An Implemention of a Mechanism for Malicious URLs Detection

The users accessing the internet is increasing day by day, therefore, there is also a significant increase in criminal activities. Therefore, to check the credibility of website user uses traditional blacklist approach. Most antivirus, spam filter and intrusion detection system use a Blacklisting Method. However, this method is inconsequential because the behaviour of the website changes frequently. Hence, it is difficult to differentiate between the site which is benign or malicious. In our paper, we proposed a machine learning approach for detecting malicious URLs in place of the traditional blacklisting approach. In the proposed paper, we compared different supervised machine learning algorithms namely Random Forest, K-Nearest Neighbor, and Support Vector Machine and Decision Tree. The Artificial Neural Network from the domain of deep learning would also be used along with these classification algorithms. The dataset is going to have multiple features and we will use the most significant and impactful features to train the classifier model. All these algorithms are binary classifier accustomed to building a predictive model to detect a high number of malicious URLs. Since many people use Google Chrome browser most frequently hence, we plan to create a chrome extension which will give result whether the site is malicious or not.