The research introduces a hybrid behavioural analysis technique for early detection of anomalous user behavior observed on web applications. This strategy involves statistical probability modeling and sequence- based deep learning to design interpretable and robust anomaly detection. A probability baseline has been obtained as a probabilistic basis using KDE (Kernel Density Estimation) and longitudinal time series patterns are sampled using an LSTM network. The hybrid anomaly score combines these two models to dynamically analyze behavioural deviations. The proposed approach has been applied to synthetic behavioural data and demonstrated enhanced detection accuracy and reduced false alarms compared to independent statistical or learning-based models. The results have shown the method is capable for adaptive, transparent intrusion detection in web environments, and it can be effectively adopted by contemporary cybersecurity solutions.
[1] Borowiec, M., & Rak, T. (2023). Advanced examination of user behavior recognition via log dataset analysis of web applications using data mining techniques. Electronics, 12(21), 4408. DOI:https://doi.org/10.3390/electronics12214408.
[2] Folino, G., Otranto Godano, C., & Pisani, F. S. (2023). An ensemble-based framework for user behaviour anomaly detection and classification for cybersecurity. Journal of Supercomputing, 79(11). DOI: https://doi.org/10.1007/s11227-023-05049-x.
[3] Vavryk, Y., & Opirskyy, I. Artificial Intelligence: Cybersecurity оf The New Generation. Ukrainian Scientific Journal of Information Security, 30(2), 242- 254. DOI: https://doi.org/10.18372/2225-5036.30.19235.
[4] S Semenenko, Y. (2025). Using artificial intelligence technologies to analyse consumer behaviour in e-commerce. Ekonomichnyy analiz, 35(1), 475-483. Econa: https://www.econa.org.ua/index.php/econa/article/view/6303.
[5] Kushnerov O., Pozovna I., Sokol V.(2024). The influence of neural networks on the development of cyber security in the conditions of regulatory changes. Ukrainian Scientific Journal of Information Security, 30(2), 261-269. DOI: https://doi.org/10.18372/2225-5036.30.19238.
[6] Benova, L., & Hudec, L. (2024). Comprehensive analysis and evaluation of anomalous user activity in web server logs. Sensors, 24(3), 746. DOI:https://doi.org/10.3390/s24030746.
[7] Xing, L., Li, S., Zhang, Q., Wu, H., Ma, H., & Zhang, X. (2024). A survey on social network’s anomalous behavior detection. Complex & Intelligent Systems, 10(4), 5917-5932. DOI: https://doi.org/10.1007/s40747-024-01446-8.
[8] Kudin, A., Tkach, V., Baranovskyi, O., & Carbunar, B. (2025). A Distributed System for Early Intrusion Detection and Assessment of Cybersecurity. DOI: https://doi.org/10.3390/electronics12030721.
[9] Syrovatchenko, M. (2024). Legal aspects of cybersecurity in Ukraine: current challenges and the role of national legislation. Bulletin of Lviv Polytechnic National University. Series: Legal Sciences, 1(41), 314-320. DOI: https://doi.org/10.23939/law2024.41.314.
[10] Khadzhiradieva, S., Bezverkhniuk, B., Nazarenko, O., Bazyka, S., & Dotsenko, T. (2024). Personal data protection: Between human rights protection and national security. Social and Legal Studios, 3(7), 245-256. DOI: https://doi.org/10.32518/sals3.2024.245.
[11] Zaplatynskyi, N., Lub, P., & Zaporozhtsev, S. (2024). Improving cybersecurity with artificial intelligence. Bulletin of Cherkasy State Technological University, 29(4), 53-61. DOI: https://doi.org/10.62660/bcstu/4.2024.53.
[12] Kravchuk, M., Kravchuk, V., Hrubinko, A., Podkovenko, T., & Ukhach, V. (2024). Cyber security in Ukraine: Theo- retical view and legal regulation. Law, Policy and Security, 2(2), 28-38. DOI: https://doi.org/10.62566/lps/2.2024.28.
[13] Rishniak M..(2025). hybridModel. [Source code]. GitHub. GitHub: https://github.com/martari03/hybridModel.