Fake It Till You Get Caught Phishing Website Detection Using Machine Learning Algorithms

With the increasing reliance on digital platforms, phishing attacks have become a significant cybersecurity threat, exploiting user vulnerabilities through deceptive websites that mimic legitimate services. Traditional security measures like antivirus programs and firewalls often fail to prevent such attacks, especially zero-day exploits. This paper proposes a machine learning-based phishing detection system, employing four different algorithms to analyze URLs and leveraging  distinct datasets for comprehensive evaluation. The study compares the performance of various machine learning techniques in classifying phishing websites, demonstrating high accuracy in detection. The results highlight the effectiveness of anomaly-based approaches in identifying phishing attempts, offering a dynamic solution to enhance cybersecurity. The findings contribute to the ongoing efforts to mitigate phishing risks, providing a robust framework for real-world applications.