BibTeX:

@article{IJIRSTV1I9063,
     title={Using Support Vector Machine and Naive Bayes Classification for Intrusion Detection },
     author={Aman Mudgal and Rajiv Munjal},
     journal={International Journal for Innovative Research in Science & Technology},
     volume={1},
     number={9},
     pages={224--227},
     year={},
     url={http://www.ijirst.org/articles/IJIRSTV1I9063.pdf},
     publisher={IJIRST (International Journal for Innovative Research in Science & Technology)},
}

For identifying the various attacks involved in a particular network, Intrusion Detection System is used. The role of Intrusion Detection System (IDS) is identified and prevent the malicious or un-authorized user to access the system. The data present on web or other data on the system or the data on a network of systems have many complicated and structural relations. To uniquely and properly identify the intrusion in a network it is necessary to understand the structural relationship between the data types and many algorithms based for clustering data neglect the relation among the individual data types. A defensive layer, or in other words the preventive layer is made between the data and data types by Intrusion Detection System; this ensures the unauthorized and unauthenticated users to have access for the system. These unauthorized users can be easily detected by sensing and sending a Alert Message, Operator to the Administrator. The Support Vector Machine and Naïve Bayes Classification are defined as the two techniques which can be used to create a valuable Intrusion Detection System. The proposed System accepts Fuzzy K-Means clustering algorithm which accepts either Support Vector Machine or Naïve Bayes Classifier to detect the malicious Users. The administrator would have the complete data set which will help out to detect the malicious users after complete scanning and authorization of user. If any type of user with different data sets trying to access system, then can be considered as a Intrusion Detection. In this paper we combine two of the efficient data mining algorithms and make a hybrid technique for the detection of intrusion called fuzzy k-means, Naïve Bayes Classification and Support vector machine.

Intrusion Detection, Fuzzy K-Mean, SVM, Naïve Bayes Classification