Feature selection is always beneficial to the field like Intrusion Detection, where vast amount of features extracted from network traffic needs to be analysed. All features extracted are not informative and some of them are redundant also. We investigated the performance of three feature selection algorithms Chi-square, Information Gain based and Correlation based with Naive Bayes (NB) and Decision Table Majority Classifier. Empirical results show that significant feature selection can help to design an IDS that is lightweight, efficient and effective for real world detection systems.