Intrusion Detection System using KDD Cup 99 Dataset
Ch. Aishwarya1, N. Venkateswaran2, T. Supriya3, M. Sreekar4, V. Sreeja5
1Ch. Aishwarya*, Department of Computer Science and Engineering, Jyothismathi Institute of Technology and Science, Karimnagar, Telangana, India.
2N. Venkateswaran, Associate Professor Department of Computer Science and Engineering, Jyothismathi Institute of Technology and Science, Karimnagar, Telangana, India.
3T. Supriya, Department of Computer Science and Engineering, Jyothismathi Institute of Technology and Science, Karimnagar, Telangana, India.
4M. Sreekar, Department of Computer Science and Engineering, Jyothismathi Institute of Technology and Science, Karimnagar, Telangana, India.
5V. Sreeja, Department of Computer Science and Engineering, Jyothismathi Institute of Technology and Science, Karimnagar, Telangana, India.
Manuscript received on January 12, 2020. | Revised Manuscript received on January 22, 2020. | Manuscript published on February 10, 2020. | PP: 3169-3171 | Volume-9 Issue-4, February 2020. | Retrieval Number: D2017029420/2020©BEIESP | DOI: 10.35940/ijitee.D2017.029420
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. 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: Intrusion Detection System is a vital feature of protecting network infrastructure from unauthorized users or hackers. Intrusion detection system is used to identify several types of malicious activities that could effect the safety of network and to reduce network traffic. Because of faster growth of Internet, networks are growing rapidly in every area of society. As a result, large amount of data is travelling across many networks which may lead to vulnerability of integrity and confidentiality of data. Many Machine learning models are opened up providing new opportunity to classify traffic in network. In quest to select a good learning model, this paper illustrates performance between J48, Naive Bayes and Random forest classification models. The KDD Cup 99 dataset is used for experimental analysis to identify which classification model improves correctness of data and attains highest accuracy.
Keywords: Intrusion Detection, Machine Learning, KDD dataset, Classification models, Naive-bayes, J48, Random Forest, WEKA.
Scope of the Article: Machine Learning