基于特征选择和支持向量机的异常检测方法  

Anomaly detection method based on feature selection and support vector machine

作　　者：张昭[1] 张润莲[1] 蒋晓鸽[1] 曾兵[2] 

机构地区：[1]桂林电子科技大学信息与通信学院,广西桂林541004 [2]保密通信重点实验室,四川成都610041

出　　处：《计算机工程与设计》2013年第9期3046-3049,3162,共5页Computer Engineering and Design

基　　金：保密通信重点实验室基金项目(9140C110404110C1106);广西自然科学基金项目(2012GXNSFAA053224);广西研究生教育创新计划基金项目(2010105950810M18);广西教育厅基金项目(201010LX156;CD10066X)

摘　　要：针对异常检测系统虚警率高、检测率低以及冗余特征对检测系统造成负担的问题,提出一种基于特征选择和支持向量机相结合的异常检测方法。该方法通过构造一种基于分类模型分类准确率计算的特征选择算法,筛选出能够获得分类准确率最高的特征组合,并与支持向量机分类算法相结合,实现数据的异常检测。仿真测试结果表明,该方法具有较高的检测准确率和较低的检测时间,并通过去除噪声特征,降低了系统的数据处理难度。There are many problems in anomaly detection system, such as high false alarm rate, low detection rate, and bad in- fluence of redundant features. To solve the problems, an anomaly detection method based on the feature selection and support vector machine is proposed. In order to select the feature sets with the highest classification accuracy, a feature selection algo- rithm is designed, which can compute classification accuracy of the constructed classification model based on features. By eombi ning the selected feature sets and support vector machine, the anomaly detection method can detect and identify whether the data is normal or not. The simulation test results show that the method can improve the detection accuracy, reduce the detection time, and reduce the difficulty processing data by removing noise features.

关 键 词：异常检测 特征选择 支持向量机 分类准确率 分类模型 

分 类 号：TP393.08[自动化与计算机技术—计算机应用技术]