A program which shows signs of intelligent behaviour can be classified as an AI program. Game Theory has significant influence in the area of AI. One of the areas in which game theory is typically employed are two player games like Chess, Othello, Push, Tic-Tac-Toe etc. The domain model of a two player strategy game can expressed in terms of board, rules, pieces, players and strategies. It's quite interesting to note that there are algorithms which tackle two player games, which implies that if we can sufficiently abstract the game, we can use the same algorithm (engine) to play any type of two player game by plugging in the game rules and strategies.

In this program, I have tried to create a generic framework for the computer to play any two player strategy game. As an example, I have included applets for Tic-Tac-Toe and Chess being built upon the same core framework. I have used alpha-beta pruning as the core algorithm and have supplimented it with game specific rules and strategies. It is quite interesting to note that for a simple game like Tic-Tac-Toe, where the solution space is limited, it is impossible for a human player to win against the computer. Where as, for a game with a huge solution space, the algorithm can show decent signs of intelligence by selectively pruning the game tree to a reasonable depth.

You can follow the links below to explore more. I have included the source code, documentation and applets. Please note that this source code was developed sometimes during 2001 and has not been updated since.

Framework | Tic-Tac-Toe | Chess | Source Code | Rose Model | Javadocs (download) | Javadocs (online)