An Efficient Approach for URL Phishing Attack Detection using Random Forest Algorithm
Phishing is a cyber-attack where attackers aim to steal user’s personal information, login credentials and passwords, bank account details, location, etc., from naive internet users. Phishing attacks are the leading cause to information theft and other financial information theft. Both companies and daily users of the internet are affected by this malicious practice which illegally steals user information. We propose a machine learning approach to detect online phishing attacks using Uniform Resource Locator (URL) features. In this system, we have considered about 12 URL attributes to determine whether a website is benign, spam or malicious. The system is trained using about 4000 phishing and legitimate URLs using SVM and Random forest data classifiers. Our system is able to detect the nature of the website of up to 90% accuracy using SVM data classifier.