Journal of Computers, Mechanical and Management

G. Jadhav,* P. N. Mahalle, T. Desale, T. Deshmukh, S. Dhaytonde, S. Hajare, V. Gheware, A hybrid search–learning framework for artificial intelligence in board games, Journal of Information and Communications Technology: Algorithms, Systems and Applications, 2025, 1(3), 25315, doi: https://doi.org/10.64189/ict.25315.

Rights and permissions

© The Author(s) 2025.

Open Access

This article is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License, which permits the non-commercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as appropriate credit to the original author(s) and the source is given by providing a link to the Creative Commons License and changes need to be indicated if there are any. The images or other third-party material in this article are included in the article's Creative Commons License, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons License and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this License, visit: httpshttps://creativecommons.org/licenses/by-nc/4.0/

Abstract

Artificial Intelligence (AI) has played a transformative role in the evolution of board games by enabling machines to exhibit strategic reasoning, long-term planning, and adaptive decision-making. Board games such as Chess, Go, and Checkers provide well-defined environments with complex state spaces, making them ideal benchmarks for evaluating AI techniques. Early rule-based systems relied heavily on handcrafted heuristics and exhaustive search strategies, while recent advances leverage deep neural networks and reinforcement learning to achieve superhuman performance. This paper presents a hybrid study that combines a focused review of classical and modern AI approaches in board games with the design and evaluation of a proposed hybrid search–learning architecture. The proposed system integrates Minimax, Monte Carlo Tree Search (MCTS), and deep reinforcement learning within a modular four-layer framework to achieve scalability, adaptability, and efficient real-time decision-making. Extensive experimental evaluation on Chess, Go, and Checkers demonstrates that the proposed architecture achieves improved win rates, reduced inference latency, and measurable Elo rating gains compared to traditional and baseline AI systems. Beyond gaming, the findings highlight the broader applicability of board-game AI techniques in strategic planning, optimization, and human-AI interaction domains.

Graphical Abstract

Novelty statement

This research introduces a comprehensive feature engineering framework that substantially enhance linear regression performance for residential property valuation.