An Intelligent Phishing Website Detection System Using Machine Learning Algorithms

Phishing attacks are one of the most popular cyber threats, where attackers design a copy of a genuine website to steal confidential information like usernames, passwords, and bank account details. It is quite difficult for common users to identify genuine and phishing websites, resulting in loss of money and data breaches. This project work presents a Machine Learning-based phishing website detection system that examines the URL of the website as well as its HTML structure. It identifies features like URL length, number of links, forms, scripts and external resources. Different algorithms like Random Forest, Support Vector Machine (SVM), Decision Tree, Naive Bayes, and K-Nearest Neighbours (KNN) are used and compared. Among them, Random Forest gave the best accuracy. The system is automated, accurate, and able to identify new phishing websites, thus improving the security of online users.