Enhanced Intrusion Detection System for Detecting Rare Class Attacks using Correlation based Dimensionality Reduction Technique

Background/Objective: With Fast growing internet world the risk of intrusion has also increased, as a result Intrusion Detection System (IDS) is the admired key research field. IDS are used to identify any suspicious activity or patterns in the network or machine, which endeavors the security features or compromise the machine. IDS major use all the features of the data. It is a keen observation that all the features are not of equal relevance for the detection of attacks. Moreover every feature does not contribute in enhancing the system performance significantly. The aim of the work done is to find out the smallest subset of most important attributes to design an efficient IDS model. Methods/Statistical Analysis: By implementing Correlation Feature Selection (CFS) mechanism using 6 search algorithms, a smallest set of features sis selected with all the features that are selected very frequently. Findings: The smallest subset of features chosen is the most nominal among all the feature subset found i.e.12 features. Further, the performances using Naïve Bayes and Random Tree classifiers is compared for 7 subsets found by filter model and 41 attributes. Results: The outcome indicates a remarkable improvement in the performance metrics used for comparison of the two classifiers. The simulation results with enhanced classifiers accuracy is approx. 82% to 86% for Random tree and 33% to 65% for Naïve Bayes with 41 and 12 features respectively. There is a noticeable improvement in classifiers accuracy and exposure of U2R attacks s for the proposed smallest subset in comparison to other six subsets as shown in the result. Application: The proposed work with such an improved detection rate and lesser classification time and larger merits of the minimal subset found will play a vital role for the network administrator in choosing efficient IDS.