Phishing Website Detection Using Machine Learning

Abstract

Phishing attacks have become increasingly sophisticated, with criminals creating convincing clones of legitimate websites to steal sensitive information from unsuspecting users. Phishing websites are cleverly disguised platforms designed to trick people into sharing sensitive information like passwords,bankdetails, andpersonaldata.Traditional detection methods such as blacklists and rule-based systems often fall short, especially against newer and more advanced phishing techniques. Current systems predominantly use Random Forest and Decision Tree algorithms, which often suffer from low latency and lack of a user-friendly interface This projectintroduces a smart and proactive approach to phishing detection using the XGBoost machine learning algorithm. By analyzing features from website URLs and related data, our system takes advantage of XGBoost’s accuracy and speed to reliably spot phishing threats.