A Survey on Various Techniques for Classifying Attacks in NIDS

Intrusion Detection System (IDS) is the most powerful system that can handle the intrusions of the computer environments by triggering alerts to make the analysts take actions to stop this intrusion. IDSs are based on the belief that an intruders behavior will be noticeably different from that of a legitimate user. A variety of intrusion detection systems (IDS) have been employed for protecting computers and networks from malicious attacks by using traditional statistical methods to new data mining approaches in last decades. The conventional system is not efficient for unseen data and they need to be updated frequently to work properly. There are several techniques which classify data into only normal and threat or attack type, further classification is not done which leads to less accuracy. In several approaches dimensionality of input set is large which makes the problem complex and redundancy might increase. So basically in todays world of internet and automation, it is important to maintain a security, authenticity. There must be a proper efficient technique which can detect attacks and classify them into proper attack categories. The further classification into sub-attack categories plays a vital role in IDS as likewise preventive actions can be taken. So basically in this paper we are going to focus on various techniques for classifying attacks in NIDS.