A Comparative Study of Classification Techniques for Intrusion Detection Using Nsl-kdd Data Sets

Data Mining is a technique to drilling the database for giving meaning to the approachable data. It involves systematic analysis of large data sets. And the classification is used to manage data, sometimes tree modeling of data helps to make predictions about new data. Recently, we have increasing in the number of cyber-attacks, detecting the intrusion in networks become a very tough job. In Network Intrusion Detection System (NIDS), many data mining and machine learning techniques are used. However, for evaluation, most of the researchers used data set DARPA 2000, which has widely criticized not suitable for current network situation. We have labeled a network dataset and also an improved version of KDD Cup datasets, called NSL-KDD dataset. In NSL-KDD data set, every instant is labeled as normal (no attack), attack (Dos, U2R, R2L, and Probe). In NSL-KDD dataset we have only a selected dataset to provide a good analysis on various machine learning techniques for intrusion detection. This analysis explains discussion of Random Forest, J48, ZeroR, and Naïve Bayes. Among them we get best classification algorithm for the given dataset.