Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{189610,
        author = {Bharath R and Meghana P and Sake Ajay and Srinith N and Dr. B. Vani},
        title = {An Explainability Focused Computational Pipeline for Anti-Cheat and Churn Monitoring in Online Gaming Platforms},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {12},
        number = {7},
        pages = {7020-7032},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=189610},
        abstract = {Online Multiplayer games face regular difficulty related to cheating and player churn, both of which negatively impact Equality, engagement and platform sustainability. Existing detection mechanisms often rely on single-modal analysis or operate as black-box systems, limiting adaptability and trust. This paper presents an explainable multi-view computational framework that jointly addresses cheating detection and churn prediction by integrating player telemetry features, behavioral patterns, social interaction structures, and gameplay images. Four complementary views are included in the proposed system: (i) an image-based view that uses convolutional neural networks to detect visual cheating artefacts; (ii) a behavioural view for churn prediction based on temporal engagement indicators; (iii) a social view that models player interactions as networks to identify influence-driven risk; and (iv) a portrait view that uses supervised machine learning on aggregated player features. Using SHAP for tabular models, network-based visual reasoning for social analysis, and Grad-CAM for CNN-based image predictions, explainability is integrated throughout the pipeline. The framework is implemented as a Flask-based web application with interactive dashboards, persistent logging, and real-time inference. Strong performance is demonstrated by experimental evaluation across all modules, with high detection accuracy, low false-positive rates, and understandable explanations for humans. The findings show that explainable AI in conjunction with multi-view learning offers a reliable and practical method for contemporary online game analytics.},
        keywords = {SHAP, Grad-CAM, CNN, explainable AI, multi-view learning, player churn prediction, online game cheating detection, and machine learning.},
        month = {December},
        }