Comparison of Machine Learning Techniques for Phishing Detection

Abstract

Phishing is the act of stealing personal data through a false Web page. Users are asked by a supposedly legitimate company to enter their private information in order to verify their identity. This research work compared different machine learning techniques and determined that the random forest technique is the most effective one when the characteristics of the pages have an exact value, and the decision tree is the most effective one when the characteristics of the pages have been analyzed and a classification has been determined based on such characteristics.