A Roughset Based Ensemble Framework for Network Intrusion Detection System

A Roughset Based Ensemble Framework for Network Intrusion Detection System

Sireesha Rodda (Department of CSE, GITAM (Deemed to be University), Visakhapatnam, India) and Uma Shankar Erothi (Department of CSE, GITAM (Deemed to be University), Visakhapatnam, India)
Copyright: © 2018 |Pages: 18
DOI: 10.4018/IJRSDA.2018070105


Designing an effective network intrusion detection system is becoming an increasingly difficult task as the sophistication of the attacks have been increasing every day. Usage of machine learning approaches has been proving beneficial in such situations. Models may be developed based on patterns differentiating attack traffic from network traffic to gain insight into the network activity to identify and report attacks. In this article, an ensemble framework based on roughsets is used to efficiently identify attacks in a multi-class scenario. The proposed methodology is validated on benchmark KDD Cup '99 and NSL_KDD network intrusion detection datasets as well as six other standard UCI datasets. The experimental results show that proposed technique RST achieved better detection rate with low false alarm rate compared to bagging and RSM.
Article Preview

The problem of designing an efficient NIDS has garnered lot of attention from the research community lately.

(Bhavsar & Waghmare, 2013) proposed a model for IDS using Support Vector Machine (SVM) with the three kernel functions (Gaussian Radial Basis Function (RBF) kernel, polynomial kernel and sigmoid kernel). Experiments on NSL_KDD intrusion dataset is analyzed in terms of time taken to build different SVM kernel models. SVM classification with RBF kernel and tenfold cross validation with re-evaluation using supplied test-set achieved less time to build model and higher accuracy.

(Thanasekaran,2011) presented ensemble multi-boosting and binary classification algorithm using C4.5 for Network Intrusion Detection System (NIDS). NIDS using Artificial Neural Network (ANN) technique reduced false alarm rate and achieved higher attack detection. The experimental results show ensemble binary classification with multi-boosting achieved higher accuracy than the C4.5. The time taken to detect attacks has been reduced with the use of dynamic multi-boosting.

Complete Article List

Search this Journal:
Open Access Articles: Forthcoming
Volume 6: 4 Issues (2019): 1 Released, 3 Forthcoming
Volume 5: 4 Issues (2018)
Volume 4: 4 Issues (2017)
Volume 3: 4 Issues (2016)
Volume 2: 2 Issues (2015)
Volume 1: 2 Issues (2014)
View Complete Journal Contents Listing