Intrusion Detection Using Krill Herd Optimization Based Weighted Extreme Learning Machine

With the improvement in computer network and technology, network attacks have increased drastically, as a result network intrusion becomes an important topic to work on and to find a solution to stop these network attacks. Advancement in artificial intelligence, can be utilized to find a solution for network intrusion. In this paper we are using Krill Herd Optimization (KHO) algorithm based Weighted Extreme Learning Machine (WELM) to detect the intrusion occurring in the network. Extreme Learning Machine (ELM) randomly assigns weight for neural network which is followed by the network training activity and finally the output weight is obtained. There is a need for optimization of weights used in ELM, for this purpose we are using krill herd optimization algorithm. NSL-KDD dataset is used to compare and analyze the performance of the model proposed in this paper. The experimental results show that krill herd optimization based on WELM performed better in identifying the intrusion in the network and minimize the false positive and false negative rates.