Intrusion Detection System using Machine Learning Algorithms: A Comparative Study

In recent years, the extensive usage of the internet leads to an exponential increase in the volume of information exchanged among various devices and the number of new ways of network attacks. The conventional methods like firewall, which focused on the filtering of data, may not suitable to identify all types of attacks on time. For effective handling and timely identification of these types of attacks, intrusion detection systems (IDS) based on machine learning algorithms are very effective to efficiently process the large volume of data for identifying any malicious activity. Machine learning based IDS are used to analyze all network activities for any malicious behavior. The proposed paper mainly focuses on providing the analytical studies of such existing intrusion detection system. Also, this work explores the useful data sets with different existing ways to create an effective IDS using single, hybrid and ensemble machine learning algorithms. Then, the approaches in the literature have been analyzed based on different data sets and compared with aiming to provide a simple path and guidance for effective future work.