increasingly hot topic for research. Scholars discuss the Grand Canal as linear cultural heritage, drawing comparisons with European cultural route theories and American heritage corridor theories, while also debating whether these places necessarily exhibit the form of linear cultural heritage\u003csup\u003e[\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e, \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e]\u003c/sup\u003e. Currently, academic research on linear cultural heritage and cultural routes including the Grand Canal mainly focuses on summarizing foreign experiences\u003csup\u003e[\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e, \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e]\u003c/sup\u003e, exploring the origins of the concept\u003csup\u003e[\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e]\u003c/sup\u003e, studying their functions and values\u003csup\u003e[\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]\u003c/sup\u003e, as well as empirical research on specific park planning and construction\u003csup\u003e[\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]\u003c/sup\u003e. However, there has been relatively little research on visitor behavior characteristics and preference within World Cultural Heritage sites, making it essential to study the scene preference of visitor in these heritages. Research on visitor\u0026rsquo;s scene preference in World Cultural Heritage sites based on digital footprints not only provides reference and support for the management and planning of heritage sites but also deepens understanding of visitor needs and behaviors, thereby facilitating the sustainable development of heritage.\u003c/p\u003e\u003cp\u003eIn the field of digital footprint research, Negroponte, N. proposed the concept of the \"Slug trail\" in the 1990s, which is generally considered as a precursor to the \"digital footprint\"\u003csup\u003e[\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]\u003c/sup\u003e. In 2007, Stephen, D. first introduced the concept of the \"digital footprint,\" defining it as the data traces left by individuals interacting with digital environments or media\u003csup\u003e[\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e]\u003c/sup\u003e. Since then, numerous scholars have studied this concept and proposed different definitions\u003csup\u003e[\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]\u003c/sup\u003e. Regarding the application of digital footprints in tourism, current research both domestically and internationally often relies on data obtained from online texts such as travelogues and microblogs, as well as online photos\u003csup\u003e[\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e]\u003c/sup\u003e, while research on obtaining User-Generated Content (UGC) videos from emerging short video media platforms is scarce\u003csup\u003e[\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]\u003c/sup\u003e. Therefore, this study incorporates short videos as a data source into digital footprint research to address the lack of academic research in this specific field.\u003c/p\u003e\u003cp\u003eThe theoretical framework of this study encompasses theories of tourist preferences and tourist experiential scenes. Tourist preference refers to the cognitive and affective components of tourists' inclinations towards a specific tourism product or destination, manifested by their emotional and intentional factors. It not only involves explicit preferences displayed during the travel process but also encompasses abstract and emotional tourism preferences\u003csup\u003e[\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]\u003c/sup\u003e. Tourist preferences can be categorized into two aspects: explicit and stated preferences\u003csup\u003e[\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e]\u003c/sup\u003e. This study focuses on identifying tourists' preferences for attractions based on their geographic location information, falling under the category of explicit preferences research.\u003c/p\u003e\u003cp\u003eTourist experience scenes refer to the comprehensive perception of all tangible environments and intangible services that tourists encounter throughout their travel behaviors, a concept derived from Bitner's service scene\u003csup\u003e[\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]\u003c/sup\u003e. Building upon this foundation, Zhang Hui et al. proposed the concept of tourist experience scenes, which encompasses four dimensions: physical, social, socio-symbolic, and natural dimensions\u003csup\u003e[\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e]\u003c/sup\u003e. Meanwhile, research on tourist scenes and visitor behavior characteristics has developed relatively mature theoretical frameworks, with rich research content and methods. However, these aspects of research lack the incorporation of emerging short video media as a data source. Therefore, this study will conduct analysis based on short video media.\u003c/p\u003e"},{"header":"2 Research Design and Process","content":"\u003cp\u003eThe study aims to acquire visitor digital footprints of the Tongzhou section of the Grand Canal World Cultural Heritage through short video platforms, analyze visitors' preferences for Tourist Scenes, and attempt to construct the mechanism and logical origins of visitor scene preferences in heritage sights. This analysis seeks to understand the strengths and weaknesses of tourism space development in the Tongzhou section of the Grand Canal, with the goal of providing targeted opinions and strategies for the planning, enhancement, construction, and management of the heritage.\u003c/p\u003e\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003e2.1 Overview of Research Destination\u003c/h2\u003e\u003cp\u003eThe Grand Canal in China is a dual heritage of nature and culture, boasting distinct natural and cultural ecosystems\u003csup\u003e[\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]\u003c/sup\u003e. China\u0026rsquo;s Grand Canal is both a natural and cultural World Heritage site, characterized by its distinct ecological and cultural systems. It functions not only as an ecological corridor linking northern and southern China but also as a cultural corridor that connects and integrates six major cultural regions: Beijing-Tianjin, Yan-Zhao, Qi-Lu, Central Plains, Huaiyang, and Wu-Yue. The canal embodies cultural diversity, reflects the essence of Chinese civilization, and carries profound national memory and cultural identity.\u003c/p\u003e\u003cp\u003eTherefore, the cultural heritage of the Grand Canal is not intended to serve merely as a conventional cultural landmark. Rather, it is built upon the canal\u0026rsquo;s iconic tangible and intangible cultural heritage and aims to establish a symbolic representation of Chinese national culture in the new era. Guided by the strategic goal of promoting national spirit, the canal serves as a platform for public cultural expression. Through the effective preservation, inheritance, and utilization of its heritage, it fosters a unique system of cultural symbols and discourse rooted in contemporary Chinese context.\u003c/p\u003e\u003cp\u003eAccording to a comprehensive survey conducted in 2008, the heritage structure of the Grand Canal consists of 1,154 heritage elements. These include not only the canal\u0026rsquo;s adjacent facilities such as docks and bridges but also industrial enterprises like factories and storage facilities that have developed directly or indirectly due to the canal. Additionally, it encompasses administrative and residential buildings historically associated with the canal\u0026rsquo;s development.