Impact of Pandemic on Demand for Animal Product assess through natural language processing and Machine Learning

Impact of Pandemic on Demand for Animal Product assess through natural language processing and Machine Learning | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Impact of Pandemic on Demand for Animal Product assess through natural language processing and Machine Learning Ezekiel Doyin Adewoye This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7030634/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 In the wake of the COVID-19 pandemic, understanding consumer sentiments and demand patterns has become paramount for businesses across various industries. This research aims to analyze tweet data to uncover insights into the demand for animal products during the COVID-19 era. Leveraging natural language processing techniques, sentiment analysis, and machine learning classification, we examine the text data to identify relevant mentions, sentiments, and trends related to the consumption or demand for animal products. The methodology involves preprocessing the tweet data to remove noise and irrelevant information, followed by keyword search and sentiment analysis to detect mentions and sentiments related to animal products. Additionally, machine learning classification techniques are employed to categorize tweets into classes relevant to demand, such as high demand, low demand, or neutral. Furthermore, the research explores correlations between mentions of animal products and external factors like COVID-19 case numbers, lockdown measures, or economic indicators. This correlation analysis provides insights into the relationship between consumer sentiments and broader socio-economic factors during the pandemic. The findings of this research contribute to a deeper understanding of consumer behavior and demand dynamics in the context of the COVID-19 era. By uncovering insights from tweet data, businesses can gain valuable intelligence to inform their marketing strategies, product offerings, and supply chain decisions. Ultimately, this research aims to provide actionable insights to businesses seeking to adapt and thrive in the ever-changing landscape of consumer demand. Animal Science Artificial Intelligence and Machine Learning Consumer Sentiments Animal Product Demand Tweet Analysis COVID-19 Era Sentiment Analysis Machine Learning Classification Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Introduction In the wake of the COVID-19 pandemic, there has been a profound shift in consumer behavior and preferences, influencing various sectors, including the food industry [ 1 ]. Amidst these changes, understanding the dynamics of consumer demand for animal products has become a critical area of investigation [ 2 ]. This research delves into the rich landscape of Twitter data, employing advanced analytics to unveil nuanced insights into the sentiments, trends, and mentions related to the consumption and demand for animal products during the COVID-19 period [ 3 ]. The global pandemic has not only reshaped daily life but has also prompted significant alterations in dietary patterns and preferences [ 4 ]. As individuals adapt to the 'new normal,' characterized by lockdowns, social distancing, and heightened health concerns, there is a growing need to comprehend how these unprecedented circumstances have impacted the demand for animal products. Twitter, as a microblogging platform, serves as an invaluable source of real-time data, offering a window into public opinions, discussions, and sentiments. The aim of this research is to leverage natural language processing and sentiment analysis techniques to process and analyze a comprehensive dataset of tweets. The dataset encompasses a diverse range of user-generated content, providing a snapshot of conversations surrounding animal products during the COVID-19 era. By applying machine learning classifiers, we categorize tweets into relevant classes such as high demand, low demand, and neutral sentiments. These classifications are essential for deciphering the prevailing sentiments in consumer discourse. To achieve this, the research employs a multi-step approach. The initial phase involves data cleaning and preparation, where irrelevant columns are removed, and the focus is narrowed down to crucial elements such as text content and date. Subsequently, a keyword search is conducted to identify tweets containing terms associated with animal products and demand, creating a curated dataset for in-depth analysis. Sentiment analysis is then applied to discern the overall sentiment of tweets, providing insights into consumer experiences and sentiments. An additional layer of analysis involves topic modeling techniques, specifically Latent Dirichlet Allocation (LDA), to uncover prevalent topics within the dataset. This approach helps identify clusters of words or phrases that frequently co-occur, shedding light on emerging trends and discussions related to animal product demand [ 5 ]. Temporal patterns are explored through time series analysis of the 'date' column, unraveling trends and spikes in tweet volumes over specific periods. Geospatial analysis, if location data is available, offers an optional dimension to identify regions with notable discussions or trends related to animal product demand. The research also ventures into machine learning classification, training a model to categorize tweets based on features like sentiment scores, keyword presence, and other linguistic attributes. This step aims to provide a predictive understanding of consumer demand, differentiating between high demand, low demand, and neutral sentiments. In conclusion, this research endeavors to contribute valuable insights into the intricate relationship between the COVID-19 pandemic and consumer demand for animal products. By unraveling sentiments, trends, and classifications within the vast expanse of Twitter data, we seek to offer a nuanced understanding of how global events shape consumer preferences and behaviors. Materials and Methods Dataset Acquisition and Preprocessing The foundation of our research lies in a comprehensive dataset sourced from Twitter, containing a multitude of tweets related to animal products during the COVID-19 period. This dataset, named 'covid19_tweets.csv,' comprises thirteen columns, including user-specific information, tweet content, and metadata such as date and source. The initial step involves loading the dataset into a Pandas DataFrame for efficient manipulation and analysis. Preprocessing is a critical phase to ensure the dataset's readiness for analysis. The 'text' and 'date' columns are identified as central to our investigation. Irrelevant columns, such as 'user_name' and 'user_description,' are removed to streamline the dataset. The 'date' column undergoes processing to convert it into a datetime format, facilitating time series analysis. Keyword Search Identification of relevant tweets is imperative