计算机工程与应用 ›› 2007, Vol. 43 ›› Issue (18): 177-180.

• 数据库与信息处理 • 上一篇    下一篇

一种面向电子邮件分类的特征值处理方法

邹 娟,周经野,邓 成,陈 静   

  1. 湘潭大学 信息工程学院,湖南 湘潭 411105
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2007-06-21 发布日期:2007-06-21
  • 通讯作者: 邹 娟

Characteristic value extractive method for e-mail categorization

ZOU Juan,ZHOU Jing-ye,DENG Cheng,CHEN Jing   

  1. Information Engineering College,Xiangtan University,Xiangtan,Hunan 411105,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2007-06-21 Published:2007-06-21
  • Contact: ZOU Juan

摘要: 利用电子邮件的特点提出了一种面向电子邮件分类处理的特征值提取方法。本方法根据电子邮件文法随意性的特点,利用模糊集合对其同义词和多义现象都进行了处理,使得所得到的特征值能更好的契合文本的特点。通过与其它特征值提取方法的比较实验,以及在不同分类算法中应用实验结果都证明文中提出的特征值提取方法能够提高电子邮件分类处理的正确率,并达到有效降低特征向量维数的目的。

Abstract: Based on the e-mail,we propose a method of characteristic value extraction in this paper.We process synonym and polysemant that use of the fuzzy theory.So the characteristic value that using the method of characteristic value extraction in this paper can the better denote text characteristic.Finally,we present the results of the experiments comparing with other characteristic value extraction method and the results of applying the method in the different classified algorithm,which illustrate that the method in this paper improve the correct rate of e-mail categorization and reduce the dimensions effectively.