\u003c/p\u003e\u003cp\u003eThe Tongzhou section of the Grand Canal (also known as the Tonghui River section) has played a significant role in the development of Beijing, witnessing the dramatic changes of the city and leaving behind rich historical remnants. It carries valuable cultural memories and is recognized as a segment of the World Heritage Site. Currently, the Tongzhou section of the Grand Canal includes three core attractions: the \"Three Temples and One Pagoda,\" the Canal Olympic Park, and the Grand Canal Forest Park. Additionally, it extends to locations such as the City Green Heart Forest Park, Universal Studios Theme Park, Luxian Ancient City, Tongzhou Ancient City, Zhangjiawan Ancient Town, and Xihaizi Park.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec4\" class=\"Section2\"\u003e\u003ch2\u003e2.2 Research Design\u003c/h2\u003e\u003cp\u003eIn recent years, capturing and sharing short videos on social media platforms during travel has become a new trend, thanks to the rapid development of short video platforms. These short videos, posted on platforms such as Douyin (TikTok), not only document the scenery along the way but also provide important information about the visitors\u0026rsquo; route and points of interest\u003csup\u003e[\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]\u003c/sup\u003e. The rapid development and integration of online sharing platforms and geographic information systems have provided technical support for tracking the production and collection of data on tourists' spatial and temporal behavior distributions. Currently, mainstream short video platforms in China include Douyin, Kuaishou, WeChat Video Accounts, and Rednotes. Rednotes was excluded from this study due to its content often reflecting the perspective of content providers rather than visitors, which may differ from the analysis of \"visitor scene preference.\" Douyin, being the most active and widely used short video platform in China, will be the primary focus of analysis in this study.\u003c/p\u003e\u003cp\u003eOn the Douyin platform, using keywords such as \"Tongzhou\", \"Grand Canal\", \"Grand Canal Forest Park\", and \"Canal Park\" for searching, dozens of topics such as #Grand Canal Forest Park, #Tongzhou Grand Canal, and #Tongzhou Canal were identified. Among them, there were 54 topics with over 10,000 views. The topic with the highest views was #Touring Tongzhou along the Grand Canal, with 2.622\u0026nbsp;million views; followed by the #Three Temples and One Pagoda topic, with 2.264\u0026nbsp;million views. However, compared to other popular Douyin topics in Tongzhou, such as #Wanda Tongzhou, #Tongzhou Marathon, and #Eating in Tongzhou, which have millions of views, the exposure of the Grand Canal topics is relatively low. Additionally, some topics, although not directly related to the Tongzhou Grand Canal, still contribute to its popularity, such as #Eating, Drinking, Having Fun, and Shopping in Tongzhou (with 16.88\u0026nbsp;million views).\u003c/p\u003e\u003cp\u003eThe research primarily employs the method that extracting keyframes from the videos and further extracting text information related to the destination's image from these keyframes, thereby analyzing the image characteristics of the cultural heritage in the short videos. Tourism videos often cover multiple scenes with frequent transitions and poor visual continuity, indicating significant differences between adjacent frames. Therefore, this study chooses an algorithm based on timed screenshots to extract the main scene images from the videos, converting dynamic videos into static images for analysis.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec5\" class=\"Section2\"\u003e\u003ch2\u003e2.3 Research Process\u003c/h2\u003e\u003cdiv id=\"Sec6\" class=\"Section3\"\u003e\u003ch2\u003e2.3.1 Data Collection\u003c/h2\u003e\u003cp\u003eAt the individual video level, sorting by the number of likes within the identified hashtags shows that, as of April 2025, the most popular single video had received 5,700 likes. In terms of video posting locations, videos published in the Tongzhou Grand Canal Cultural Tourism Area amassed a total of 9.558\u0026nbsp;million views, while those from the Grand Canal Forest Park received 7.56\u0026nbsp;million views. Both destinations ranked among the most popular scenic spots in Beijing according to viewer engagement on the platform.Douyin (TikTok) short video platform was selected as the source of material, with keywords such as \"Grand Canal World Cultural Heritage\" and \"Tongzhou Grand Canal\" used for searching. In the search results, the \"video\" category was chosen, and the results were sorted by the \"most popular\" tag. From January 2019 to December 2024, 200 videos relevant to the Tongzhou section of the Grand Canal were selected as video samples, while videos with less relevance to the Grand Canal, and those featuring only people or language descriptions, were excluded. Considering that some videos contain marketing content that may not objectively reflect the objects of visitors' attention, these videos were also excluded to ensure the accuracy of the research results. Finally, 113 short videos were obtained as data sources.\u003c/p\u003e\u003cp\u003eProprietary software platform were employed to analyze 113 tourism videos. To ensure the accuracy of the research results, redundant frames such as black screens and author Douyin IDs were removed. A total of 1327 keyframes were extracted from the videos, serving as the direct data source for this study. Finally, information regarding the video's release date and location was recorded from the short video platform and saved as an XLS file.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec7\" class=\"Section3\"\u003e\u003ch2\u003e2.3.2 Data Processing\u003c/h2\u003e\u003cp\u003eConsidering the limitations of manual image coding, this study introduces \"image recognition\" technology to automatically identify and classify a large number of images, with each image being assigned different keywords.\u003c/p\u003e\u003cp\u003eIn previous studies, static images were typically manually identified and classified, with the advantage of high accuracy but the drawback of a large workload and the potential influence of individual subjective factors on classification. Therefore, a proprietary software platform was developed, which utilized the image recognition technology provided by Baidu AI Cloud to identify keyframes and extract the main visual elements from images. The study primarily used the \"landmark recognition,\" \"text recognition,\" and \"general object and scene recognition\" APIs to identify keyframes. Landmark recognition extracted destination landmark information from the images, while general object and scene recognition extracted other scene information, including scene categories and specific scenes. These pieces of information represent the specific attributes of the destination image portrayed by the video. This method can save researchers a significant amount of work and effectively avoid the influence of subjective factors on classification results, showing high development potential.\u003c/p\u003e\u003cp\u003eThe study utilized a self-developed analysis tool based on the Python language to analyze video content. The tool first read the videos downloaded from the short video platform in advance from the local path, regularly captured keyframes of the videos at intervals of 10 seconds, and saved them in a folder for subsequent data processing. Using the \"landmark recognition\" and \"general object and scene recognition\" API interfaces, the tool batch-analyzed keyframe images of the videos and outputted keywords and classification information. It finally outputted descriptions of image content and exported them in Excel format; it also utilized the \"text recognition\" API interface to recognize subtitles and other text in keyframe images and exported them in TXT format. Finally, the entire process was integrated into one program, enabling the analysis of video screenshots, extraction of text descriptions, and recognition of text content to be performed at once after inputting the video file.