Malware Detection In Url Using Machine Learning Approach

Abstract

The most common technique to host fraudulent or harmful content, such as spam, malicious ads, etc., is using a Uniform Resource Locator (URL). Malicious URLs hold harmful contents that cause loss of information, malware installation, and monetary loss of the victims. Hence, it is necessary to detect such URLs and take some action on such threats. Earlier, one database is maintained to blacklist such URLs and a URL is compared with the available database of blacklisted URLs. If the URL is found in the database then the browser considers the URL suspicious and blocks it. But, this method is ineffective in finding newly discovered URLs. By suggesting a solution based on machine learning, this problem can be resolved. This research aims to investigate how machine-learning techniques can be used to identify harmful URLs. © 2023, Springer Nature Switzerland AG.