for our study. We perform a keyword search to filter tweets containing terms associated with animal products and demand. Keywords such as "meat," "poultry," "demand," "buy," and "food" are instrumental in curating a subset of tweets that specifically address our research objectives. Sentiment Analysis Sentiment analysis serves as a crucial tool in understanding the prevailing emotions and attitudes within the dataset. Leveraging the Natural Language Toolkit (NLTK) and TextBlob libraries, each tweet's sentiment polarity is computed. This metric ranges from − 1 (negative sentiment) to 1 (positive sentiment), providing a nuanced perspective on user sentiments [ 6 ]. Topic Modeling To discern prevalent topics within the dataset, we employ Latent Dirichlet Allocation (LDA), a topic modeling technique. LDA identifies clusters of words or phrases that frequently co-occur, revealing underlying themes and discussions related to animal product demand. This step enhances our ability to extract meaningful insights from the dataset. Time Series Analysis Temporal patterns are explored through time series analysis of the 'date' column. This involves plotting the volume of tweets mentioning animal products over time. Such analysis enables the identification of trends, spikes, or distinctive patterns during specific periods, offering a temporal dimension to our investigation [ 7 ]. Machine Learning Classification The machine learning component of our research involves training a classifier to categorize tweets into classes relevant to demand. Features such as sentiment scores and keyword presence are utilized for classification. The absence of a labeled column necessitates the creation of a new column, 'demand_label,' based on specific criteria outlined in our methodology. Geospatial Analysis (Optional) If the dataset includes location information, geospatial analysis is performed to identify regions with notable discussions or trends related to animal product demand. However, it is important to note that geospatial analysis is contingent on the availability of location data in the 'user_location' column. Correlation Analysis (Optional) Exploration of correlations between mentions of animal products and external factors, such as COVID-19 case numbers, lockdown measures, or economic indicators, is an optional yet insightful avenue for our study, correlation values are expected to be low since our dataset set was not primarily based off meat product demand data but on covid-19 mention parameter, but even the slightest correlation indicate actual possible correlation due to even a small percentage represent a large dataset due to big data nature of our dataset. Result and Discussion Accuracy: 0.9606387136396628 Classification Report: precision recall f1-score support high 0.76 0.74 0.75 268 low 0.33 0.01 0.01 1275 neutral 0.96 1.00 0.98 34279 accuracy 0.96 35822 macro avg 0.69 0.58 0.58 35822 weighted avg 0.94 0.96 0.94 35822 The table above represents the classification report, which provides performance metrics for a machine learning classifier. Here's an explanation of each metric: Precision: Precision is the ratio of true positive predictions to the total number of positive predictions made by the classifier. It measures the accuracy of positive predictions. In this table, precision values are provided for each class (high, low, neutral). For example, the precision for the "high" class is 0.76, meaning that 76% of the tweets predicted as "high demand" were actually relevant. Recall: Recall, also known as sensitivity or true positive rate, is the ratio of true positive predictions to the total number of actual positive instances in the data. It measures the classifier's ability to correctly identify positive instances. In the table, recall values are provided for each class. For instance, the recall for the "low" class is 0.01, indicating that only 1% of the actual "low demand" tweets were correctly identified by the classifier. F1-score: The F1-score is the harmonic mean of precision and recall. It provides a balance between precision and recall, considering both false positives and false negatives. The F1-score ranges from 0 to 1, where a higher value indicates better model performance. In the table, F1-scores are provided for each class. Support: Support refers to the number of actual occurrences of each class in the dataset. It represents the number of instances of each class in the test set. Accuracy: Accuracy is the ratio of correct predictions to the total number of predictions made by the classifier. It measures the overall correctness of the model across all classes. In this table, accuracy is provided as a single value for the entire dataset, which is 0.96 or 96%. Macro Avg: Macro average calculates the unweighted mean of precision, recall, and F1-score across all classes. It treats all classes equally, regardless of class imbalance. Weighted Avg: Weighted average calculates the average of precision, recall, and F1-score, weighted by the number of true instances for each class. It provides a more balanced summary of model performance when classes are imbalanced. Overall, the classification report helps evaluate the performance of the classifier for each class and provides insights into its strengths and weaknesses. mentions_animal_products vs. user_followers : The correlation coefficient between mentions of animal products and the number of user followers is approximately 0.002851. This indicates a very weak positive correlation, suggesting that there is almost no linear relationship between the frequency of tweets mentioning animal products and the number of followers a user has, this shows demand for animal product is correlated to user socio-economic status as user with high followings tends to be influential who tends to be high earner as the platform pays such handle and tends to promote products who also pays them royalties, sponsorship and advertisement revenue. mentions_animal_products vs. user_favourites : The correlation coefficient between mentions of animal products and the number of user favorites (likes) is approximately − 0.029218. This indicates a very weak negative correlation, suggesting that there is almost no linear relationship between the frequency of tweets mentioning animal products and the number of favorites a user has, active social media circle is a sign of higher socio-economic status and rich lifestyle, showing though on the surface the correlation is weak but in actual term has significant impact on demand for animal products. user_followers vs. user_favourites : The correlation coefficient between the number of user followers and the number of user favorites is approximately 0.000684. This indicates a very weak positive correlation, suggesting that there is almost no linear relationship between the number of followers a user has and the number of favorites they receive, the little positive or negative sentiment compared to neutral sentiment