\u003c/p\u003e\u003cp\u003eThe data obtained from the capture were identified, resulting in 6638 keywords belonging to 199 different categories from the recognition of 1327 keyframes, and were exported for further processing.\u003c/p\u003e\u003cp\u003eDue to the large volume of data and the presence of significant interference, further manual processing of the data is required. According to the recognition results, keyword entries are generally more detailed and specific, while category entries are more generalized. For example, when the main subject of an image is a plant, the keyword entry will provide the specific name of the plant, while the category entry will only provide \"Plants - Forests/Trees/Shrubs/Flowers.\" In this study, the demand for precise identification of image content is relatively low, while the demand for batch classification of image content is higher. Comparatively, category entries are more aligned with the requirements of this study. Therefore, this study will primarily analyze the recognition results based on category entries.\u003c/p\u003e\u003cp\u003eThis study carried out manual corrections to the data in the following aspects:1. Removal of Irrelevant Categories: Categories clearly unrelated to the research focus on tourist behavior were excluded. Some of these categories appeared frequently but had no analytical relevance, thereby potentially distorting the results. For example, categories such as \"Products\u0026ndash;Wearables,\" \"Products\u0026ndash;Clothing,\" and \"Products\u0026ndash;Eyewear\" appeared over a hundred times, mainly due to the presence of people in the videos. These entries overlapped significantly with the \"People\u0026ndash;Close-Up\" category and were therefore removed to avoid redundancy and bias. 2.Merging Overly Detailed Categories: Some categories were excessively subdivided. These were merged into broader, more meaningful groups based on their suffixes. For instance, \u0026ldquo;Plants\u0026ndash;Leafy Plants\u0026rdquo; and \u0026ldquo;Plants\u0026ndash;Ericaceae\u0026rdquo; were categorized under the more general labels of \u0026ldquo;Plants\u0026ndash;Flowers\u0026rdquo; or \u0026ldquo;Plants\u0026ndash;Trees\u0026rdquo; to streamline analysis. 3.Correction of Low-Accuracy Classifications: Categories with low image recognition accuracy were manually revised and reclassified. For example, \"Natural Scenery\u0026ndash;Ocean\" was reassigned to \"Natural Scenery\u0026ndash;Lake,\" and \"Non-Natural Image\u0026ndash;Design Rendering\" was corrected to \"Non-Natural Image\u0026ndash;Map.\" These adjustments helped improve the consistency and accuracy of the dataset used in this study.\u003c/p\u003e\u003cp\u003eBased on the location information annotated in the videos and the location information obtained from \"landmark recognition,\" one to two nodes were selected as representatives for each section of the canal. Ultimately, nine nodes were selected, including the Grand Canal Forest Park, Grand Canal Forest Park Viewing Platform, Canal Olympic Park, Canal Cultural Square, Three Temples and One Pagoda Scenic Area, Canal Ecological Park, Canal Sightseeing Wharf, Xihaizi Park, and Wenyu River Forest Park. This study believes that there is a certain correlation between the number of video releases and the number of visitors. The quantity of short videos related to a particular node can to some extent reflect the level of attention from visitors to that node. The number of videos associated with each node was counted, with each node individually counted for videos that included multiple nodes.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e"},{"header":"3 Research Findings","content":"\u003cdiv id=\"Sec9\" class=\"Section2\"\u003e\u003ch2\u003e3.1 temporal characteristics of tourist behaviour\u003c/h2\u003e\u003cp\u003eThis study analyzes the temporal characteristics of short videos posted on short video platforms. The time-related metadata of the videos is categorized by year, month, and season. In addition, videos containing night scenes are specifically identified and tagged. By locating keyframes labeled as \u0026ldquo;Architecture \u0026ndash; Night View,\u0026rdquo; the corresponding source videos are traced. From these, ten representative night tourism videos are selected for further analysis using the data processing methods described earlier. Relevant keywords, categories, and captions associated with night tourism are then extracted and examined.\u003c/p\u003e\u003cdiv id=\"Sec10\" class=\"Section3\"\u003e\u003ch2\u003e3.1.1 Interannual Variation Characteristics\u003c/h2\u003e\u003cp\u003eBy analyzing the publication dates of sample short videos, the study attempts to reveal the interannual variation in tourist activity. Using year as the statistical unit, the number of video samples collected each year is shown in Fig.\u0026nbsp;10. The data indicate that the number of videos has increased year by year; however, this trend is primarily influenced by the content distribution mechanisms of short video platforms. Platforms like Douyin tend to push newer videos more prominently to users. Additionally, due to limitations in the platform's display algorithms and the large volume of data, very few videos posted before 2020 were retrievable through chronological searches. This results in a noticeable increase in video quantity after 2020. Given the limited time span covered, no clear interannual trends can be established, and the year-to-year data offer limited reference value.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec11\" class=\"Section3\"\u003e\u003ch2\u003e3.1.2 Intermonth Variation Characteristics\u003c/h2\u003e\u003cp\u003eFrom a seasonal perspective, tourist activity in the Tongzhou section of the Grand Canal is influenced by climatic conditions. The site lies within a temperate monsoon climate zone, characterized by hot, rainy summers and cold, dry winters, resulting in significant seasonality in tourism activity.\u003c/p\u003e\u003cp\u003eAs shown in Fig.\u0026nbsp;11, April, May, August, and October see the highest number of video postings, coinciding with major public holidays such as Qingming Festival, Labor Day, summer vacation, and National Day. There is a clear \"holiday effect\" on tourism activity. The highest number of videos is published in summer (June\u0026ndash;August), accounting for approximately 33% of the annual total. Winter (December\u0026ndash;February) has the lowest count, at around 17%, while spring and autumn show roughly equal levels. This indicates that tourist visits to the Grand Canal World Cultural Heritage site in Tongzhou are concentrated mainly in summer, followed by spring and autumn, with the lowest activity in winter.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eOn one hand, the summer influx of tourists from outside Beijing creates a spillover effect that increases visitor numbers at the Tongzhou section. While the site is less attractive to non-local tourists compared to traditional hotspots like the Forbidden City, the Great Wall, or the Summer Palace\u0026mdash;and captions rarely mention keywords related to tourists from outside Beijing\u0026mdash;some video texts indicate that this influx impacts the leisure spaces of local residents, prompting them to seek less-crowded destinations like the Grand Canal site for recreation. As tourist numbers increase across Beijing, the Tongzhou section likewise sees a rise in visitation.