计算机工程与应用 ›› 2010, Vol. 46 ›› Issue (19): 94-96.DOI: 10.3778/j.issn.1002-8331.2010.19.027

• 网络、通信、安全 • 上一篇    下一篇

基于C-SVM和KPCA的垃圾邮件检测研究

秦玉平1,耿姝1,孙宗宝2   

  1. 1.渤海大学信息科学与工程学院,辽宁锦州121000
    2.哈尔滨理工大学计算机科学与技术学院,哈尔滨150080
  • 收稿日期:2008-12-17 修回日期:2009-03-06 出版日期:2010-07-01 发布日期:2010-07-01
  • 通讯作者: 秦玉平

Spam detection approach based on Support Vector Machine and Kernel Principal- Component Analysis

QIN Yu-ping1,GENG Shu1,SUN Zong-bao2   

  1. 1.College of Information Science and Engineering,Bohai University,Jinzhou,Liaoning 121000,China
    2.Computer Science and Technology College,Harbin University of Technology,Harbin 150080,China
  • Received:2008-12-17 Revised:2009-03-06 Online:2010-07-01 Published:2010-07-01
  • Contact: QIN Yu-ping

摘要: 现有的垃圾邮件检测算法存在小样本情况下泛化能力差的问题。提出了利用核主成分分析和支持向量机结合进行垃圾邮件检测的方法。与传统算法相比,该方法与邮件异构有很高的检测率、更强的泛化能力和更高的检测效率。实验证明了方法的实用性和高效性。

Abstract: Current spam filtering has poor generalization ability as given less priority knowledge.The KPCA and SVM are
adopted to implement spam filtering detection.Compared with traditional algorithms this method can achieve higher detection
rate and better generalization,and decrease time of performance.The experiment on data set shows the effectiveness and excellent
performance of the method.

中图分类号: