PLAYER BEHAVIOR PREDICTION MODEL IN MOBILE GAMES BASED ON MACHINE LEARNING
DOI:
https://doi.org/10.32689/maup.it.2025.4.9Keywords:
player behavior, mobile games, machine learning, prediction, neural networks, ensemble methodsAbstract
In the modern mobile gaming industry, accurate prediction of player behavior is an important aspect for improving user retention, optimizing gameplay mechanics, personalizing content, and enhancing marketing efficiency. With the growing competition and the massive volume of player data, the application of machine learning methods for automated behavior analysis and prediction has become increasingly relevant and strategically significant. Objective. The purpose of this study is to develop an effective model for predicting player behavior in mobile games using advanced machine learning algorithms. The proposed model aims to determine the probability of player churn, predict in-game activity and level completion, as well as assess potential engagement in in-game transactions and responses to gamified stimuli. Methodology. The research is based on the analysis of a large set of anonymized behavioral data collected from players, including session duration, login frequency, number of actions performed, task completion rate, in-game purchases, inactivity periods, and other behavioral metrics. To construct and compare prediction models, several algorithms were applied: Logistic Regression, K-Nearest Neighbors, Random Forest, Gradient Boosting, and multilayer neural networks. Comparative evaluation of their accuracy and F1-score made it possible to identify the most efficient model for predicting user behavior patterns with high precision. Scientific novelty. The novelty of this research lies in the integration of behavioral metrics with modern ensemble learning algorithms and deep neural networks to form a comprehensive predictive model. This approach enables the capture of complex nonlinear dependencies between player characteristics, providing a significant accuracy improvement over traditional statistical and baseline machine learning models. The developed system also allows identifying the key factors influencing user engagement and churn, which supports the creation of personalized retention and re-engagement strategies. Conclusions. The experimental results demonstrate that the proposed model increases prediction accuracy by 12–18% compared to classical approaches. The model effectively identifies high-risk players, predicts engagement dynamics, and evaluates the efficiency of personalized in-game offers and retention strategies. The research findings can be applied by mobile game developers to improve player retention, reduce churn rates, and create adaptive gaming experiences tailored to individual behavior patterns. Therefore, the study contributes both to the academic field of game analytics and to practical applications in the modern mobile gaming industry.
References
Завгородній В. В., Завгородня Г. А., Валявська Н. О., Адаменко В. С., Дороговцев Є. В., Несмачний П. В. Метод автоматичної генерації контенту на основі процедурних алгоритмів. Вчені записки Таврійського національного університету імені В. І. Вернадського. Серія: Технічні науки. 2022. Т. 33 (72), № 1. С. 91–96. DOI: https://doi.org/10.32838/2663-5941/2022.1/15
Завгородній В. В., Завгородня Г. А., Валявська Н. О., Герасименко О. О., Калюжний О. В., Степовий А. В. Пошук аномалій у даних за допомогою машинного навчання. Вчені записки ТНУ імені В. І. Вернадського. Серія: Технічні науки. 2022. Т. 33 (72), № 3. С. 39–43. URL: https://tech.vernadskyjournals.in.ua/journals/2022/3_2022/6.pdf
