Performance Analysis of Machine Learning Algorithms in Intrusion Detection Systems

With the developing technology, need for dissemination and protection of information is becoming increasingly important. Recently, attacks on systems have increased significantly. In addition to rise in number attacks, different types pose a great threat systems. As result these institutions users suffer serious damages. At this point, Intrusion Detection Systems (IDS) very important position. The pre-detection preparation necessary reports can reduce impact threats that may be encountered future. Recent studies are carried out so as increase performance IDS. paper, classification was made using NSL-KDD dataset SVM, KNN, Bayesnet, NavieBayes, J48 Random Forest algorithms, it aimed compare classifications by WEKA. Consequently, has been reached KNN algorithm had best with an accuracy rate 98.1237 %. addition, effect increasing folds neighborhoods examined comparatively.