ARCHITECTURE OF A WEB-BASED SYSTEM FOR THE ANALYSIS OF TACTICAL PATTERNS AND THE AUTOMATED DETECTION OF ANOMALOUS DEVIATIONS IN CHESS GAMES

Abstract

The article discusses approaches to building a distributed web-based information system for the intelligent analysis of sports data that combines tactical pattern categorization and anomalous play detection within a single software solution. The aim of the study is to design a conceptual microservice architecture, develop the mathematical framework, and implement a scalable web application capable of automating the full cycle of processing, retrospective analysis, and statistical evaluation of digital chess records in PGN format.

The methodological foundation consists of software engineering patterns, distributed systems theory, and mathematical statistics, in particular the calculation of average centipawn loss, the Z-score statistical deviation index based on the three-sigma rule, and the match rate metric relative to top chess engine evaluation lines.

A modular approach to design was used: the user management business logic, the analytical core, relational data storage, and inter-service communication mechanisms were separated. As a result of the research, a system topology was proposed that ensures the integrated execution of two key tasks-formalizing an individual playing style and ensuring academic or sporting integrity against computer assistance (cheating). The paper describes the core relational system entities, the mechanism for asynchronous non-blocking request dispatching via HTTP protocol, and the approach to service deployment, in which the base API is implemented in C# (.NET Core), and the resource-intensive analytical processing module is implemented as a separate service on FastAPI (Python) interacting with the Stockfish engine via the UCI protocol.

The practical significance of the results lies in the ability to use the proposed solution by chess referees, coaches, and analysts for monitoring fair play, conducting deep performance evaluation, and automated generation of analytics reports without the need for local infrastructure deployment. The developed system can serve as a foundation for further improvement through the integration of time-series analysis for time management, adaptation of models to shortened time controls (blitz and bullet), and the implementation of deep learning methods for advanced tactical style clustering.

References

1)          Campbell M., Hoane A. J., Hsu F. H. Deep Blue. Artificial Intelligence. 2002. Vol. 134, Iss. 1-2. P. 57–83. DOI: 10.1016/S0004-3702(01)00129-1.

2)          Silver D., Hubert T., Schrittwieser J. etc. A general reinforcement learning algorithm that masters chess, shogi, and Go through self-play. Science. 2018. Vol. 362, Iss. 6419. P. 1140–1144. DOI: 10.1126/science.aar6404.

3)          Romstad T., Costalba M., Kiiski J., Linscott G. Stockfish: Strong open-source chess engine. URL: https://stockfishchess.org (accessed: 25.04.2026).

4)          Regan K. W., Haworth G. M. Intrinsic Chess Rating. AAAI Conference on Artificial Intelligence: proceedings of the Twenty-Fifth AAAI Conference, San Francisco, California, USA, August 7–11, 2011. Palo Alto: AAAI Press, 2011. P. 834–839. DOI: 10.1609/aaai.v25i1.7951 (accessed: 25.04.2026).

5)          Biswas T., Regan K. Measuring Level-K Reasoning, Satisficing, and Human Error in Game-Play Data. 2015 IEEE 14th International Conference on Machine Learning and Applications (ICMLA), Miami, FL, USA, 9–11 December 2015. 2015. DOI: 10.1109/icmla.2015.233 (accessed: 26.04.2026).

6)          Rosales Pulido H. A. Predicting the outcome of a chess game by statistical and machine learning techniques : Projecte Final de Màster Oficial / UPC, Facultat d'Informàtica de Barcelona, Departament d'Estadística i Investigació Operativa. Barcelona, 2016. URL: https://upcommons.upc.edu/server/api/core/bitstreams/0b6f26cb-0b1d-4c82-817e-a3f3440f466d/content (accessed: 26.04.2026).

7)          Cheng Isaac, Camargo Chico. Machine Learning to Study Patterns in Chess Games. 2023. DOI: 10.13140/RG.2.2.30894.52807

8)          Linville R. Understanding average centipawn loss in chess. chess.com. URL: https://www.chess.com/blog/raync910/average-centipawn-loss-chess-acpl (accessed: 30.04.2026).

9)          Chess games. lichess.org. URL: https://database.lichess.org (accessed: 02.05.2026).

10)      Fiekas N. Сhess 1.11.2. PyPI. URL: https://pypi.org/project/chess/ (accessed: 02.05.2026).