| 一种基于学习向量量化网络的垃圾邮件过滤方法 |
| |
| 引用本文： | 詹川,卢显良,周旭,侯孟书.一种基于学习向量量化网络的垃圾邮件过滤方法[J].计算机科学,2005,32(3):66-68. |
| |
| 作者姓名： | 詹川 卢显良 周旭 侯孟书 |
| |
| 作者单位： | 电子科技大学计算机科学与工程学院,成都,610054 |
| |
| 基金项目： | 信息产业部电子工业生产发展基金,编号，[2002]11006 |
| |
| 摘 要： | 伴随着电子邮件的广泛使用，垃圾邮件泛滥成灾，严重影响了人们正常的学习、工作和生活。拳文针对目前的垃圾邮件主要是由多种商业或政治性类别的垃圾邮件组成的特点，利用学习向量量化网络能把多个子类合并成一个复杂大类的特性，构建了一个反垃圾邮件的LVQ神经网络模型，我们对该LVQ网络模型进行了与其他算法的对比试验，试验表明它比基于贝叶斯公式算法和基于神经网络BP算法的过滤器有更好的性能。
|
| 关 键 词： | 学习向量量化网络 垃圾邮件 过滤方法 互信息 向量空间模型 |
An Anti-spam E-mail Filter Based on LVQ Network |
| |
| ZHAN Chuan,LU Xian-Liang,ZHOU Xu,HOU Meng-shu.An Anti-spam E-mail Filter Based on LVQ Network[J].Computer Science,2005,32(3):66-68. |
| |
| Authors: | ZHAN Chuan LU Xian-Liang ZHOU Xu HOU Meng-shu |
| |
| Affiliation: | ZHAN Chuan,LU Xian-Liang,ZHOU Xu,HOU Meng-Shu College of Computer Scieme and Engineering,UESTC of China,Chengdu 610054 |
| |
| Abstract: | Along with wide application of E-mail nowadays,a large number of spam E-mails flood into people's life and bring catastrophe to their study and life. This paper presents a novel anti-spam e-mail filter based-LVQ network in terms of spam e-mail's characteristic which are mainly made up of several kinds commercial or political spam emails at present. Our experiment has proved that our filter based-LVQ is superior to anti-spam email filter based-Bayes and that based-BP in total performances apparently. |
| |
| Keywords: | LVQ Anti-spam E-mail filtering Mutual information Vector space model |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
点击此处可从《计算机科学》下载全文 |
