Machine game has always been considered to be the most challenging task in the field of artificial intelligence, taking the Chinese chess machine game as an example, in order to achieve a deeper level of game tree search in a limited time, an improved pruning strategy is proposed. Based on the traditional αβ search algorithm, by introducing iterative deepening ,history table and other auxiliary means of pruning, discard worthless node search, effectively reduce the number of nodes, make the pruning occurs earlier to shorten the search time. Experimental results show that the improved strategy search algorithm significantly improve the efficiency of pruning, at the same time, the search depth is higher than the original search algorithm, the number of nodes is also increased.