摘要
钓鱼检测方式一般只是通过比较网页间特征的相似度判断钓鱼网站,容易被攻击者根据特征提取过程反检测。因此,提出依据网页之间的关系来检测钓鱼网站,通过设立网页集合、结合钓鱼网站与其他网站间的链接关系改进TrustRank算法来检测钓鱼网站。实验证明,改进的TrustRank算法可以将钓鱼网站的信任值集中在一个范围内,并且与其他方法相比具有较低的误判率、漏判率和较高的速度,可以很好地检测钓鱼网站。
Anti-phishing methods just generally detect phishing sites based on comparing the similarity of pages fea- tures, which could be anti-detected when attackers are clear about the extraction process of features. Therefore, a method based on web pages relationship was proposed to detect phishing websites. According to this method, phish- ing websites were detected by an improved TrustRank algorithm which set up a collection of web pages and used the link relationship between phishing websites and other websites. Experiments show that the improved TrustRank algo- rithm can focus the trust values of phishing websites on some range with the lower false positive rate, the lower false negative rate and the higher speed compared to other methods.
作者
韩浩
刘博文
林果园
HAN Hao, LIU Bowen, LIN Guoyuan(School of Computer Science and Technology, China University of Mining and Technology, Xuzhou 221116, Chin)
出处
《电信科学》
2018年第3期86-94,共9页
Telecommunications Science
基金
江苏省产学研前瞻性联合研究项目(No.BY2016026-04)~~
