摘要
分析目前基于邮件内容过滤技术存在的缺陷,根据垃圾邮件的大量发送和不请自来的行为特征,提出了一种垃圾邮件行为过滤技术.该技术将各邮件服务器组成一个垃圾邮件协作过滤网络,邮件服务器发送邮件时判断其发送行为,然后将发送行为信息加密;在MTA会话阶段,接收方先解密邮件发送行为信息,然后将不请自来的群发行为垃圾邮件进行过滤.实验结果表明,该技术在MTA会话通信阶段能过滤大量不请自来的垃圾邮件,具有较高的准确率和查全率,处理速度也较快,节省大量的网络资源,具有良好的过滤性能.
Analysis of the shortcomings existed in the techniques of filtering based on the content of email,this paper proposed a spam behavior filtering technology（ SBFT）according to the sending behavior characteristics of spam. A spam collaborative filtering network is structured between mail servers.Mail servers judge their sending behavior when they send messages,and encrypt the information of sending behavior.In the process of MTA communication,mail receivers first decrypt the information of sending behavior,then a large number of uninvited spam are filtered.The experiment result shows that the SBFT can filter a large number of uninvited spam in the process of MTA communication,have higher accuracy and recall,speed up the process,save a lot of network resources,and have well filtering capability.
出处
《南华大学学报（自然科学版）》
2016年第4期94-99,共6页
Journal of University of South China：Science and Technology
基金
湖南省教育厅科学研究重点项目(14A121)