\u003c/p\u003e\u003cp\u003eOn the other hand, local residents\u0026rsquo; demand for nearby outings also shifts seasonally. Due to climate comfort, urban dwellers tend to engage in nearby leisure travel during the more pleasant spring, summer, and autumn seasons. Additionally, available free time significantly influences travel behavior: summer vacation is a peak season for student and family travel, while holidays like May Day, National Day, and Qingming Festival mark major peaks in local travel. This confirms that both climate and leisure time significantly impact visitor behavior at the Grand Canal heritage site.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec12\" class=\"Section3\"\u003e\u003ch2\u003e3.1.3 Night Tourism Behavior\u003c/h2\u003e\u003cp\u003eNight tourism behavior is also commonly observed in videos related to the Tongzhou section of the Grand Canal. Video recognition results show that the category \u0026ldquo;Architecture \u0026ndash; Night Scenes\u0026rdquo; appeared 197 times, ranking 9th among all identified categories\u0026mdash;only behind categories such as \u0026ldquo;Portraits,\u0026rdquo; \u0026ldquo;Lakes and Rivers,\u0026rdquo; \u0026ldquo;Trees,\u0026rdquo; and \u0026ldquo;Modern/Traditional Architecture.\u0026rdquo; As a time-sensitive indicator of tourist activity, the frequent appearance of \u0026ldquo;night scenes\u0026rdquo; strongly suggests the popularity of nighttime tourism at this site.\u003c/p\u003e\u003cp\u003eFurther analysis of the night-themed videos reveals several frequently occurring categories, as illustrated in Fig.\u0026nbsp;13. Activities such as \u0026ldquo;Sports,\u0026rdquo; \u0026ldquo;Artistic Performances,\u0026rdquo; and \u0026ldquo;Ceremonial Events\u0026rdquo; ranked 9th, 11th, and 12th respectively under the \u0026ldquo;Human Activity\u0026rdquo; category, while the appearance of \u0026ldquo;Watercraft \u0026ndash; Boats\u0026rdquo; was also notably high. The remaining high-frequency terms largely align with those in the overall dataset.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eTwo key factors explain the prominence of night tourism at this location. First, parts of the Tongzhou section serve as evening leisure spaces for residents, hosting activities like square dancing, night jogging, and casual walks. Second, the site has benefited from promotional efforts such as the Beijing (International) Grand Canal Cultural Festival, which has played a significant role in encouraging nighttime visits. According to the video content, the festival features various events, including canal boat tours, immersive performances, archery and ceremonial sport experiences, intangible cultural heritage exhibitions, and academic forums. Among the 10 sampled night tourism videos, 6 were directly related to the festival\u0026mdash;suggesting a strong correlation between festival events and nighttime visitor activity.\u003c/p\u003e\u003cp\u003eIn summary, interannual variation in tourist activity at the Grand Canal site in Tongzhou is difficult to assess due to limitations in the availability of earlier video data; thus, no reliable trend can be established. In contrast, intermonth variation shows clear seasonality, with video posting peaks aligning with major holidays and Beijing\u0026rsquo;s peak tourist seasons. These trends are driven both by the seasonal influx of non-local tourists and the changing recreational needs of local residents. Regarding night tourism, visitor engagement is notably high, particularly in cultural, ceremonial, and sports-related activities, influenced by both everyday leisure habits and organized festival events.\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e\u003ch2\u003e3.2 Determination of Scene Type Indicators\u003c/h2\u003e\u003cp\u003eTourist scenes refer to the comprehensive tangible and intangible environmental elements that tourists perceive at tourist destinations. Before analyzing the characteristics of visitors' scene preferences, it is necessary to determine the dimensions of scenes to which the elements obtained by machine recognition of video keyframes belong. In this study, we refer to the classification research on dimensions of tourist experience scenes by Zhang Hui et al.\u003csup\u003e[\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e]\u003c/sup\u003e, and categorize tourist scenes into historical and cultural experiences, natural experiences, consumption experiences, sports experiences, urban views, festival activities, etc. According to this classification method, the \"root\" keywords obtained by machine recognition of video keyframes are reclassified. When classifying, irrelevant content related to landscape types is removed, and some content that is difficult to classify is manually processed. If multiple content units appear in one keyframe, they are counted separately.\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\u003escene type indicator system\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"2\"\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\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eScene type\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eCorresponding content of machine recognition\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eHistorical and cultural experience scenes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eConstructions-traditional construction/cultural heritage\u0026hellip;;\u003c/p\u003e\u003cp\u003egoods- craft sculpture/painting\u0026hellip;\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eConsumption experience scenes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eGoods-food/toy/agricultural material/ sporting goods\u0026hellip;;\u003c/p\u003e\u003cp\u003econstructions-store and mall\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNatural experience scenes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eNatural landscapes-river/lake/waterfall/spring\u0026hellip;;\u003c/p\u003e\u003cp\u003eanimals-fish/bird/dog/beetle\u0026hellip;;\u003c/p\u003e\u003cp\u003eplants-tree/shrub/flower/grass\u0026hellip;;\u003c/p\u003e\u003cp\u003econstructions-garden/natural park\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSports experience scenes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eHuman activities-sport activities/agricultural production\u0026hellip;;\u003c/p\u003e\u003cp\u003etransportation vehicles-bicycle/ships\u0026hellip;\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eUrban landscape scenes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eConstructions-modern constructions/architectural night scenery/landscape sketch\u0026hellip;\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eFestive events scenes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eHuman activities- Literary and artistic activities/ceremonial activities\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec14\" class=\"Section2\"\u003e\u003ch2\u003e3.3 Overall Analysis of Scene Preferences\u003c/h2\u003e\u003cp\u003eThis study completed the classification of scene categories contained in video keyframes through a combination of machine recognition and manual classification. The table below shows the number of scene types contained in keyframes.