多阶段聚类—朴素贝叶斯的异常检测  

Anomaly detection based on the multi-phase clustering and naive bayes

作　　者：姜立标[1] 马乐[2] 余建伟[2] 刘永花[1] 

机构地区：[1]哈尔滨工业大学(威海)汽车工程学院 [2]哈尔滨工业大学(威海)计算机科学与技术学院,山东威海264209

出　　处：《重庆大学学报（自然科学版）》2009年第8期983-986,共4页Journal of Chongqing University

基　　金：山东省自然科学基金资助项目(Y2007G19);哈尔滨工业大学(威海)研究基金资助项目(HIT(WH)ZB200813)

摘　　要：针对异常检测手段用来标定数据集中明显的不同于其他数据的对象,提出多阶段聚类旨在解决噪声数据的引入和缺失属性样本的处理,并改变传统的贝叶斯分类的被动学习为主动学习的方式来构造性能优越的分类器。在数据预处理阶段,利用密度聚类滤去噪声点,密度聚类的输出作为下一阶段的K-means聚类的输入,提高了K-means的分类准确率。K-means负责对缺失属性的样本进行处理。在分类阶段,利用adaboost学习算法优化朴素贝叶斯分类器,使其获得较好的分类效果。Anomaly detection method was used for calibration data concentration significantly different from other data objects. In this paper, the multi-phase clustering aimed at resolving the import of noise data and the lack of the attributive sample, and changing the traditional passive learning of bayes for active learning ways to structure the superior performance classifier. In the pre-processing stage, a clustering algorithm based on density is introduced to handle noise data. And the output of the density-based clustering algorithm can be used as the input of K-means, which responsible for handling the training samples with absent values. At classification time, we introduce adaboost algorithm into naive bayes to generate a more effective classifier.

关 键 词：聚类 朴素贝叶斯 主动学习 K—means算法 

分 类 号：TP301[自动化与计算机技术—计算机系统结构]