Network Intrusion Detection Based on Hybrid Rice Algorithm Optimized Extreme Learning Machine

Intrusion detection is an important issue for network security. Many machine learning based detection technologies have been put forward and obtained rather good results, such as Bayesian classifiers, Extreme learning machine. However, the performance of these methods is dependent on parameters to a great extent. Aiming at learning the optimal parameters of extreme learning machine for network intrusion detection, the paper proposes a new network intrusion detection based on hybrid rice algorithm optimized extreme learning machine (HRO-ELM). First, the extreme learning machine parameters are encoded as rice gene location, and the accuracy of test results represents the fitness value of the algorithm. Then the optimal parameters of the extreme learning machine are found by simulating rice breeding behavior, and a network intrusion detection classifier is established. Finally, the KDD99 data set is used and simulation results show that the HRO-ELM improves the accuracy of network intrusion detection and can meet the requirements of network intrusion online detection.