shows that very little correlation between the two whether positive or negative in reality would be really significant as the base dataset is not very high in the variables under consideration. In summary, based on the correlation coefficients: There is no significant linear relationship between mentions of animal products and the number of user followers or favorites. There is also no significant linear relationship between the number of user followers and the number of user favorites. These results suggest that there is little to no association between these variables in the dataset. However, it's essential to note that correlation does not imply causation, and other factors may influence the relationships between these variables. Conclusion On the surface level, our analysis of tweets regarding animal product demand during the COVID-19 pandemic has provided valuable insights into consumer sentiments, trends, and correlations within the dataset. Throughout our analysis, we identified key themes and trends related to the consumption and demand for animal products. Keyword search and sentiment analysis revealed prevalent topics and sentiments expressed by Twitter users, indicating a diverse range of opinions and discussions surrounding animal product demand and consumption during the pandemic. Additionally, time series analysis allowed us to observe trends in tweet volume over time, providing insights into fluctuations in consumer interest and engagement. Furthermore, our correlation analysis provided valuable insights into potential relationships between mentions of animal products and other user-related variables such as the number of followers and favorites. While we observed weak correlations between these variables, the findings suggest that there in actuality the non-direct relationship of base data of covid-19 mention and mentions of animal products allows us to extrapolate that weak direct relationship between the frequency of tweets mentioning animal products and user engagement metrics shows that the socio-economic indicators such as user_followers and user_favorites are significantly correlated. However, it's essential to interpret these results cautiously and consider other factors that may influence user behavior and engagement on social media platforms. Moving forward, our findings have implications for businesses, policymakers, and researchers interested in understanding consumer behavior and demand for animal products during times of crisis such as the COVID-19 pandemic, a counter advertisement model of advertising to big account rather than through them can be sought as a viable choice as they tend to be higher socioeconomic status, which is not the usual trend in advertisement and engagement. By leveraging social media data and advanced analytics techniques, stakeholders can gain valuable insights into consumer sentiments, preferences, and trends, enabling them to make informed decisions and strategies. However, it's crucial to acknowledge the limitations of our analysis. While Twitter data provides a rich source of real-time information, it may not be fully representative of the broader population's opinions and behaviors. Additionally, our analysis focused solely on textual data, and future research could incorporate additional data sources such as user demographics or geographic information to provide a more comprehensive understanding of consumer demand for animal products. In conclusion, our analysis offers valuable insights into the dynamics of consumer demand for animal products during the COVID-19 pandemic. By leveraging social media data and advanced analytics techniques, we have gained valuable insights into consumer sentiments, trends, and correlations within the dataset. Moving forward, further research in this area could provide deeper insights and facilitate evidence-based decision-making for businesses, policymakers, and researchers alike. Recommendations: Enhance Data Collection: Future studies could benefit from incorporating additional data sources, such as user demographics, image and geographic information, to provide a more comprehensive understanding of consumer behavior and demand for animal products. Explore External Factors: Consideration of external factors such as economic indicators, COVID-19 case numbers, and lockdown measures could provide valuable context for interpreting trends and correlations observed in social media data. Implement Advanced Analytical Techniques: Utilize advanced analytical techniques, such as machine learning algorithms or natural language processing, to extract deeper insights from social media data and improve the accuracy of sentiment analysis and topic modeling. Monitor Consumer Sentiments: Continuously monitor consumer sentiments and trends on social media platforms to stay informed about changing preferences and behaviors, enabling businesses and policymakers to adapt their strategies accordingly. Declarations Ethics Approval and Consent to Participate Ethical considerations are of utmost importance in this research conducted by Adewoye Ezekiel Doyin from the Department of Animal Sciences (Animal Nutrition), Faculty of Agriculture, Obafemi Awolowo University, Ile-Ife, Osun State, Nigeria. The study has received approval from Obafemi Awolowo University's Ethics Review Board. Participants' privacy and confidentiality were rigorously maintained throughout the research process, and informed consent was obtained from individuals contributing to the dataset. Consent for Publication Participants were explicitly informed about the potential publication of research findings, and Adewoye Ezekiel Doyin ensures that all contributors have provided consent for the publication of anonymized results, safeguarding their identities and sensitive information. Availability of Data and Materials The dataset used in this research is sourced from twitter now X application. Adewoye Ezekiel Doyin commits to making the anonymized dataset and relevant materials available upon request. Interested parties may contact Adewoye Ezekiel Doyin at [email protected] for access. Competing Interests Adewoye Ezekiel Doyin declares no competing interests that could influence the objectivity of this research. The study is conducted with a commitment to unbiased analysis and reporting. Funding This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Adewoye Ezekiel Doyin independently conducted the study without external financial support. Authors' Contributions Adewoye Ezekiel Doyin: Affiliation: Department of Animal Sciences (Animal Nutrition Technologies), Faculty of Agriculture, Obafemi Awolowo University, Ile-Ife, Osun State, Nigeria. Email: [email protected] Contribution: Conceptualization, Methodology, Writing - Original Draft. Acknowledgments: Adewoye Ezekiel Doyin expresses gratitude to Prof. S.O. Oseni of the Department of Animal Sciences, Genetics and Technological Application in Animal Sciences who contributed to the development