\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\u003eNumber of each scene type\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"2\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eScene type\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003enumber\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNatural experience scenes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e1735\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eHistorical and cultural experience scenes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e854\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eUrban landscape scenes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e804\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eConsumption experience scenes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e269\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSports experience scenes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e108\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eFestive events scenes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e89\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\u003eAmong all scene classifications, the \"natural experience scenes\" category accounts for the highest proportion, approximately 36% of the total. This category includes elements such as natural parks, rivers, vegetation, and weather. Upon further segmentation of this category, it is evident that vegetation elements dominate, including grasslands and trees along the canal, as well as most scenes in the Grand Canal Forest Park. Visitors often focus on vegetation when experiencing nature. The next most prominent element is rivers. As one of the core elements of the Grand Canal World Cultural Heritage, rivers play an important role in natural experience scenes. Even if visitors do not consider rivers the primary scene element, they still pay considerable attention to them while touring along the Grand Canal. While animals, weather, and climate are also important parts of natural experience scenes, their significance is relatively low in the Tongzhou section of the Grand Canal.\u003c/p\u003e\u003cp\u003e\"Historical and cultural experience scenes\" appeared 854 times, with scene elements focusing on traditional architecture, handicraft sculptures, and paintings, among other material cultural heritage. The Tongzhou section of the Grand Canal boasts rich material and intangible cultural heritage. It is a UNESCO World Heritage site with abundant historical and cultural relics and significant potential for cultural activities. However, compared to natural experience scenes, historical and cultural experience scenes receive less attention, which does not match the rich heritage resources of the Tongzhou section of the Grand Canal. Many excellent heritage resources have not been fully developed and utilized, and the creation of historical and cultural experience scenes is still immature. Therefore, in future Tourist Scene creation, the cultural characteristics of national cultural parks should be fully leveraged, and the cultural resources along the canal should be effectively utilized to create historical and cultural experience scenes that can attract visitors.\u003c/p\u003e\u003cp\u003e\"Urban landscape scenes\" encompass surrounding buildings, bridges, park facilities, and more. Upon further segmentation of the elements in this category, it is evident that modern architecture dominates, including buildings along the Grand Canal, bridges spanning the canal, sports venues, etc. Following closely are pathways and urban roads, which constitute basic infrastructure. During the process of shooting short videos, visitors often find it difficult to avoid including surrounding buildings in their shots. However, this also reflects the significant role of surrounding buildings in shaping the scene of the Grand Canal: due to the vague boundaries of the Grand Canal, visitors not only focus on the scenery within the park but also inevitably pay attention to the landscape along a certain distance of the canal.\u003c/p\u003e\u003cp\u003e\"Consumption experience scenes,\" \"sports experience scenes,\" and \"festive events scenes\" appear relatively infrequently. The connection between consumption experience scenes and festive events scenes is close, but both are currently receiving low attention from visitors in the Tongzhou section of the Grand Canal. As for sports experience scenes, upon closer examination of specific videos, it is evident that due to the limited participation of out-of-town visitors in sports activities and the lack of prominent sports features in the Tongzhou section of the Grand Canal, only a few sports events attract special attention from out-of-town visitors. Therefore, most of the participants in sports activities are local residents, and the intensity of sports activities they engage in is relatively low, usually involving activities such as square dancing, cycling, walking, and jogging. These sports activities typically occur in everyday scenarios, and the desire of local residents to share these daily life activities is far less than that of out-of-town visitors to share their travel experiences, resulting in fewer such contents in short videos. In fact, in the process of creating tourist scenes, although community residents receive less attention on media platforms, their needs and interests must not be ignored.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec15\" class=\"Section2\"\u003e\u003ch2\u003e3.4 Analysis of Scene Preferences at Different Nodes\u003c/h2\u003e\u003cp\u003eThe collected short videos were divided according to nodes, with each video related to a specific node marked. Then, utilizing the data processing method described in the last chapter, the elements depicted in the videos were categorized into different scenes, thus obtaining the scene preference characteristics of different nodes. By categorizing the elements contained in the short videos corresponding to nodes such as the Grand Canal Forest Park, the Three Temples and One Pagoda, the Canal Olympic Park, and the Canal Cultural Square into the aforementioned scene types, the following results were obtained:\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\u003eNumber of each scene type in the Grand Canal Forest Park\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"2\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eScene type\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003enumber\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNatural experience scenes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e1328\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eUrban landscape scenes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e397\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eConsumption experience scenes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e96\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eHistorical and cultural experience scenes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e79\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSports experience scenes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e64\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eFestive events scenes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e34\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\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\u003eNumber of each scene type in the Three Temples and One Pagoda\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"2\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eScene type\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003enumber\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eHistorical and cultural experience scenes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e612\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNatural experience scenes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e257\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eUrban