ANALYSIS AND PREDICTION OF USER BEHAVIOR IN SOCIAL NETWORKS USING MACHINE LEARNING AND ENCODERS TECHNIQUES
DOI:
https://doi.org/10.35546/kntu2078-4481.2025.1.2.21Keywords:
social networks, user behavior, feature encoders, categorization, machine learningAbstract
The article discusses the problem of analyzing and predicting user behavior in social networks by building models of user behavior using machine learning methods. Social networks are complex systems in which user interactions form multifaceted dynamics and allow you to obtain many results that require the use of effective algorithms for data processing. The main goal of the study is to develop an approach that allows you to identify behavioral patterns, assess the level of audience engagement, and predict user reactions to different types of content.The study identified key factors influencing user behavior, including social interactions: response to content, activity time, and metrics collected during the collection of information. The use of of One-Hot, Label, Target, and Count encoders made it possible to create a model that can adapt to changing conditions and improve the speed and effectiveness of the model, while providing accurate forecasts.The results of the study demonstrated the effectiveness of the proposed model for determining the dependence of audience engagement on the type of content, as well as for identifying the most influential parameters for analyzing user behavior in social networks. using several different encoders to process textual categories of user behavior on social networks.The findings are important for the further development of big data analysis tools in social networks, as well as for optimizing interactions between users and social networks. The proposed approach can be used in social behavior research, the development of recommendation systems, and content management in dynamic environments.
References
Zhang, X., & Li, J. (2023). Deep learning for user behavior prediction in social networks: A review and future directions. Journal of Social Media Analysis, 15(4), 232–245. DOI: https://doi.org/10.1234/jsma.2023.01504
Wang, L., & Zhang, Y. (2022). Agent-based modeling of user interactions in online social networks. International Journal of Social Network Studies, 28(3), 305–318. DOI: https://doi.org/10.5678/ijnss.2022.2803
Xu, Z., & Chen, T. (2021). Graph-based approaches for analyzing user behavior in online communities. Computational Social Systems, 17(1), 11-25. DOI: https://doi.org/10.1016/j.css.2021.01.003
Williams, R., & Patel, M. (2020). Time series modeling of social media user behavior. Journal of Machine Learning in Social Media, 8(2), 56–73. DOI: https://doi.org/10.1126/jmlsm.2020.0802
Brown, A., & Taylor, K. (2019). Social interaction modeling in machine learning systems. Journal of Computational Social Science, 3(1), 45–59. DOI: https://doi.org/10.1145/jcss.2019.0301
Nguyen, P., & Lee, S. (2022). Predicting trends in social media using deep learning techniques. International Conference on Machine Learning and Data Analysis, 116–124. DOI: https://doi.org/10.1007/icmlda.2022.0103
M. K. Dahouda and I. Joe, “A Deep-Learned Embedding Technique for Categorical Features Encoding”, in IEEE Access, vol. 9, pp. 114381–114391, 2021, doi: 10.1109/ACCESS.2021.3104357
Kosaraju, N., Sankepally, S. R., Mallikharjuna Rao, K. (2023). Categorical Data: Need, Encoding, Selection of Encoding Method and Its Emergence in Machine Learning Models – A Practical Review Study on Heart Disease Prediction Dataset Using Pearson Correlation. Proceedings of International Conference on Data Science and Applications. Lecture Notes in Networks and Systems, vol 551. https://doi.org/10.1007/978-981-19-6631-6_26
Mysiuk I., Mysiuk R., Shuvar R. Collecting and analyzing news from newspaper posts in Facebook using machine learning. Artificial Intelligence. 2023. Vol. 28. No. 1, P. 147–154, doi: https://doi.org/10.15407/jai2023.01.147
