In order to eliminate irrelevant and redundant genes and improve the classification accuracy and efficiency, this paper proposed a cancer gene selection method by hybrid swarm algorithm based on cultural algorithm. Firstly, ReliefF algorithm is used to select the primary gene subset, and then hybrid swarm algorithm based on cultural algorithm is used to select the optimal information gene. Finally, the performance is tested by three standard tumor information gene data set. The simulation results show that the proposed method can remove noise and useless and reduce the computing complexity, the classification accuracy can reach 100%, and has good practical value.
