K-Means Clustering Approach to Analyze NSL-KDD Intrusion Detection Dataset
Vipin Kumar1, Himadri Chauhan2, Dheeraj Panwar3
1Vipin Kumar, Department of Computer Science and Engineering, Graphic Era University, Dehradun, India.
2Himadri Chauhan, Department of Computer Science and Engineering, Graphic Era University, Dehradun, India.
3Dheeraj Panwar, Department of Computer Science and Engineering, DCMTE, Dehradun, India.
Manuscript received on August 03, 2013. | Revised Manuscript received on August 29, 2013. | Manuscript published on September 05, 2013. | PP: 1-4 | Volume-3, Issue-4, September 2013. | Retrieval Number: D1742093413
Open Access | Ethics and Policies | Cite
© The Authors. Published By: Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Clustering is the most acceptable technique to analyze the raw data. Clustering can help detect intrusions when our training data is unlabeled, as well as for detecting new and unknown types of intrusions. In this paper we are trying to analyze the NSL-KDD dataset using Simple K-Means clustering algorithm. We tried to cluster the dataset into normal and four of the major attack categories i.e. DoS, Probe, R2L, U2R. Experiments are performed in WEKA environment. Results are verified and validated using test dataset. Our main objective is to provide the complete analysis of NSL-KDD intrusion detection dataset.
Keywords: Clustering, K-means, NSL-KDD Dataset, WEKA.