An improved harmony search based extreme learning machine for intrusion detection system

摘要: Abstract Intrusion Detection System (IDS) is one of the best ways to combat several cybercrimes, both at edge network and inside segments internal network. This article proposes a hybrid learning approach namely Improved Harmony Search Extreme Learning Machine based IDS (IHSELMIDS) classify NSL KDD dataset. dataset polished version its predecessor i.e., Knowledge Discovery in Databases (KDD). Since 1999, many researchers have relied on for evaluation anomaly detection system. Thus, it popularly known as KDD”99 has been used boost weights input latent biases more robust stable (ELM). In addition, generalized inverse Moore – Penrose systematically evaluate output. Additionally, address curse high dimensionality dataset, correlation-based feature selection with greedy hill climbing proposed which reduces time complexity while increases computational efficiency. A series performance measures such training testing accuracy, True Positive Rate (TPR), Negative (TNR), G-mean, F-score, False Alarm (FAR), Receiver Operating Characteristic curve (ROC) Confusion Matrix into consideration contrast examine efficiency, flexibility reliability model. The experimental result showed that IHSELMIDS outperforms all benchmark models considered this study.