Combination of data mining techniques for intrusion detection system

As Internet continues to influence our day to day activities like eCommerce, eGoverence, eEducation etc. the threat from hackers has also increased. Due to which many researcher thinking intrusion detection systems as fundamental line of defense. However, many commercially available intrusion detection systems are predominantly signature-based that are designed to detect known attacks. These systems require frequent updates of signature or rules and they are not capable of detecting unknown attacks. One of the solution is use of anomaly base intrusion detection systems which are extremely effective in detecting known as well as unknown attacks. One of the major problem with anomaly base intrusion detection systems is detection of high false alarm rate. In this paper, we provide solution to increase attack detection rate while minimizing high false alarm rate by combining various data mining techniques.