Phishing URL Detecting Website Using Machine Learning

Computer Engineering,Machine Learning

Phishing attacks remain a significant concern for computer system defenders, often serving as the initial phase in a multi-stage attack. Despite significant advancements in phishing detection, certain phishing emails can circumvent filters by altering the message's structure and meaning. To address this issue, we implemented a machine learning classifier on a large corpus of legitimate and phishing emails. Our system, SAFEPC (Semi-Automated Feature Generation for Phish Classification), extracts features, some of which are elevated to higher-level features, to outsmart conventional phishing email detection strategies. To evaluate SAFE-PC, we obtained a substantial corpus of phishing emails from a tier-1 university's central IT organization. Our implementation of SAFE-PC on the dataset revealed previously unknown insights into phishing campaigns targeted at university users. SAFE-PC surpasses a state-of-the-art email filtering tool, detecting more than 70% of phishing emails.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.