An Improved Detection System Using Genetic Algorithm and Decision Tree

These days, intrusion detection systems (IDSs) are gaining a lot of attention as an essential component of system defense. To safeguard the network, intrusion detection systems (IDSs) gather data on network traffic from various locations inside the computer system or network. It is quite tough and takes a lot of effort to discern between typical and intrusive network traffic operations. To determine the order of the network connection intrusion, an analyst needs examine all the extensive and varied data. It thus requires a method to represent the current network traffic that can identify network intrusion. In this paper, a unique approach to utilize data mining techniques and evolutionary algorithms for machine learning to identify intrusion characteristics for IDS was developed. Classification using a generative algorithm of decision trees is the method used to produce rules. These rules can identify the features of an intrusion and then be used as preventative measures in the genetic algorithm. To prevent intrusions in addition to identifying their presence.