SSA-CFS:An Effective Feature Selection Method for Intrusion Detection System

The development of network has brought great convenience to our lives, but the number of network intrusions has also increased sharply at the same time. Intrusion detection system(IDS) is one of the most effective ways to solve this problem. IDS based on machine learning is an important issue of current research. Feature selection is a commonly used method in IDS to improve model accuracy and timeliness. In this paper, we propose a heuristic algorithm SSA-CFS based on SSA and CFS, and build a two-layer classification model based on this algorithm. We conduct experiments on the data set named NSL-KDD to evaluate the performance of the algorithm and IDS, and compare them with other traditional algorithms and models. The experimental results show that the feature selection algorithm and IDS model we proposed have significantly improved the time and detection accuracy.