Abstract

The rising use of web services has led to a massive exposure of users to malicious websites that exploit unsafe permissions, insecure scripts, and domain features. Conventional web browser safety features and technologies have long depended extensively on static blacklisting or manual analysis, which have proved inefficient against the constantly evolving nature of web threats like phishing attacks, rogue domains, and malicious JavaScript injections. This literature review examines existing research efforts from 2020 to 2025 regarding the development of intelligent web browser safety systems that utilize machine learning classifiers for the automated identification of malicious websites based on multi-feature analysis. Upon examining the literature available regarding the identification of dubious JavaScript code, domain aging verification, the evaluation of HTTPS and SSL certificate information, and the whitelisting of trusted websites, the research indicates the importance of using a package of multiple intelligent web browser safety features together with machine learning classifiers.