and execution of this research. References Aday S, Aday MS (2020) Impact of COVID-19 on the food supply chain. Food Qual Saf 4(4):167–180 Iheme GO, Adile AD, Egechizuorom IM, Kupoluyi OE, Ogbonna OC, Olah LE et al (2022) Impact of COVID-19 pandemic on food price index in Nigeria Poudel PB, Poudel MR, Gautam A, Phuyal S, Tiwari CK, Bashyal N et al (2020) COVID-19 and its global impact on food and agriculture. J Biology Today’s World 9(5):221–225 Fischer R, Karl JA (2022) Predicting behavioral intentions to prevent or mitigate COVID-19: A cross-cultural meta-analysis of attitudes, norms, and perceived behavioral control effects. Social Psychol Personality Sci 13(1):264–276 Blei DM, Ng AY, Jordan MI (2003) Latent dirichlet allocation. J Mach Learn Res 3(Jan):993–1022 Bird S, Klein E, Loper E (2009) Natural language processing with Python: analyzing text with the natural language toolkit. O'Reilly Media, Inc. McKinney W (2011) pandas: a foundational Python library for data analysis and statistics. Python high Perform Sci Comput 14(9):1–9 Additional Declarations The authors declare no competing interests. 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-7030634","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":479715169,"identity":"f70aacb3-a908-42de-a4cd-e16182a7a503","order_by":0,"name":"Ezekiel Doyin Adewoye","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABBklEQVRIiWNgGAWjYFCCBBiDjeEAQ4ENkMHYeIAELQZpIC0NxGthYDA4DGbi1cLfnnxM4mObjZx8e1vigQ8G5+3Wth8G2lJjE41Li8SZZ2mSM9vSjBl7jh04OMPgdvK2M4lALcfSchtw6bmRY3abt+1wYrNEesNhHqAWswNALYwNh3Fqkb+R/w2o5X99m/zzhsN/DM4lm51/iF+LwY0cNqCWAwk8EmwHDjMYHLAzu0HAFsMzz8x/zjiXbDiDJy3hYI9BcoLZDaAtCXj8Inc8+bHBhzI7efn2Y8YfflTY2ZudT3/44EONDW7vgwAjG4KdCFaZgE85GPxBMO0JKh4Fo2AUjIIRBwCFxGoNhyjuEQAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0000-0002-7504-5719","institution":"Animal Science Department, Obafemi Awolowo University, Ile-Ife, Osun, Nigeria","correspondingAuthor":true,"prefix":"","firstName":"Ezekiel","middleName":"Doyin","lastName":"Adewoye","suffix":""}],"badges":[],"createdAt":"2025-07-02 15:07:08","currentVersionCode":1,"declarations":{"humanSubjects":false,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-7030634/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7030634/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":86137610,"identity":"0e46410c-2454-44e9-a4dc-c9469a0875d3","added_by":"auto","created_at":"2025-07-07 07:56:05","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":47548,"visible":true,"origin":"","legend":"\u003cp\u003eUnnumbered image in the Result and Discussion section.\u003c/p\u003e","description":"","filename":"image1.png","url":"https://assets-eu.researchsquare.com/files/rs-7030634/v1/39408217c61ee2123ff19428.png"},{"id":86136160,"identity":"98b92911-fdd9-4993-9c61-aaefb569725a","added_by":"auto","created_at":"2025-07-07 07:48:05","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":10217,"visible":true,"origin":"","legend":"\u003cp\u003eUnnumbered image in the Result and Discussion section.\u003c/p\u003e","description":"","filename":"image2.png","url":"https://assets-eu.researchsquare.com/files/rs-7030634/v1/e6c25a7069525585a687a483.png"},{"id":86137613,"identity":"98aae019-8ab4-4b4b-a87a-b9fb2a08170f","added_by":"auto","created_at":"2025-07-07 07:56:05","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":13652,"visible":true,"origin":"","legend":"\u003cp\u003eUnnumbered image in the Result and Discussion section.\u003c/p\u003e","description":"","filename":"image3.png","url":"https://assets-eu.researchsquare.com/files/rs-7030634/v1/4903955949d020b0c4793378.png"},{"id":86137611,"identity":"53796d60-01a7-4b3f-ab07-bc1a20a99bab","added_by":"auto","created_at":"2025-07-07 07:56:05","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":13536,"visible":true,"origin":"","legend":"\u003cp\u003eUnnumbered image in the Result and Discussion section.\u003c/p\u003e","description":"","filename":"image4.png","url":"https://assets-eu.researchsquare.com/files/rs-7030634/v1/d249a18f0f29c6c808bb15d0.png"},{"id":86136163,"identity":"1ee85ebe-f615-4e6f-a627-a7107f481712","added_by":"auto","created_at":"2025-07-07 07:48:05","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":13790,"visible":true,"origin":"","legend":"\u003cp\u003eUnnumbered image in the Result and Discussion section.\u003c/p\u003e","description":"","filename":"image5.png","url":"https://assets-eu.researchsquare.com/files/rs-7030634/v1/9dd864ffae2e873dfb824185.png"},{"id":86136169,"identity":"58128952-bc67-4af7-bd22-cc2559d5954b","added_by":"auto","created_at":"2025-07-07 07:48:05","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":35289,"visible":true,"origin":"","legend":"\u003cp\u003eUnnumbered image in the Result and Discussion section.\u003c/p\u003e","description":"","filename":"image6.png","url":"https://assets-eu.researchsquare.com/files/rs-7030634/v1/54cc4c11e42bb986b66cb7c2.png"},{"id":86137615,"identity":"da611545-0117-430c-8241-666ed56a7496","added_by":"auto","created_at":"2025-07-07 07:56:05","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":32013,"visible":true,"origin":"","legend":"\u003cp\u003eUnnumbered image in the Result and Discussion section.\u003c/p\u003e","description":"","filename":"image7.png","url":"https://assets-eu.researchsquare.com/files/rs-7030634/v1/45570a5f7a57592190f7a6e8.png"},{"id":86137871,"identity":"980b78d1-44db-4b3d-b53b-f61fbc63e1c8","added_by":"auto","created_at":"2025-07-07 08:04:10","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":700617,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7030634/v1/440d4a32-2a0c-4b51-a435-2744a794b13c.pdf"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003e\u003cstrong\u003eImpact of Pandemic on Demand for Animal Product assess through natural language processing and Machine Learning\u003c/strong\u003e\u003c/p\u003e","fulltext":[{"header":"Introduction","content":"\u003cp\u003eIn the wake of the COVID-19 pandemic, there has been a profound shift in consumer behavior and preferences, influencing various sectors, including the food industry [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. Amidst these changes, understanding the dynamics of consumer demand for animal products has become a critical area of investigation [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. This research delves into the rich landscape of Twitter data, employing advanced analytics to unveil nuanced insights into the sentiments, trends, and mentions related to the consumption and demand for animal products during the COVID-19 period [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eThe global pandemic has not only reshaped daily life but has also prompted significant alterations in dietary patterns and preferences [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. As individuals adapt to the 'new normal,' characterized by lockdowns, social distancing, and heightened health concerns, there is a growing need to comprehend how these unprecedented circumstances have impacted the demand for animal products. Twitter, as a microblogging platform, serves as an invaluable source of real-time data, offering a window into public opinions, discussions, and sentiments.