A Similarity-Based Framework for Detecting Phishing Websites

Phishing, which is categorized as social engineering, is turning to a hotbed for modern fraudulence. Since the first phishing attack in the 90s, many solutions were suggested. List-based methods are a common solution, mostly in commercial methods; although they are prone to zero-day attacks. Researchers introduced many methods to detect the phishing by its properties. In this paper a novel framework is introduced, which is based on textual similarity of the phishing website to the original one; this scans the possible meant-to-visit websites and by calculating the similarity, decides the destination. This framework was implemented and tested against real-world websites and provides decent accuracy. This framework not only detects the phishing can guide the user to the original website.