landscape scenes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e165\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eConsumption experience scenes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e54\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eFestive events scenes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e43\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSports experience scenes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e21\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\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\u003eNumber of each scene type in the Canal Olympic Park\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"2\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eScene type\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003enumber\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSports experience scenes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e43\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eUrban landscape scenes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e41\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNatural experience scenes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e21\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eConsumption experience scenes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e13\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eHistorical and cultural experience scenes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e6\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eFestive events scenes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e2\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\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\u003eNumber of each scene type in the Canal Cultural Square\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"2\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eScene type\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003enumber\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eUrban landscape scenes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e78\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eFestive events scenes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e57\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eConsumption experience scenes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e44\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eHistorical and cultural experience scenes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e42\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSports experience scenes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e19\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNatural experience scenes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e14\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\u003eIt can be observed that, due to the dense presence of natural elements in the Grand Canal Forest Park, natural experiential scenes receive the most attention at this node. This node stands out with a higher proportion of natural scene preferences compared to the overall Tongzhou section. The park features extensive green spaces, a variety of plant species, and water bodies that together form a serene and immersive environment, offering visitors a strong sense of natural immersion and relaxation. This appeal is particularly evident among urban residents seeking respite from the pressures of city life. In addition, the Forest Park Observation Deck serves as a strategic viewpoint that offers panoramic views of the surrounding urban skyline and iconic bridge nightscapes. As a result, interest in urban landscape scenes at this node is also notably high, as visitors are drawn to the contrast between natural tranquility and urban vitality. However, other types of scenes\u0026mdash;particularly historical and cultural experiential scenes\u0026mdash;receive relatively limited attention at this site. This is primarily because the area lacks substantial historical and cultural heritage elements, resulting in comparatively lower visitor engagement with such content.\u003c/p\u003e\u003cp\u003eIn contrast, the Three Temples and One Pagoda node demonstrates a pronounced increase in visitors\u0026rsquo; preference for historical and cultural experiential scenes. This trend can be attributed to the concentration of well-preserved heritage assets in the area, such as ancient temples, religious structures, and traditional architectural elements that embody the region's historical significance. This node serves as a cultural core of the Tongzhou section, offering rich interpretive potential and educational value. Visitors to this node are more likely to engage with the site's cultural narratives, take part in heritage tours, and capture content reflecting traditional aesthetics. As interest in cultural experiences rises at this location, it naturally leads to a decline in attention to other scene categories such as natural, sports, or leisure-based scenes, indicating a focused rather than diversified visitor engagement pattern.\u003c/p\u003e\u003cp\u003eAt the Canal Olympic Park node, a different pattern emerges. This area is characterized by extensive sports infrastructure and the availability of water-based recreational activities such as kayaking and rowing. Consequently, visitor preferences here skew toward sports experiential scenes, as tourists and local residents alike participate in or observe athletic events and outdoor fitness activities. In addition, the park's open layout and integration with waterfront spaces provide scenic views of the canal and surrounding urban developments, leading to a heightened interest in urban landscape scenes as well. Nevertheless, other types of scenes\u0026mdash;particularly those tied to culture or history\u0026mdash;are underrepresented at this node, reflecting its modern and activity-oriented positioning within the broader heritage landscape.\u003c/p\u003e\u003cp\u003eThe Canal Cultural Square is another key node that displays a distinct pattern of visitor preference. Architecturally, the square is marked by tall, colorful archways, artistic landscape sculptures, and intricately carved granite pavement, all of which contribute to its visual identity and photogenic appeal. These features make the square a favored location for capturing urban landscape content, particularly during evening hours when the bridge lighting and city skyline create a dramatic backdrop. Beyond its architectural appeal, the square also functions as a major site for public events, community celebrations, and informal recreational activities such as square dancing, musical performances, and seasonal gatherings. These events contribute to a higher frequency of festival and event-related scenes in visitor-generated content. As such, the Canal Cultural Square reflects a dual preference among visitors\u0026mdash;for urban aesthetics and socially driven experiences\u0026mdash;making it one of the more dynamic and multifunctional spaces along the Tongzhou section.\u003c/p\u003e\u003cp\u003eFor other nodes, due to the limited mentions in the short videos and insufficient data volume, there is no clear preference relationship between scenes. It is difficult to draw effective conclusions based on existing data. In future research, data retrieval based on the names of other nodes can be conducted to enrich the results of this study.