KNOWLEDGE-BASED APPROACH TO DETECT POTENTIALLY RISKY WEBSITES

Abstract-Malicious website causes huge money losses and irreparable damage for companies and particulars. To face this situation, governments have approved multiple law projects. The first component is a previously built knowledge base and the second one complements the system with a binary classifier. In this project, we describe an approach to this problem based on automated URL classifications, using statistical methods. The proposed system uses the logistic regression and host-based properties of malicious website URLs. These methods are highly predictive models be extracting and automatically analyzing features of suspicious URLs. This program will predict the malicious website as a CSV file in the database and that risky website information will be sent to the cyber security department by using SMTP protocol.