Ying Pan
Singapore Management University
Singapore

Xuhua Ding
Singapore Management University
Singapore

Though many anti-phishing schemes have been proposed recently in
literature, the phishing attacks become an even bigger threat to
cyberspace security. One of the main reasons is that phishing
attackers have the adaptability to change their tactics with
little cost. In this paper, we propose a novel approach, which is
independent of any specific phishing implementation. Our idea is
to examine the anomaly in web pages, in particular, the
discrepancy between a web site's identity and its structural
features and HTTP transactions. It demands neither user expertise
nor prior knowledge of the website. The evasion of our phishing
detection entails prohibitively high cost to the adversary. As
shown by the experiments, our phishing detector functions with low
miss rate and low false-positive rate.

Keywords: Anti-Phishing, Security, Anomaly detection

Read Paper (in PDF)