基于粗糙集的加权朴素贝叶斯邮件过滤方法  

Weighted Nave Bayes Spam Filtering Method Based on Rough Set

作　　者：邓维斌[1,2] 王国胤[2] 洪智勇[2] 

机构地区：[1]重庆邮电大学电子商务与现代物流重点实验室,重庆400065 [2]西南交通大学信息科学与技术学院,成都610031

出　　处：《计算机科学》2011年第2期218-221,共4页Computer Science

基　　金：国家自然科学基金(60773113);重庆市自然科学基金重点项目(2008BA2017);重庆邮电大学自然科学基金(A2008-38)资助。

摘　　要：邮件过滤中有两个关键问题,一是如何选择有效的邮件特征集,二是设计较好的邮件过滤算法。在对邮件特性进行分析的基础上,综合邮件头及邮件内容的主要形象特征给出了一种新的邮件特征集提取方法。用粗糙集的信息观点度量了各属性的重要性,并以此为权重进行加权朴素贝叶斯垃圾邮件过滤,有效地解决了朴素贝叶斯分类中的条件依赖性问题。通过在中英文邮件集上的测试实验,证明了所提出的邮件过滤方法的有效性。Using a classifier based on a specific machine-learning technique to automatically filter out spam email has drawn many researchers＇ attention.In a spam filtering process,how to selecting the features of emails and how to design a good filtering algorithm are two key issues.A new method of features selecting was proposed,which include the head and the other main features of emails.Furthermore,the features＇ importance degree was measured according to information viewpoint of rough set.With it,a new weighted nave Bayes spam filtering was put forward.It can solve the conditional dependence of nave Bayes efficiently.Simulation results on two email data sets in English and Chinese respectively illustrate the efficiency of this method.

关 键 词：垃圾邮件过滤 特征选择 粗糙集 加权朴素贝叶斯 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]