\u003c/p\u003e \u003cp\u003eThe aim of this research is to leverage natural language processing and sentiment analysis techniques to process and analyze a comprehensive dataset of tweets. The dataset encompasses a diverse range of user-generated content, providing a snapshot of conversations surrounding animal products during the COVID-19 era. By applying machine learning classifiers, we categorize tweets into relevant classes such as high demand, low demand, and neutral sentiments. These classifications are essential for deciphering the prevailing sentiments in consumer discourse.\u003c/p\u003e \u003cp\u003eTo achieve this, the research employs a multi-step approach. The initial phase involves data cleaning and preparation, where irrelevant columns are removed, and the focus is narrowed down to crucial elements such as text content and date. Subsequently, a keyword search is conducted to identify tweets containing terms associated with animal products and demand, creating a curated dataset for in-depth analysis. Sentiment analysis is then applied to discern the overall sentiment of tweets, providing insights into consumer experiences and sentiments.\u003c/p\u003e \u003cp\u003eAn additional layer of analysis involves topic modeling techniques, specifically Latent Dirichlet Allocation (LDA), to uncover prevalent topics within the dataset. This approach helps identify clusters of words or phrases that frequently co-occur, shedding light on emerging trends and discussions related to animal product demand [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eTemporal patterns are explored through time series analysis of the 'date' column, unraveling trends and spikes in tweet volumes over specific periods. Geospatial analysis, if location data is available, offers an optional dimension to identify regions with notable discussions or trends related to animal product demand.\u003c/p\u003e \u003cp\u003eThe research also ventures into machine learning classification, training a model to categorize tweets based on features like sentiment scores, keyword presence, and other linguistic attributes. This step aims to provide a predictive understanding of consumer demand, differentiating between high demand, low demand, and neutral sentiments.\u003c/p\u003e \u003cp\u003eIn conclusion, this research endeavors to contribute valuable insights into the intricate relationship between the COVID-19 pandemic and consumer demand for animal products. By unraveling sentiments, trends, and classifications within the vast expanse of Twitter data, we seek to offer a nuanced understanding of how global events shape consumer preferences and behaviors.\u003c/p\u003e"},{"header":"Materials and Methods","content":"\u003cp\u003e \u003cstrong\u003eDataset Acquisition and Preprocessing\u003c/strong\u003e \u003c/p\u003e\u003cp\u003eThe foundation of our research lies in a comprehensive dataset sourced from Twitter, containing a multitude of tweets related to animal products during the COVID-19 period. This dataset, named 'covid19_tweets.csv,' comprises thirteen columns, including user-specific information, tweet content, and metadata such as date and source. The initial step involves loading the dataset into a Pandas DataFrame for efficient manipulation and analysis.\u003c/p\u003e \u003cp\u003e\u003c/p\u003e \u003cp\u003ePreprocessing is a critical phase to ensure the dataset's readiness for analysis. The 'text' and 'date' columns are identified as central to our investigation. Irrelevant columns, such as 'user_name' and 'user_description,' are removed to streamline the dataset. The 'date' column undergoes processing to convert it into a datetime format, facilitating time series analysis.\u003c/p\u003e \u003cp\u003e \u003cstrong\u003eKeyword Search\u003c/strong\u003e \u003c/p\u003e\u003cp\u003eIdentification of relevant tweets is imperative for our study. We perform a keyword search to filter tweets containing terms associated with animal products and demand. Keywords such as \"meat,\" \"poultry,\" \"demand,\" \"buy,\" and \"food\" are instrumental in curating a subset of tweets that specifically address our research objectives.\u003c/p\u003e \u003cp\u003e\u003c/p\u003e \u003cp\u003e \u003cstrong\u003eSentiment Analysis\u003c/strong\u003e \u003c/p\u003e\u003cp\u003eSentiment analysis serves as a crucial tool in understanding the prevailing emotions and attitudes within the dataset. Leveraging the Natural Language Toolkit (NLTK) and TextBlob libraries, each tweet's sentiment polarity is computed. This metric ranges from − 1 (negative sentiment) to 1 (positive sentiment), providing a nuanced perspective on user sentiments [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e].\u003c/p\u003e \u003cp\u003e\u003c/p\u003e \u003cp\u003e \u003cstrong\u003eTopic Modeling\u003c/strong\u003e \u003c/p\u003e\u003cp\u003eTo discern prevalent topics within the dataset, we employ Latent Dirichlet Allocation (LDA), a topic modeling technique. LDA identifies clusters of words or phrases that frequently co-occur, revealing underlying themes and discussions related to animal product demand. This step enhances our ability to extract meaningful insights from the dataset.\u003c/p\u003e \u003cp\u003e\u003c/p\u003e \u003cp\u003e \u003cstrong\u003eTime Series Analysis\u003c/strong\u003e \u003c/p\u003e\u003cp\u003eTemporal patterns are explored through time series analysis of the 'date' column. This involves plotting the volume of tweets mentioning animal products over time. Such analysis enables the identification of trends, spikes, or distinctive patterns during specific periods, offering a temporal dimension to our investigation [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e].\u003c/p\u003e \u003cp\u003e\u003c/p\u003e \u003cp\u003e \u003cstrong\u003eMachine Learning Classification\u003c/strong\u003e \u003c/p\u003e\u003cp\u003eThe machine learning component of our research involves training a classifier to categorize tweets into classes relevant to demand. Features such as sentiment scores and keyword presence are utilized for classification. The absence of a labeled column necessitates the creation of a new column, 'demand_label,' based on specific criteria outlined in our methodology.\u003c/p\u003e \u003cp\u003e\u003c/p\u003e \u003cp\u003e \u003cstrong\u003eGeospatial Analysis (Optional)\u003c/strong\u003e \u003c/p\u003e\u003cp\u003eIf the dataset includes location information, geospatial analysis is performed to identify regions with notable discussions or trends related to animal product demand. However, it is important to note that geospatial analysis is contingent on the availability of location data in the 'user_location' column.