Detecting Phishing Websites Using Machine Learning

Phishing is one of the most widespread forms of cybercrime, tricking users into divulging sensitive information through deceptive websites. Traditional methods for detecting phishing sites rely heavily on blacklists, which are often outdated. This study applies machine learning techniques to detect phishing websites more effectively. Using a dataset of 11,000 website features and labels, various machine learning algorithms—including Decision Tree, Random Forest, and Support Vector Machine—were trained and evaluated. A structured questionnaire was distributed to 350 respondents to assess awareness and experiences related to phishing. The research achieved an accuracy of 97.1% using the Random Forest model. The survey findings provided supporting evidence on user awareness and exposure to phishing threats. The results suggest that machine learning models are capable of identifying phishing sites with high precision and could be integrated into modern web browsers and security tools.