计算机工程与应用 ›› 2008, Vol. 44 ›› Issue (4): 135-137.

• 网络、通信与安全 • 上一篇    下一篇

协同分类器及其在邮件过滤中的应用

路 梅1,2,叶澄清2   

  1. 1.徐州师范大学 计算机科学与技术学院,江苏 徐州 221116
    2.浙江大学 计算机科学与技术学院,杭州 310027
  • 收稿日期:2007-05-28 修回日期:2007-08-03 出版日期:2008-02-01 发布日期:2008-02-01
  • 通讯作者: 路 梅

Anti-spam technique with SVM and K-NN cooperation method

LU Mei1,2,YE Cheng-qing2   

  1. 1.College of Computer Science and Technology,Xuzhou Normal University,Xuzhou,Jiangsu 221116,China
    2.College of Computer Science and Technology,Zhejiang University,Hangzhou 310027,China
  • Received:2007-05-28 Revised:2007-08-03 Online:2008-02-01 Published:2008-02-01
  • Contact: LU Mei

摘要: 提出了一种基于支持向量机的改进分类方法.该方法将特征空间分类超平面附近的样本分类,交由特征空间和样本空间中的K-近邻集体投票表决。方法应用于垃圾邮件的过滤之中,邮件合法性误判发生的概率可被有效降低。最后通过垃圾邮件过滤实例验证了该方法的有效性。

关键词: 垃圾邮件, 智能过滤, 支持向量机

Abstract: This paper presents an improved classific algorithm which combines the Support Vector Machine with the K-Nearest Neighbor (K-NN).In this algorithm,the classification of samples in the critical zone is decided by the voting of K-NN set in the feature space and the sample space.Using the method,the filtering error rate will decrease obviously.Finally,lots of examples show that the method is feasible.

Key words: spam, intelligent filtering, Support Vector Machine