Identification of Potential Risks in Websites by Using Knowledge-based Machine Learning

The identification of wide variety of potential risks such as security vulnerabilities, privacy breaches, and malicious content risks is crucial for safeguarding users and maintaining the integrity of online platforms. Unsupervised web page categorization is the problem of organizing web pages within a website into clusters, each of which has a set of web pages that can be classified using a different class. The current proposals for web page classification do not meet several requirements that would make them appropriate for enterprise web information integration. To fulfil the requirement of the authentication, unsupervised learning techniques can be utilized which does not need pre-classified pages for training data. The regular operations of website should be avoided with lightweight crawling and utilized the features from outside the pages. The current research presents a novel technique for the classification of webpages based on URL classifiers. The URL patterns have been classified for checking the authenticity of the websites. So, to validate the authenticity of the websites, several websites have been tested.