基于深度信念网络和三支决策的入侵检测算法

With the diversification and intelligentization of network intrusion behaviors，traditional intrusion detection algorithms are difficult to extract the features contained in the intrusion behaviors，and there are some deficiencies in intrusion detection performance. Therefore，this paper proposes an intrusion detection algorithm based on Deep Belief Network (DBN) and three⁃way decisions. Firstly，we construct a multi⁃granularity feature space by extracting features from high⁃dimensional data with deep belief network. Then，a classifier based on the three⁃way decisions theory to make immediate decisions on the intrusion behavior or normal behavior，and KNN classifier to furtherly analyze the uncertain network behavior in the boundary region according to different granularity characteristics are used. Experimental results on NSL⁃KDD dataset show that the proposed algorithm can improve the performance of intrusion detection system.

Keywords： Deep Belief Network ; Three⁃Way Decisions ; feature extraction ; intrusion detection