dc.description.abstract | Game playing is a popular sub-field of the artificial intelligence (AI) branch of
Computer Science. AI researchers have paid particular attention to "deterministic
two-player games of perfect information". The ancient African game of Mancala is
such a game. There are many variations of the game; the Kenyan Ajua defines one
variation. In order for a computer to playa game of this nature, it needs to use an
evaluation function: this is a function that evaluates how favourable or unfavourable a
given position is. The heart of many game player programs is the minimax algorithm.
For efficiency, the minimax algorithm often is enhanced with alpha-beta cut-offs, thus
forming the alpha-beta search algorithm.
This project designed and implemented a computer program to play the board game
Ajua using the minimax search algorithm that uses an evaluation function to choose
its moves. The program allows both human/human and human/computer games to be
played, with the computer player being competitive and enjoyable to play against | en |