Browsafe: Approach to Enhance Trustworthiness for Safe and Secure Web Browsing

The quick growth in internet usage has brought more users in contact with web-based attacks, including phishing and malicious browser extensions, that can break their trust and security. The continuity of human engagement is last and injecting malwares to the user’s system is rapid with unsecured sites. These attacks exploit human issues and technical limitations, and get around a multitude of security measures, including signature filters. Machine learning (ML) has proved to be a revolutionary method to address such emerging attacks using pattern recognition, predictive modeling, and adaptive learning. The work proposed here discusses the application of ML methods to detection of malicious web content, including three generic themes: phishing web sites, scam mail, and insecure browser extensions. The prediction accuracy of phishing websites and attack type is enhanced to 95.35%.