Safe Browse Guardian

Abstract

Malicious websites offer unsolicited content and lure unsuspecting users into committing fraud. A quick investigation and action on such threats is essential. However, blacklists detect Uniform Resource Locators (URLs) which is another malicious invention. Blacklisting and machine learning techniques using feature extraction were explored in this framework to improve common malicious URL detectors. Blacklist feature takes less processing time and also relies on external data (list of malicious websites) in detecting malicious websites while feature removal method takes more time and does not rely on external data so in detecting new malicious websites through web browser extensions The system was implemented by Several malicious and non-optimal communication protocols were used to test the system. The system has three main layers: users, web extensions, and databases. The web browser extension layer uses two methods (Blacklist feature and feature extraction) to detect highly malicious websites. The performance of the malicious website detection system using blacklist and feature removal means that it provides a robust, secure and easy way to detect malicious websites in real time.