Intrusion detection system (IDS) was designed to monitor the network activity and it identifies the normal and abnormal behavioral pattern in the network. If there was any abnormal pattern, it indicates the system is in attack by compromising the confidentiality, availability or integrity of the computer system. IDS perform three functions namely monitoring, detecting and responding for malicious activity. Experiment is based on kdd99 dataset to categorize normal and abnormal pattern. Goal of this paper is to compare three classification techniques by considering two classifiers from each technique and find out the best one based on the true positive, false positive and average accuracy.