Efficient intrusion detection is needed as a defense of the network system to detect the attacks over the network. A feature selection and classification based Intrusion Detection model is presented, by implementing feature selection, the dimensions of NSL-KDD data set is reduced then by applying machine learning approach, we are able to build Intrusion detection model to find attacks on system and improve the intrusion detection using the captured data. With the increasing number of new unseen attacks the purpose of this model is to develop a system for intrusion detection, and the model will be capable of detecting new and previously unseen attacks using the basic signatures and the features of known attack.