A new pruning algorithm for game tree in Chinese Chess Computer Game

In the endgame stage of Chinese Chess Computer Game (CCCG), the complexity and diversity of positions make the endgame database always very huge. Thus, it is unsuitable and inefficient to develop an intelligent search engine based on the learning for the master players´ endgame database. In addition, the master players will stop the search of the best move for the current position if it can match with a remembered endgame pattern. However, the existing search engines select the best move based on the position values of leaf nodes of game tree, without considering the endgame patterns. Inspired by this process, we design a new pruning algorithm to select the best move for CCCG in the endgame stage. In this new algorithm, the refined master players endgame patterns have been fused into the search engine to prune the game tree. The experimental results demonstrate that our designed pruning algorithm is feasible and effective.