Eighth International Conference on Document Analysis and Recognition (ICDAR'05) (2005)
Aug. 31, 2005 to Sept. 1, 2005
Liu Wenyin , Dept. of Computer Science, City University of Hong Kong
Guanglin Huang , Translation & Linguistics City University of Hong Kong
Liu Xiaoyue , Translation & Linguistics City University of Hong Kong
Xiaotie Deng , Dept. of Computer Science, City University of Hong Kong
Zhang Min , Tsinghua Univ.
An approach to detection of phishing webpages based on visual similarity is proposed, which can be utilized as a part of an enterprise solution to anti-phishing. A legitimate webpage owner can use this approach to search the Web for suspicious webpages which are visually similar to the true webpage. The approach first decomposes the webpages into salient (visually distinguishable) block regions. The visual similarity between two webpages is then evaluated in three metrics: block level similarity, layout similarity, and overall style similarity. A webpage is reported as a phishing suspect if any of them (with regards to the true one) is higher than its corresponding preset threshold. Preliminary experiments show that the approach can successfully detect those phishing webpages with few false alarms at a speed adequate for online application.
Anti-Phishing, Web document analysis, Information filtering
Z. Min, L. Wenyin, X. Deng, G. Huang and L. Xiaoyue, "Phishing Webpage Detection," Eighth International Conference on Document Analysis and Recognition (ICDAR'05)(ICDAR), Seoul, Korea, 2005, pp. 560-564.