OMUI-HML: An optimal malicious URL detection and classification using hybrid machine learning technique
Abstract
As a result of the rapid development of web attacks, many web applications are vulnerable to various security threats and network attacks.Detecting URL security is always a web security hub.Multiple web application resources can be accessed by entering a single URL or by clicking on a link in a browser.Therefore, it is important to improve the reliability and security of web applications by accurately fixing malicious URLs. This paper proposes the identification of optimal malicious URLs using Hybrid Machine Learning (OMUI-HML) technology.First, we extract few features from datasets and select optimal features using enhanced swallow swarm optimization (ESSO) algorithm. Then, a hybrid deep Q neural network (DQNN) classifier used to identify the normal and malicious. Finally, the performance of proposed OMUI-HML technique is evaluated through two different standard datasets such as phishing websites and URL reputation. The proposed OMUI-HML technique has performed well enough in terms of malicious detection .