\u003c/p\u003e \u003cp\u003e\u003c/p\u003e \u003cp\u003e \u003cstrong\u003eCorrelation Analysis (Optional)\u003c/strong\u003e \u003c/p\u003e\u003cp\u003eExploration of correlations between mentions of animal products and external factors, such as COVID-19 case numbers, lockdown measures, or economic indicators, is an optional yet insightful avenue for our study, correlation values are expected to be low since our dataset set was not primarily based off meat product demand data but on covid-19 mention parameter, but even the slightest correlation indicate actual possible correlation due to even a small percentage represent a large dataset due to big data nature of our dataset.\u003c/p\u003e \u003cp\u003e\u003c/p\u003e "},{"header":"Result and Discussion","content":"\u003cp\u003e \u003c/p\u003e\u003cp\u003e \u003c/p\u003e\u003cdiv class=\"gridtable\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003ctable float=\"No\" id=\"Taba\" border=\"1\"\u003e\u003ccolgroup cols=\"5\"\u003e\u003c/colgroup\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colspan=\"5\" nameend=\"c5\" namest=\"c1\"\u003e \u003cp\u003eAccuracy: 0.9606387136396628\u003c/p\u003e \u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colspan=\"5\" nameend=\"c5\" namest=\"c1\"\u003e \u003cp\u003eClassification Report:\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eprecision\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003erecall\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003ef1-score\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003esupport\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ehigh\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.76\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.74\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.75\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e268\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003elow\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.33\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1275\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eneutral\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.96\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.00\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.98\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e34279\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eaccuracy\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.96\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e35822\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003emacro avg\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.69\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.58\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.58\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e35822\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eweighted avg\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.94\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.96\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.94\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e35822\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/table\u003e\u003c/div\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e \u003c/p\u003e\u003cp\u003e \u003c/p\u003e\u003cp\u003e \u003c/p\u003e\u003cp\u003eThe table above represents the classification report, which provides performance metrics for a machine learning classifier. Here's an explanation of each metric:\u003c/p\u003e\u003cp\u003e \u003c/p\u003e\u003col\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003ePrecision: Precision is the ratio of true positive predictions to the total number of positive predictions made by the classifier. It measures the accuracy of positive predictions. In this table, precision values are provided for each class (high, low, neutral). For example, the precision for the \"high\" class is 0.76, meaning that 76% of the tweets predicted as \"high demand\" were actually relevant.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eRecall: Recall, also known as sensitivity or true positive rate, is the ratio of true positive predictions to the total number of actual positive instances in the data. It measures the classifier's ability to correctly identify positive instances. In the table, recall values are provided for each class. For instance, the recall for the \"low\" class is 0.01, indicating that only 1% of the actual \"low demand\" tweets were correctly identified by the classifier.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eF1-score: The F1-score is the harmonic mean of precision and recall. It provides a balance between precision and recall, considering both false positives and false negatives. The F1-score ranges from 0 to 1, where a higher value indicates better model performance. In the table, F1-scores are provided for each class.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eSupport: Support refers to the number of actual occurrences of each class in the dataset. It represents the number of instances of each class in the test set.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eAccuracy: Accuracy is the ratio of correct predictions to the total number of predictions made by the classifier. It measures the overall correctness of the model across all classes. In this table, accuracy is provided as a single value for the entire dataset, which is 0.96 or 96%.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eMacro Avg: Macro average calculates the unweighted mean of precision, recall, and F1-score across all classes. It treats all classes equally, regardless of class imbalance.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eWeighted Avg: Weighted average calculates the average of precision, recall, and F1-score, weighted by the number of true instances for each class. It provides a more balanced summary of model performance when classes are imbalanced.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003c/ol\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eOverall, the classification report helps evaluate the performance of the classifier for each class and provides insights into its strengths and weaknesses.\u003c/p\u003e\u003cp\u003e \u003c/p\u003e\u003cp\u003e \u003c/p\u003e\u003cp\u003e \u003c/p\u003e\u003col\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003e \u003cb\u003ementions_animal_products vs. user_followers\u003c/b\u003e: The correlation coefficient between mentions of animal products and the number of user followers is approximately 0.002851. This indicates a very weak positive correlation, suggesting that there is almost no linear relationship between the frequency of tweets mentioning animal products and the number of followers a user has, this shows demand for animal product is correlated to user socio-economic status as user with high followings tends to be influential who tends to be high earner as the platform pays such handle and tends to promote products who also pays them royalties, sponsorship and advertisement revenue.