Web attacks such as spamming, phishing, and malware are common on the Internet. Unsuspecting users can become victims, affecting commercial, financial, and social sites. Features including lexical, host-based, content-based, DNS, and popularity are used to generate feature representations of URLs. This research develops a multi-class classification model to categorize URLs as potential threats to system security by combining multiple features to achieve an optimal machine learning model. © 2022 Elsevier B.V., All rights reserved.