<![CDATA[URLs are very much on the network of computer systems. Moreover, nowadays all activities use an online system. Starting from social media, and marketplaces to group chat applications. An early prevention system from malicious URL attacks is needed to counteract the large number of URLs circulating in the online system. Previously detection of malicious URLs based on blacklisting and UURLs are very much on the network of computer systems. Moreover, nowadays all activities use an online system. Starting from social media, marketplaces to group chat applications. An early prevention system from malicious URL attacks is needed to counteract the large number of URLs circulating in the online system. Previously, malicious URL detection based on Blacklisting and Heuristic URLs could not recognize the new type of malicious URL without first being analyzed. For this reason, a technique is needed to detect malicious URLs using machine learning. The lack of machine learning in the detection of malicious URLs is that it is not 100% able to detect malicious URLs precisely. This study will use an improved random forest approach with a random forest as a classifier to detect malicious URLs. Improved Random Forest is a Random Forest that is used using evaluator features and filter instances to improve the accuracy of ordinary random forests. This study concluded that both methods of improved random forest and ordinary random forest have an accuracy value above 98%.]]>
Copyrights © 2023
