Abstract
Phishing is a form of cybercrime. Sensitive information is collected by using forged webpages intended to mimic a legitimate webpage. Most of the sensitive information collected by these phishing websites will eventually be used to access the victims’ money. Phishing websites can mimic the content of legitimate websites by attempting to look extremely similar to them, making it hard for users to distinguish them by their appearance. In this paper, we propose a heuristic based approach to detect phishing websites. Heuristic features based on URL and literal and conceptual similarity between URL and web content were derived. Our approach achieved 96% accuracy, showing that the heuristic features selected for our work are effective at detecting various forms of phishing attacks.