高技术通讯2012，Vol.22Issue(9)：889-896,8.DOI:10.3772/j.issn.1002-0470.2012.09.001
基于用户关系挖掘和信誉评价的垃圾邮件识别算法
A spam identification algorithm based on user relationship mining and reputation evaluation
摘要
Abstract
According to behavior-based spam identification ideas, a spam identification algorithm based on user relationship mining and reputation evaluation is proposed. First, a mail user relationship description model is established on the basis of analyzing historical communication in the mail user group. When mail reputation is evaluated, the potential user communication relationship is mined using the model proposed, and the mail communication path collections are established according to mail receiving users and mail fingerprints. The comprehensive reputation value is calculated to identify spam according to historical evaluation of mail fingerprints given by users in the path collections. When a mail receiver gives his feedback, the reputation of user and mail fingerprints is updated. The experimental results show that the proposed algorithm can identify spam correctly and effectively. The availability is considered in the design of the algorithm, which makes it work in actual spam identification systems.关键词
垃圾邮件/行为分析/用户关系挖掘/信誉评价Key words
spam, behavior analysis, user relationship mining, reputation evaluation引用本文复制引用
王巍,苘大鹏,玄世昌,杨武,邱文真..基于用户关系挖掘和信誉评价的垃圾邮件识别算法[J].高技术通讯,2012,22(9):889-896,8.基金项目
863计划(2007AA01Z473)和中央高校基本科研业务费专项资金(HEUCF100601)资助项目. （2007AA01Z473）
