Many advances in computer systems and IT infrastructures increases the risks associated with the use of these technologies. Specifically, intrusion into computer systems by unauthorized users is a growing problem and it is very challenging to detect. Intrusion detection technologies are therefore becoming extremely important to improve the overall security of computer systems. In the past decades, most of the intrusion detection systems designed suffer from the problem of high false negative and low efficiency rate. A powerful intrusion detection system (IDS) should be implemented to solve these issues and it is necessary to collect, reduce and analysis the data automatically. The integration of machine learning and artificial intelligence techniques serves this purpose in this paper. A use of particle swarm optimization (PSO) selects the optimal number of clusters and the integration of k-means based artificial neural network (ANN) achieves maximum efficiency when the number of clusters selected optimally. The proposed IDS are t bested with NSL-KD dataset and the experiment result shows the significance of the proposed IDS.