Design of Intrusion Detection Models using an Improved Bayesian Networks

摘要：

This paper concerns the design of intrusion detection model with the aid of Bayesian Network (BN). An algorithm for learning Bayesian network is proposed. In comparison with the conventional approaches (such as anomaly-based approaches), the proposed algorithm can build robust models of acceptable behavior which may not result in a large number of false alarms. In this algorithm, i database is input while the constructing BN structure is output. The algorithm is realized by means of conditional independence (Cl) tests,. The construction process is based on the computation of mutual information of attribute pairs. Experimental results show that the proposed model leads to improved performance in terms of accuracy and effectiveness of the Bayesian Networks Classification when compared with the wellknown K Nearest Neighbors (KNN) algorithm reported in the literature.

关键词： bayesian networks bayesian structure learning conditional independence intrusion detection

作者: Hui He

作者单位: Electronic Commerce Department of Business School JiangXi Normal University, NanChang, China

会议类型: 国际会议

会议名称: 2011 3rd International Conference on Computer and Network Technology(ICCNT 2011)(2011第三届IEEE计算机与网络技术国际会议)

会议地点: 太原

会议语种:英文

页码: 628-632

在线出版日期: 2011-02-26（万方平台首次上网日期，不代表论文的发表时间）