Nowadays, computer security has become very familiar question in the society as nearly everyone has connected their computers to internet to get access to information from various informative sources and send or transmit messages in today's much complex computer networking world. The most common security threats are intruder which is generally referred as hacker or cracker and the other is virus. To protect the computer on network from intruders, Intrusion Detection Systems are very much important defensive measure component. In this paper we propose a hybrid approach which is the combination two algorithms for clustering and classification that are K-Means and Naïve Bayes respectively. Using KDD Cup'99 dataset we'll be evaluating the performance of our proposed approach. The evaluation will show that new type of attack can be detected effectively in the system and efficiency and accuracy of IDS will improve in terms of detection rate along with its reasonable prediction time.

1. AditiPurohit, Hitesh Gupta, "Hybrid Intrusion Detection System Model using Clustering, Classification and Decision Table" IEEE 2013.
2. Jonathon Ng, Deepti Joshi, Shankar M. Banik, "Applying Data Mining Techniques to Intrusion Detection" IEEE 2015.
4. TruptiPhutane, Prof. ApashabiPathan, "Intrusion Detection System using Decision Tree &Apriori Algorithm"IJCET 2015.
5. Dr. SaurabhMukherjeea, Neelam Sharma, "Intrusion Detection using Naive Bayes Classifier with Feature Reduction" IEEE 2012.
6. Duanyang Zhao, QingxiangXu, ZhilinFeng, "Analysis and Design for Intrusion Detection System Based on Data Mining" IEEE 2010.
7. Dr. M. Hanumanthappa, Manish Kumar, Dr. T. V. Suresh Kumar, "Intrusion Detection System Using Decision Tree Algorithm" IEEE 2012.

Data Mining Intrusion Detection System K-means Naïve Bayes Sensors.