Enhanced relevant feature selection model for intrusion detection systems

With the increased amount of network threats and intrusions, finding an efficient and reliable defence measure has a great focus as a research field. Intrusion detection systems (IDSs) have been widely deployed as effective defence measure for existing networks. IDSs detect anomalies based on features extracted from network traffic. Network traffic has many features to measure. The problem is that with the huge amount of network traffic we can measure many irrelevant features. These irrelevant features usually affect the performance of detection rate and consume the IDSs resources. In this paper, we proposed an enhanced model to increase attacks detection accuracy and to improve overall system performance. We measured the performance of the proposed model and verified its effectiveness and feasibility by comparing it with nine-different models and with a model that used the 41-features dataset. The results showed that, our enhanced model could efficiently achieves high detection rate, high performance rate, low false alarm rate, and fast and reliable detection process.