Завгородній В. В., Завгородня Г. А., Демченко І. В., Крамаренко К. С., Шевченко І. О., Юрченко А. В. Метод створення штучних текстур із заданими параметрами. Вчені записки ТНУ імені В. І. Вернадського. Серія: Технічні науки. 2022. Т. 33 (72), № 2. С. 86–90. DOI: https://doi.org/10.32838/2663-5941/2022.2/14
Завгородній В. В., Завгородня Г. А., Дроботович К. Є., Тенігін О. В., Шматко М.М. Математичне моделювання у методах формального дослідження. Вчені записки ТНУ імені В. І. Вернадського. Серія: Технічні науки. 2021. Т. 32 (71), № 6. С. 75–79. DOI: https://doi.org/10.32838/2663-5941/2021.6/12
Barros e Sá G. C., Madeira C.A.G. Deep reinforcement learning in real-time strategy games: a systematic literature review. Applied Intelligence. 2025. Vol. 55. P. 243. DOI: https://doi.org/10.1007/s10489-024-06220-4
Drachen A., Lundquist E.T., Kung Y., Rao P.S., Klabjan D., Sifa R., Runge J. Rapid Prediction of Player Retention in Freeto-Play Mobile Games. arXiv preprint. 2016. DOI: https://doi.org/10.48550/arXiv.1607.03202
Farrokhi Maleki M., Zhao R. Procedural Content Generation in Games: A Survey with Insights on Emerging LLM Integration. arXiv. 2024. URL: https://arxiv.org/html/2410.15644v1
Holt J. I. Integrating Player-Centric Procedural Content Generation. University of Utah Technical Report. Salt Lake City, 2025. URL: https://www-old.cs.utah.edu/docs/techreports/2025/UUCS-25-002.pdf
Latif A., Zuhairi M. F., Khan F. Q., Randhawa P. A Critical Evaluation of Procedural Content Generation Approaches for Digital Twins. Journal of Sensors. 2022. Vol. 2022 (4). P. 1–13. DOI: https://doi.org/10.1155/2022/5629645
Liu J., Snodgrass S., Khalifa A., Risi S., Yannakakis G.N., Togelius J. Deep Learning for Procedural Content Generation. Neural Computing and Applications. 2020. DOI: https://doi.org/10.1007/s00521-020-05383-8
Mulla R., Potharaju S., Tambe N., Joshi S., Kale K., Bandishti P., Patre R. Predicting Player Churn in the Gaming Industry: A Machine Learning Framework for Enhanced Retention Strategies. Journal of Current Science and Technology. 2025. Vol. 15 (2). P. 103. DOI: https://doi.org/10.59796/jcst.V15N2.2025.103
Mustač K., Bačić K., Skorin-Kapov L., Sužnjević M. Predicting Player Churn of a Free-to-Play Mobile Video Game Using Supervised Machine Learning. Applied Sciences. 2022. Vol. 12 (6). P. 2795. DOI: https://doi.org/10.3390/app12062795
Óskarsdóttir M., Gísladóttir K. E., Stefánsson R. et al. Social networks for enhanced player churn prediction in mobile free-to-play games. Applied Network Science. 2022. Vol. 7. P. 82. DOI: https://doi.org/10.1007/s41109-022-00524-5
Paduraru C., Paduraru M., Stefanescu A. RiverGame – a game testing tool using artificial intelligence. In: Proceedings of the 2022 IEEE Conference on Software Testing, Verification and Validation (ICST). Valencia, Spain, 2022. P. 422–432. DOI: https://doi.org/10.1109/ICST53961.2022.00048
Peng L. Research on Machine Learning Models for Predicting Player Churn. In: Proceedings of the 1st International Conference on Modern Logistics and Supply Chain Management (MLSCM). Singapore, 2024. P. 112–121. DOI: https://doi.org/10.5220/0013234700004558
Rismayanti N. Predicting Online Gaming Behaviour Using Machine Learning Techniques. Indonesian Journal of Data and Science. 2024. Vol. 5 (2). DOI: https://doi.org/10.56705/ijodas.v5i2.166
Rodriguez-Diaz N., Aspandi D., Sukno F., Binefa X. Machine Learning-based Lie Detector applied to a Novel Annotated Game Dataset. arXiv preprint. 2021. URL: https://arxiv.org/abs/2104.12345
Romero-Méndez E.A., Santana-Mancilla P.C., Garcia-Ruiz M., Montesinos-López O.A., Anido-Rifón L.E. The Use of Deep Learning to Improve Player Engagement in a Video Game through a Dynamic Difficulty Adjustment Based on Skills Classification. Applied Sciences. 2023. Vol. 13 (14). P. 8249. DOI: https://doi.org/10.3390/app13148249
