Machine Learning for Multiple Stage Phishing URL Prediction

Abstract

Phishing is a fraudulent process and a form of cybercrime where an attacker tries to obtain sensitive information for malicious use. A phisher uses social engineering and technical deception to fetch private information from the web user. Previous Machine Learning (ML) approaches have been used to detect whether URLs are valid, or invalid. The purpose of this work is to detect, or predict, the three stages of Phishing URLs starting with valid, not enough info and invalid URLs. We will investigate different potential models that are trained by Machine Learning algorithms and find out which of these models has better accuracy.