Hybrid Intrusion Detection System Model using Clustering, Classification and Decision Table
Aditi Purohit · IOSR Journal of Computer Engineering · 2013
Nowadays, computer networks are so complex, nearly everyone with a computer has connected it to the Internet to access information and transmit messages and as complexity increases the question of security becomes more and more familiar as well as the depth knowledge of computer network protocols namely; Transmission Control Protocol (TCP), Internet Protocol (IP) being the most common ones amongst others like User Datagram Protocol (UDP) etc...One of the two most publicized to security is intruder (other is virus) generally referred to as hacker or cracker.Intrusion detection systems are an important component of defensive measures protecting computer systems and networks from abuse.In this paper we proposes, a new hybrid learning approach that combines K-Mean clustering, Naive Bayes (statistical) also known as KMNB with Decision Table Majority (rule based) approaches.An experiment will be Carrie out to evaluate the performance of the proposed approach using KDD Cup '99 dataset.The experimental results will shows that new type of attack can be detected effectively in the system, so that the efficiency and accuracy of intrusion detection system will improve terms of detection rate as well false positive rate with reasonable prediction time.