Quantum-inspired ant lion optimized hybrid k-means for cluster analysis and intrusion detection

Abstract

Intrusion detection maintains network security by detecting intrusion behaviors. There are many clustering algorithms that can be used directly for intrusion detection. K-means is a simple and efficient method used in data clustering. However, k-means has a tendency to converge to local optima and depends on the initial value of cluster centers. Therefore, we present an efficient hybrid clustering algorithm referred to as QALO-K, whereby, we combine k-means with quantum-inspired ant lion optimized. This algorithm combines the advantages of quantum computing and swarm intelligence algorithms to improve the k-means algorithm and make the k-means algorithm converge towards the global optimal direction. Our proposed algorithm is tested on several standard datasets from UCI Machine Learning Repository for cluster analysis and its performance is compared with other well-known algorithms. The proposed method was applied on KDD Cup 99 large datasets for intrusion detection. The simulation results infer that the proposed algorithms can be efficiently used for data clustering and intrusion detection.