Using genetic feature selection for improving cyber attack detection rate

摘要: As Internet becomes an essential tool for all kinds of business transactions, the issue detecting network intrusion has received greater attention. In this paper, we suggest a novel method based on genetic optimization that can improve detection rate attack patterns without loss due to false-positive error rate. We focus selecting robust feature subset by designing multicriteria procedure. During evaluation phase, performance proposed approach is contrasted against one state-of-the-art selection methods using k nearest neighbor classifier. Experimental results show remarkably effective than full set.