AntMiner is the first swarm intelligence algorithm based on the ant colony optimization algorithm in data mining area. Since AntMiner is easy to fall into the local optimal solution and has long convergence time, we introduce ACSMiner algorithm in this paper. ACSMiner is based on the ant colony system which constructs rules by using the state transition rules, not only updating the pheromone of each ant locally, but also updating the global optimal solution. The KDD99 dataset was used to carry out the experiment. Simulation results demonstrated that compared with the original algorithm, the variant works well in improving the accuracy and reducing false positive rate as well as shortening the training time. It shows that the method can be effectively used in the field of intrusion detection.