查看文章

The widespread adoption of the World Wide Web (WWW) has brought about a monumental transition toward e-commerce, online banking, and social media. This popularity has presented attackers with newer opportunities to scam the unsuspecting—malicious URLs are among the most common forms of attack. These URLs host unsolicited content and perpetrate cybercrimes. Hence classifying a malicious URL from a benign URL is crucial to enable a secure browsing experience. Blacklists have traditionally been used to classify URLs, however, blacklists are not exhaustive and do not perform well against unknown URLs. This necessitates the use of machine learning/deep learning as they improve the generality of the solution. In this paper, we employ a novel feature extraction algorithm using ‘urllib.parse’, ‘tld’, and ‘re’ libraries to extract static and dynamic lexical features from the URL text. IPv4 and IPv6 address …