Intrusion detection is a major research problem in network security. Due to the nonlinear nature of the intrusion attempts, unpredictable behavior of the network traffic and the large number of features in the problem space, intrusion detection systems represent a complicated problem area. Choosing effective and key features for intrusion detection is a very important topic in information security. The purpose of this study is to identify important features in building an intrusion detection system such that they are computationally efficient and effective. To improve the performance of intrusion detection system, this paper proposes an intrusion detection system that its features are optimally selected using ant colony optimization. The proposed method is easily implemented and has a low computational complexity due to use of a simplified feature set for the classification. The extensive experimental results on the KDD Cup 99 and NSL-KDD intrusion detection benchmark data sets demonstrate that the proposed method outperforms previous approaches, providing higher accuracy in detecting intrusion attempts and lower false alarm with reduced number of features.