多特征的网络钓鱼检测算法研究

摘要

Phishing or fraudulent URL attack is an attack pattern that cannot be stopped on the Internet.For such attacks,a multi-feature detection method based on machine learning is proposed.This method analyzed the website URL features,crawl the Web pages and analyzed their interaction behaviors.The comprehensive features of the phishing websites were obtained.Two kinds of machine learning algorithms such as random forest and support vector machine were used to classify and compared the detection accuracy.The experimental results show that the proposed algorithm achieves an average detection accuracy of 98%on the site dataset consisting of PhishTank and DMOZ.%网络钓鱼(Phishing)或欺诈URL攻击是互联网上无法杜绝的攻击模式.针对这类攻击,提出基于机器学习的多特征检测方法.该方法分析网站URL特征,抓取Web页面,并对其交互行为进行分析,得到该钓鱼网站的综合特征.采用随机森林和支持向量机两种机器学习算法进行分类,并比较其检测精度.实验结果表明,所提算法在由PhishTank和DMOZ构成的网站数据集上,平均能达到98%的检测精度.