Bagging eEP-based classifiers for junk mail classification

摘要：

The problem of classifying junk emails on the Internet are getting more and more attention recently.The highly skewed email datasets make it difficult to classify them correctly by using many traditional classifiers.Bagging is a method for improving the predictive power of weak classifiers,which can improve the performance of unstable classifiers.This paper uses a special kind of emerging patterns,called essential emerging pattern (eEP),to build eEP-based classifiers,and bagging them to build effective ensemble to the junk email examine,and carries out a new categorization and filtering algorithm BeEPJMC.The experiments show,the bagging methods and the combination eEP classification is a very efficient method of classification,and The classification efficiency of the algorithm BeEPJMC is higher than currently several better classification algorithm.

关键词： E-Mail Categorization Bagging Essential Emerging Patterns

作者: Yan LI

作者单位: Department of Information Engineering Henan Technical College of Construction ZhengZhou City,China

会议类型: 国际会议

会议名称: 2010 4th International Conference on Intelligent Information Techonlogy Application(第四届智能信息技术应用国际学术研讨会 IITA 2010)

会议地点: 秦皇岛

会议语种:英文

页码: 299-302

在线出版日期: 2010-11-05（万方平台首次上网日期，不代表论文的发表时间）