An Intelligent Intrusion Detection Model

The Intrsuion Detecion Systems(IDS) are required the accuracy, the adaptability, and the expansion in the information society to be changed quickly. Also, it is required the more structured, and intelligent IDS to protect the resource which is important and maintains a secret in the complicated network environment. The research has the purpose to build the model for the intelligent IDS, which creates the intrusion patterns. The intrusion pattern has extracted from the vast amount of data. To manage the large size of data accurately and efficiently, the link analysis and sequence analysis among the data mining techniqes are used to build the model creating the intrusion patterns. The model is consist of "Time based Traffic Model", "Host based Traffic Model", and "Content Model", which is produced the different intrusion patterns with each model. The model can be created the stable patterns efficiently. That is, we can build the intrusion detection model based on the intelligent systems. The rules prodeuced by the model become the rule to be represented the intrusion data, and classify the normal and abnormal users. The data to be used are KDD audit data.