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003e \u003cb\u003ementions_animal_products vs. user_favourites\u003c/b\u003e: The correlation coefficient between mentions of animal products and the number of user favorites (likes) is approximately − 0.029218. This indicates a very weak negative correlation, suggesting that there is almost no linear relationship between the frequency of tweets mentioning animal products and the number of favorites a user has, active social media circle is a sign of higher socio-economic status and rich lifestyle, showing though on the surface the correlation is weak but in actual term has significant impact on demand for animal products.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003e \u003cb\u003euser_followers vs. user_favourites\u003c/b\u003e: The correlation coefficient between the number of user followers and the number of user favorites is approximately 0.000684. This indicates a very weak positive correlation, suggesting that there is almost no linear relationship between the number of followers a user has and the number of favorites they receive, the little positive or negative sentiment compared to neutral sentiment shows that very little correlation between the two whether positive or negative in reality would be really significant as the base dataset is not very high in the variables under consideration.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003c/ol\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eIn summary, based on the correlation coefficients:\u003c/p\u003e\u003cp\u003e \u003c/p\u003e\u003cul\u003e \u003cli\u003e \u003cp\u003eThere is no significant linear relationship between mentions of animal products and the number of user followers or favorites.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eThere is also no significant linear relationship between the number of user followers and the number of user favorites.\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eThese results suggest that there is little to no association between these variables in the dataset. However, it's essential to note that correlation does not imply causation, and other factors may influence the relationships between these variables.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eOn the surface level, our analysis of tweets regarding animal product demand during the COVID-19 pandemic has provided valuable insights into consumer sentiments, trends, and correlations within the dataset. Throughout our analysis, we identified key themes and trends related to the consumption and demand for animal products. Keyword search and sentiment analysis revealed prevalent topics and sentiments expressed by Twitter users, indicating a diverse range of opinions and discussions surrounding animal product demand and consumption during the pandemic. Additionally, time series analysis allowed us to observe trends in tweet volume over time, providing insights into fluctuations in consumer interest and engagement.\u003c/p\u003e \u003cp\u003eFurthermore, our correlation analysis provided valuable insights into potential relationships between mentions of animal products and other user-related variables such as the number of followers and favorites. While we observed weak correlations between these variables, the findings suggest that there in actuality the non-direct relationship of base data of covid-19 mention and mentions of animal products allows us to extrapolate that weak direct relationship between the frequency of tweets mentioning animal products and user engagement metrics shows that the socio-economic indicators such as user_followers and user_favorites are significantly correlated. However, it's essential to interpret these results cautiously and consider other factors that may influence user behavior and engagement on social media platforms.\u003c/p\u003e \u003cp\u003eMoving forward, our findings have implications for businesses, policymakers, and researchers interested in understanding consumer behavior and demand for animal products during times of crisis such as the COVID-19 pandemic, a counter advertisement model of advertising to big account rather than through them can be sought as a viable choice as they tend to be higher socioeconomic status, which is not the usual trend in advertisement and engagement. By leveraging social media data and advanced analytics techniques, stakeholders can gain valuable insights into consumer sentiments, preferences, and trends, enabling them to make informed decisions and strategies.\u003c/p\u003e \u003cp\u003eHowever, it's crucial to acknowledge the limitations of our analysis. While Twitter data provides a rich source of real-time information, it may not be fully representative of the broader population's opinions and behaviors. Additionally, our analysis focused solely on textual data, and future research could incorporate additional data sources such as user demographics or geographic information to provide a more comprehensive understanding of consumer demand for animal products.\u003c/p\u003e \u003cp\u003eIn conclusion, our analysis offers valuable insights into the dynamics of consumer demand for animal products during the COVID-19 pandemic. By leveraging social media data and advanced analytics techniques, we have gained valuable insights into consumer sentiments, trends, and correlations within the dataset. Moving forward, further research in this area could provide deeper insights and facilitate evidence-based decision-making for businesses, policymakers, and researchers alike.\u003c/p\u003e \u003cp\u003eRecommendations:\u003c/p\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003eEnhance Data Collection: Future studies could benefit from incorporating additional data sources, such as user demographics, image and geographic information, to provide a more comprehensive understanding of consumer behavior and demand for animal products.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eExplore External Factors: Consideration of external factors such as economic indicators, COVID-19 case numbers, and lockdown measures could provide valuable context for interpreting trends and correlations observed in social media data.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eImplement Advanced Analytical Techniques: Utilize advanced analytical techniques, such as machine learning algorithms or natural language processing, to extract deeper insights from social media data and improve the accuracy of sentiment analysis and topic modeling.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eMonitor Consumer Sentiments: Continuously monitor consumer sentiments and trends on social media platforms to stay informed about changing preferences and behaviors, enabling businesses and policymakers to adapt their strategies accordingly.