期刊论文

A Spam Detection Model Based on Simulated Annealing Algorithm and Sending Behavior

中文

南华大学学报(自然科学版)

1673-0062

2017

31

1

77-80

10.19431/j.cnki.1673-0062.2017.01.016

南华大学研究生科研创新项目（2016XCX16）；

本校为第一机构

垃圾邮件给当今人们的生活带来严重的负面影响.虽然已经有很多过滤方法,但大多存在一定的不足之处,如检测时间长、召回率低等问题.本文提出了一种基于模拟退火算法和发送行为的垃圾邮件检测模型,旨在弥补已有检测方法的不足.模拟退火算法可能找到全局最优解,且收敛性强;而基于发送行为的垃圾邮件检测技术能显著提高服务器处理垃圾邮件的速度.本文尝试将二者相结合,以提高垃圾邮件的召回率及服务器处理能力.通过实验结果可以看出,该方法在垃圾邮件的召回率上有较大提升,较适于部署在小型邮件服务器上.

Spam has brought serious negative impact on people＇s life today. Many filtering methods have been proposed, but less perfect. In this essay, a detection model based on simulated annealing algorithm and sending behavior is presented to cover some shortcomings of the existing methods.Simulated annealing algorithm can find the global optimal solution, and the convergence is strong, and the spam detection technology based on sending behavior can significantly improve the speed of the server to deal with spam. Through experiments,the recall rate of...