Intrusion detection using self-adaptive Genetic algorithm

Recently with the rapid development of Internet,computer systems are facing increased of security threats.notwithstanding different protection methods, it isapproximately impossible to have a completely secured system.Therefore, intrusion detection is becoming an very importanttechnology that monitors network traffic and identifies networkintrusions such as anomalous network behaviors, unauthorizednetwork access, and malicious attacks to computer systems.different methods have been proposed so far based on theGenetic Algorithms to detect computer network attacks. thispaper presents a new method based on the self-adaptive Geneticalgorithm to solve the problem of intrusion to network; usingthis method creates new rules. the support-confidenceframework is utilized as fitness function to identify the quality ofeach rule.The generated rules are then used to detect or classifynetwork intrusions. evaluation results show that unlike thecommon genetic algorithm, the proposed method presents amore Accuracy rate, more reliability, and also fasterconvergence based on the features of KDD dataset.