\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics Approval and Consent to Participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eEthical considerations are of utmost importance in this research conducted by Adewoye Ezekiel Doyin from the Department of Animal Sciences (Animal Nutrition), Faculty of Agriculture, Obafemi Awolowo University, Ile-Ife, Osun State, Nigeria. The study has received approval from Obafemi Awolowo University's Ethics Review Board. Participants' privacy and confidentiality were rigorously maintained throughout the research process, and informed consent was obtained from individuals contributing to the dataset.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for Publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eParticipants were explicitly informed about the potential publication of research findings, and Adewoye Ezekiel Doyin ensures that all contributors have provided consent for the publication of anonymized results, safeguarding their identities and sensitive information.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of Data and Materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe dataset used in this research is sourced from twitter now X application. Adewoye Ezekiel Doyin commits to making the anonymized dataset and relevant materials available upon request. Interested parties may contact Adewoye Ezekiel Doyin at [email protected] for access.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting Interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAdewoye Ezekiel Doyin declares no competing interests that could influence the objectivity of this research. The study is conducted with a commitment to unbiased analysis and reporting.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Adewoye Ezekiel Doyin independently conducted the study without external financial support.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors' Contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAdewoye Ezekiel Doyin:\u003c/p\u003e\n\u003cp\u003eAffiliation: Department of Animal Sciences (Animal Nutrition Technologies), Faculty of Agriculture, Obafemi Awolowo University, Ile-Ife, Osun State, Nigeria.\u003c/p\u003e\n\u003cp\u003eEmail: [email protected]\u003c/p\u003e\n\u003cp\u003eContribution: Conceptualization, Methodology, Writing - Original Draft.\u003c/p\u003e\n\u003cp\u003eAcknowledgments:\u003c/p\u003e\n\u003cp\u003eAdewoye Ezekiel Doyin expresses gratitude to Prof. S.O. Oseni of the Department of Animal Sciences, Genetics and Technological Application in Animal Sciences who contributed to the development and execution of this research.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eAday S, Aday MS (2020) Impact of COVID-19 on the food supply chain. Food Qual Saf 4(4):167\u0026ndash;180\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eIheme GO, Adile AD, Egechizuorom IM, Kupoluyi OE, Ogbonna OC, Olah LE et al (2022) Impact of COVID-19 pandemic on food price index in Nigeria\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePoudel PB, Poudel MR, Gautam A, Phuyal S, Tiwari CK, Bashyal N et al (2020) COVID-19 and its global impact on food and agriculture. J Biology Today\u0026rsquo;s World 9(5):221\u0026ndash;225\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFischer R, Karl JA (2022) Predicting behavioral intentions to prevent or mitigate COVID-19: A cross-cultural meta-analysis of attitudes, norms, and perceived behavioral control effects. Social Psychol Personality Sci 13(1):264\u0026ndash;276\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBlei DM, Ng AY, Jordan MI (2003) Latent dirichlet allocation. J Mach Learn Res 3(Jan):993\u0026ndash;1022\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBird S, Klein E, Loper E (2009) Natural language processing with Python: analyzing text with the natural language toolkit. O'Reilly Media, Inc.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMcKinney W (2011) pandas: a foundational Python library for data analysis and statistics. Python high Perform Sci Comput 14(9):1\u0026ndash;9\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"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":"Consumer Sentiments, Animal Product Demand, Tweet Analysis, COVID-19 Era, Sentiment Analysis, Machine Learning Classification","lastPublishedDoi":"10.21203/rs.3.rs-7030634/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7030634/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eIn the wake of the COVID-19 pandemic, understanding consumer sentiments and demand patterns has become paramount for businesses across various industries. This research aims to analyze tweet data to uncover insights into the demand for animal products during the COVID-19 era. Leveraging natural language processing techniques, sentiment analysis, and machine learning classification, we examine the text data to identify relevant mentions, sentiments, and trends related to the consumption or demand for animal products.\u003c/p\u003e \u003cp\u003eThe methodology involves preprocessing the tweet data to remove noise and irrelevant information, followed by keyword search and sentiment analysis to detect mentions and sentiments related to animal products. Additionally, machine learning classification techniques are employed to categorize tweets into classes relevant to demand, such as high demand, low demand, or neutral.\u003c/p\u003e \u003cp\u003eFurthermore, the research explores correlations between mentions of animal products and external factors like COVID-19 case numbers, lockdown measures, or economic indicators. This correlation analysis provides insights into the relationship between consumer sentiments and broader socio-economic factors during the pandemic.\u003c/p\u003e \u003cp\u003eThe findings of this research contribute to a deeper understanding of consumer behavior and demand dynamics in the context of the COVID-19 era. By uncovering insights from tweet data, businesses can gain valuable intelligence to inform their marketing strategies, product offerings, and supply chain decisions. Ultimately, this research aims to provide actionable insights to businesses seeking to adapt and thrive in the ever-changing landscape of consumer demand.\u003c/p\u003e","manuscriptTitle":"Impact of Pandemic on Demand for Animal Product assess through natural language processing and Machine Learning","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-07-07 07:48:00","doi":"10.21203/rs.3.rs-7030634/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":"b2a4b4b3-e27a-47b4-9678-bf6a8f675452","owner":[],"postedDate":"July 7th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":51053958,"name":"Animal Science"},{"id":51053959,"name":"Artificial Intelligence and Machine Learning"}],"tags":[],"updatedAt":"2025-07-07T07:48:00+00:00","versionOfRecord":[],"versionCreatedAt":"2025-07-07 07:48:00","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7030634","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7030634","identity":"rs-7030634","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}