Abstract: An intrusion detection system is used to find the malicious activities in the network and reports to the management. It can be done in two ways. One is host based intrusion detection system and another one is network based intrusion detection system. There are two types of detection methods one is anomaly detection and another one is misuse detection. In this paper classification algorithm is used to detect the malicious activities in the network. Support vector machine concept is used to find the optimal hyper plane. Initially classification is done to classify the data whether it is normal or an attack. Then clustering concept is applied to group the similar attacks. After grouping multiple class batch processing is done.

Keywords: Intrusion detection system, Detection types, data mining, classification, Support vector machine, Multiclass batch algorithm.