A novel hybrid technique using fuzzy logic, neural networks and genetic algorithm for intrusion detection system

Abstract

As the number of people who use the internet continues to rapidly increase, security has become an increasingly important concern in the online world. Many researchers in the past have developed detection systems to identify and detect intruders using data mining. These systems can be found in use today. However, the existing methods had the disadvantages in terms of detection accuracy and time overhead. To enhance the IDS detection accuracy and reduces the required time a novel intrusion detection system is proposed that will ensure the safety of data communication by locating any unauthorized users and efficiently identifying any unwanted visitors to wireless networks. In this paper, a hybrid algorithm is proposed for the removal of uncertainty and the prediction of outcomes. Fuzzy logic is utilized in the process of removing uncertainty, whereas neural networks are utilized for the purpose of prediction. The Genetic Algorithm is utilized in order to effect improvements in the accuracy of prediction results. Experiments have been carried out in order to evaluate the proposed intrusion detection system, which has an overall detection accuracy of 99.12 %. The proposed model's performance has been put to evaluation utilizing tenfold cross-validation, which has been completed.