\u003c/p\u003e\u003c/div\u003e"},{"header":"4 Conclusion and Discussion","content":"\u003cdiv id=\"Sec17\" class=\"Section2\"\u003e\u003ch2\u003e4.1 conclusion\u003c/h2\u003e\u003cp\u003eThrough machine processing of visitors' related videos posted on the Douyin short video platform, an analysis of visitors' spatiotemporal behavior characteristics and landscape preferences leads to the following conclusions:\u003c/p\u003e\u003cp\u003eFrom the perspective of temporal distribution, annual variations in tourist volume are not clearly observable due to limitations in short video platform data, making it difficult to identify consistent interannual trends. However, on a seasonal scale, tourist activity shows pronounced seasonality, largely influenced by climate conditions and the availability of leisure time. Nighttime tourism is also notably popular in the Tongzhou section of the Grand Canal World Cultural Heritage Site. This is driven both by the recreational habits of local residents and by the promotional impact of festivals and events organized by park authorities.\u003c/p\u003e\u003cp\u003eIn terms of visitors' scene preference characteristics, visitors' preferences for various types of scenes along the canal are as follows: natural experience scenes\u0026thinsp;\u0026gt;\u0026thinsp;historical and cultural experience scenes\u0026thinsp;\u0026gt;\u0026thinsp;urban landscape experience\u0026thinsp;\u0026gt;\u0026thinsp;consumer experience scenes\u0026thinsp;\u0026gt;\u0026thinsp;sports experience scenes\u0026thinsp;\u0026gt;\u0026thinsp;festival event scenes. Further analysis of specific elements reveals that the attention to historical and cultural experience scenes does not match the rich cultural heritage resources of the Tongzhou section of the Grand Canal. Many excellent heritage resources have not been fully developed and utilized. Regarding the shaping of sports experience scenes, although visitors do not show a high preference for sports experience scenes, in reality, most of the sports activities conducted here are by community residents. Their needs should not be overlooked when shaping the scenes.\u003c/p\u003e\u003cp\u003eVisitors' scene preferences at different nodes vary according to the characteristics of the nodes themselves. At the Grand Canal Forest Park node, visitors show a strong preference for natural experiences; at the Three Temples and One Pagoda node, visitors have a higher preference for historical and cultural experience scenes; at the Canal Olympic Park node, visitors prefer urban landscape and sports experience scenes; at the Canal Cultural Square node, visitors prefer urban landscape and festival activity scenes. Meanwhile, due to the vague boundaries of the Grand Canal National Cultural Park, landscapes and cultural activities outside the scenic area have a significant impact on visitors' perceptions.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec18\" class=\"Section2\"\u003e\u003ch2\u003e4.2 discussion\u003c/h2\u003e\u003cp\u003eTo promote further development of the Tongzhou section of the Grand Canal, this study proposes the following recommendations for park managers and developers based on the research conclusions:\u003c/p\u003e\u003cp\u003e1.Focus on visitors' scene preferences. Addressing the low attention to historical and cultural experience scenes, efforts should be made to enhance the development and utilization of the rich historical and cultural heritage resources along the Tongzhou section of the Grand Canal. This can be achieved by integrating heritage resources with natural experiences, festival activities, and other scenes. Since visitors show a relatively high preference for urban landscape experiences, it is recommended to strengthen the integration of urban landscape elements throughout the entire Tongzhou section, including planning for urban landscapes and cultural activities beyond the boundaries of the scenic area, to enhance visitors' overall perception and experience. Tailor Tourist Scenes to the specific preferences of visitors and residents at different nodes, creating unique and attractive Tourist Scenes.\u003c/p\u003e\u003cp\u003e2. Pay attention to the actual needs of community residents. When developing scenes for cultural heritage, not only should the construction of scenic spots and supporting facilities along the route be considered, but also the cultural and natural environments within a certain range should be taken into account. In the planning and development process of the heritage, adopt a community-participation planning approach to understand the needs and expectations of local residents. Collect their opinions through forums, questionnaires, etc., to ensure that the park's design and services truly meet the practical needs of local residents. Understand the community residents' needs for sports and leisure, and consider adding appropriate sports facilities such as basketball courts, fitness areas, etc. Organize cultural activities such as community performances, handicraft markets, etc., to attract active participation from community residents. These activities not only enhance the social atmosphere of the park but also strengthen its role as a center for community cultural exchange.\u003c/p\u003e\u003c/div\u003e"},{"header":"Declarations","content":"\u003cp\u003eData availability\u003c/p\u003e\n\u003cp\u003eThe author confirms that all data generated or analysed during this study are included in this published article\u0026nbsp;and database. The datasets analysed during the current study are available in the Dataverse repository.\u0026nbsp;Furthermore, primary and secondary sources and data supporting the findings of this study were all publicly available at the time of submission.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;Ethical Approval\u003c/p\u003e\n\u003cp\u003eThis study was conducted in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki and its later amendments. As the research involved only the analysis of publicly available user-generated content (UGC) on short video platforms, without any direct involvement or interaction with human subjects, biological material, or personal identifiers, formal ethical approval was not required.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;Informed Consent\u003c/p\u003e\n\u003cp\u003eThis study did not involve any direct interaction with human participants, and all data were collected from publicly available short video content. No identifiable personal data, private information, or images requiring consent were collected, analyzed, or published. Therefore, informed consent was not applicable. The authors affirm that all procedures respected the rights to privacy and anonymity of individuals, and any potentially sensitive content was excluded from analysis.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eBecker R A, Caceres R, Hanson K, et al. 2011, A Tale of One City: Using Cellular Network Data for Urban Planning[J]. IEEE pervasive computing, 10(4):18-26.\u003c/li\u003e\n\u003cli\u003eBirkin M, Malleson N. 2011, Microscopic simulations of complex metropolitan dynamics[J]. the Complex City workshop, Unpublished.\u003c/li\u003e\n\u003cli\u003eDeng N, Qu L. 2022, Comparison of Destination Images Based on Video Analysis through Machine Learning\u0026mdash;\u0026mdash;A Case Study on YouTube Videos of Beijing [J]. Tourism Tribune, 37(08):70-85.\u003c/li\u003e\n\u003cli\u003eDai J, Huang X, Sun J, et al. 2023, Research on the Formation Mechanism of Tourist Cultural Identity in National Cultural Park\u0026mdash;\u0026mdash;Taking the Iconic Natural Landscapes of the Yellow River National Cultural Park as an Example [J]. Tourism Tribune, 38(01):31-41.\u003c/li\u003e\n\u003cli\u003eGirardin F, Calabrese F, Fiore F D, et al. 2008, Digital Footprinting: Uncovering Tourists with User-Generated Content[J]. IEEE pervasive computing, 7(4):36-43.