计算机科学 ›› 2007, Vol. 34 ›› Issue (9): 93-94.

• 软件工程与数据库技术 • 上一篇    下一篇

基于支持向量机的邮件过滤

  

  • 出版日期:2018-11-16 发布日期:2018-11-16

  • Online:2018-11-16 Published:2018-11-16

摘要: 随着万维网的兴起和电子邮件的快速发展,大量的垃圾电子邮件也随之在互联网上泛滥.电子邮件过滤就是要在大量邮件中过滤出垃圾邮件,帮助用户找到所需的邮件。本文讨论了基于机器学习方法实现垃圾邮件过滤的原理,提出一种改进的基于支持向量机的邮件过滤技术,该方法使用互信息度函数,结合Z-测试进行特征选择,使用SVM(支持向量机)构造分类超平面来进行文本分类。实验表明,提高了中文邮件过滤的准确性。

关键词: 支持向量机 文本分类 邮件过滤 互信息Z-测试

Abstract: With the rapid explosion of unsolicited bulk e-mail, also known as "spam", has generated a great need for reliable anti-spare e-mail filters. Mail filtering technique is to used to detect the spare and find what's useful. We discuss the theorem to impleme

Key words: SVM, Text classification, Spam filter, Mutual information, Z-test

No related articles found!
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!