عنوان مقاله [English]
Information of the enemy's military and security affairs is always one of the important parameters which is considering for defense. One of the most important sources of information is intrusion to computer networks and data base of defense system. Hence, these networks are under constant cyber-attacks of enemies. Therefore their protection is vital. Nowadays the use of intrusion detection systems to protect systems has been increased significantly. One branch of computer network intrusion detection is anomaly detection methods in data network. In order to detect the network intrusion, these methods compare the observed condition with the normal one to distinguish the differences in the incidence of acute attacks. Behavioral and relationship status, number of regular subscribers and their typical applications and routine hardware and software relations are of these kinds. In this paper, the Local Outlier Factor as one of the methods based on machine learning used for network intrusion detection, has been introduced.