关键词: 遗传算法, K均值聚类, 入侵检测, 特征选择

Abstract: In order to improve performances of intrusion detection system in terms of detection speed and detection rate，a novel unsupervised intrusion detection method based on Genetic Algorithm（GA） and feature selection mechanism is proposed.In the method，an improved GA is adopted as search strategy.On the other hand，the K-means clustering is used to classify the feature data，whose evaluation target is the ratio of the between-class scatter to the within-class scatter.Then，the optimal feature subset is found and applied to unsupervised intrusion detection.The experimental results show that the method can solve the feature selection problem of intrusion detection effectively，and it has a better detecting effect than unsupervised intrusion detection without feature selection.

Key words: Genetic Algorithm（GA）, K-means clustering, intrusion detection, feature selection