Phishing Website Detection Using Machine Learning

With the rapid growth of internet usage, cybercrime has also increased, especially phishing attacks, where fake websites are created to look like real and trusted websites in order to steal users' passwords, bank details, and personal information. Traditional phishing detection methods based on fixed rules are no longer very effective because attackers keep changing their techniques. This project presents a machine-learning-based web application that can quickly check whether a website URL is safe or phishing in real time. The system analyses different features of a website, such as the URL structure, HTTPS security, domain age, and website traffic information, and then uses four machine learning algorithms — Decision Tree, Random Forest, Logistic Regression, and Support Vector Machine to classify the website. Among these, the Random Forest algorithm gave the best performance with higher accuracy and better detection results. The proposed system is lightweight, easy to use, and accessible through a web browser, making it a practical solution to improve online safety for both individuals and organisations.