\u003c/li\u003e\n\u003cli\u003eGirardin F, Fiore F D, Ratti C, et al. 2008, Leveraging explicitly disclosed location information to understand tourist dynamics: a case study[J]. Journal of location based services, 2(1):41-56.\u003c/li\u003e\n\u003cli\u003eGong L. 2019, The Grand Canal: From Cultural Landscape Heritage to National Cultural Park [J]. The Mass Weekly, 2019(24):17-18. \u003c/li\u003e\n\u003cli\u003eGong D. 2021, The Origin and Characteristics of the Concept of National Cultural Park [J]. Chinese Landscape Architecture, 37(06):38-42.\u003c/li\u003e\n\u003cli\u003eHan Z, Ren H. 2022, Significance,Principle and Strategy of National Cultural Park Trail Construction\u0026mdash;\u0026mdash;Take the Great Wall National Cultural Park as an Example [J]. Journal of Northwest Normal University(Social Sciences), 59(05):23-30.\u003c/li\u003e\n\u003cli\u003eLi M, Ma Q, Zhao X. 2013, A Review of Researches on Servicescape and Future Prospects [J]. Foreign Economics \u0026amp; Management, 35(04):62-70. \u003c/li\u003e\n\u003cli\u003eLi X. 2017, Research of Short Video Communication Strategy and effect\u0026mdash;\u0026mdash;YiTiao TV as an example[D]. Anhui University.\u003c/li\u003e\n\u003cli\u003eLi F, Zou C, 2021, National Culture Park: Logic, Origins and Implications [J]. Tourism Tribune, 36(01):14-26.\u003c/li\u003e\n\u003cli\u003eLi S. 2021, Functions, Values, and Implementation Approaches of National Cultural Parks [J]. China Economic \u0026amp; Trade Herald, 2021(03):152-155.\u003c/li\u003e\n\u003cli\u003eLiu L, Wang Q. 2021, Reflections on the Construction Path of the Long March National Cultural Park: A Case Study of the Sichuan Section of the Long March National Cultural Park [J]. Mao Zedong Thought Study, 38(01):108-113.\u003c/li\u003e\n\u003cli\u003eNegroponte N. Being Digital[M]. 1996.\u003c/li\u003e\n\u003cli\u003eNerute K S. 2012, Digital Footprints in the Context of Professional Ethics[J]. Informatics in education, 11(1):65-79.\u003c/li\u003e\n\u003cli\u003ePan Z, Liang B. 2015, Tourist Perception of Tourism Destination Image Analysis Based on Digital Footprints:A Case Study of World Natural Heritage Tianshan Mountain [J]. Journal of Beijing International Studies University, 37(05):40-48.\u003c/li\u003e\n\u003cli\u003eShan J, 2006, Preliminary Discussion on the Protection of Large Linear Cultural Heritage: Breakthrough and Pressure [J]. Cultural Relics in Southern China, 2006(03):2-5.\u003c/li\u003e\n\u003cli\u003eSun H. 2021, Preliminary Study on National Cultural Parks: Concepts, Types, Characteristics, and Construction [J]. China Cultural Heritage, 2021(05):4-14. \u003c/li\u003e\n\u003cli\u003eSong Y, Zhang Y. 2022, On the design and construction of environmental aesthetic experience in Hangzhou Section of Grand Canal National Cultural Park [J]. Journal of Zhejiang University of Technology(Social Sciences), 21(03):308-313.\u003c/li\u003e\n\u003cli\u003eThatcher J. 2014, Living on Fumes: Digital Footprints, Data Fumes, and the Limitations of Spatial Big Data[J]. International Journal of Communication.\u003c/li\u003e\n\u003cli\u003eVaccari A, Liu L, Biderman A, 2009, et al. A holistic framework for the study of urban traces and the profiling of urban processes and dynamics, [C]. IEEE, 2009.\u003c/li\u003e\n\u003cli\u003eWeaver S D, Gahegan M. 2007, Constructing, Visualizing, and Analyzing A Digital Footprint[J]. Geographical review, 97(3):324-350. \u003c/li\u003e\n\u003cli\u003eWober K. 2007, Similarities in Information Search of City Break Travelers - A Web Usage Mining Exercise[J]. Institute for Tourism and Leisure Studies, DBLP.\u003c/li\u003e\n\u003cli\u003eWang X. 2014, The Research Review on Tourism Preference at Home and Abroad [J]. Modern Urban Research, 2014(01):110-115. \u003c/li\u003e\n\u003cli\u003eWang K, 2021, Theoretical Exploration and Practical Review of National Cultural Park [J]. enterprise.economy, 40(04):5-12.\u003c/li\u003e\n\u003cli\u003eWang X, Bai L. 2021, Logical Following and Path Exploration of the Construction of Grand Canal National Cultural Park: Based on the Dual Theoretical Perspectives of Cultural Memory and Spatial Production [J]. Zhejiang Social Sciences, 2021(10):72-80.\u003c/li\u003e\n\u003cli\u003eWu D, Liu H, Wang B. 2021, Realistic Misunderstanding and Improved Suggestions during the Construction of National Cultural Park [J]. Research on Development, 2021(03):1-7.\u003c/li\u003e\n\u003cli\u003eYu C, Lu L. 2008, A Comment on Overseas Studies of Tourists\u0026apos; Preferences for Tourism in Recent Years [J]. Journal of Anhui Normal University(Natural Science), 31(06):590-595.\u003c/li\u003e\n\u003cli\u003eZhang W, Tao Z, Qin L, et al. 2016, Space Response of Tourists\u0026apos; Digital Footprint in Suzhou Ancient Garden Based on Online Travel Note [J]. Resource Development \u0026amp; Market, 32(07):886-891. \u003c/li\u003e\n\u003cli\u003eZhang H, Xu H. 2019, A structural model of liminal experience in tourism[J]. Tourism Management, 71:84-98.\u003c/li\u003e\n\u003cli\u003eZou C, Chang M, Lai M, 2019, International Experience and Reference on the Management Model of National Cultural Parks [N]. China Tourism News, 11(05), 003\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Digital Footprints, the Grand Canal, World Cultural Heritage Site, Tourism Short Videos","lastPublishedDoi":"10.21203/rs.3.rs-6619412/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6619412/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"World Cultural Heritage sites represent the historical value and spatial significance of human civilization, balancing cultural preservation with ecological protection. They serve not only as vessels of cultural continuity but also as key resources for tourism development. To enhance the understanding of tourist scene preferences and provide targeted strategies for future planning, this study focuses on the Tongzhou section of the Grand Canal—designated as a UNESCO World Cultural Heritage site—as a case study. By collecting content from mainstream short video platforms (primarily Douyin) and developing custom analysis tools using Python and Baidu AI Cloud APIs, the study extracts and interprets keyframes and associated textual data from videos to identify the types of scenes that capture visitors' attention, as well as variations across different heritage nodes. Results show that natural experiential scenes receive the highest level of visitor interest, while historical and cultural scenes—despite the area's rich heritage—remain underappreciated. Moreover, tourist scene preferences vary significantly between nodes, influenced by the specific characteristics of each site. Based on these findings, the study offers practical recommendations for site managers to better align interpretive and experiential offerings with visitor interests, while also addressing the needs of local residents.","manuscriptTitle":"Analyzing Tourists’ Scene Preferences through Digital Footprints in World Heritage Site: A Case Study of the Grand Canal’s Tongzhou Section","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-08-05 06:50:18","doi":"10.21203/rs.3.rs-6619412/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"432e3c40-05ff-4b7a-b96b-4db5d6b052af","owner":[],"postedDate":"August 5th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":52527455,"name":"Business and commerce/Information systems and information technology"},{"id":52527456,"name":"Social science/Cultural and media studies"}],"tags":[],"updatedAt":"2025-10-02T15:53:41+00:00","versionOfRecord":[],"versionCreatedAt":"2025-08-05 06:50:18","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6619412","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6619412","identity":"rs-6619412","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}