The core of this thesis is the Research and Application of Data-mining based IntrusionDetection System. First, the existing Data-mining technology and Intrusion Detection Systemtechnology were analyzed and compared, the clustering algorithm was discussed. and theadvantages of the technology were indicated. In this paper Nearest Neighbor First (NNF)clustering algorithm is proposed to resolve this problem based on the idea that the data in samecluster must be near. The main contribution include SNN algorithm. Then we test this algorithmwith the data KDD Cup 1999. Finally, the research work was summarized, and gives theproblem to solve further. |