An Emperical Study on Support Vector Machines for Intrusion Detection

In today's world, tera-bytes of data are stored and exchanged starting from one source to another destination and vice-versa. The information when exchanged or stored over the network, there is possibility to attacks and tampered. Different procedures or applications are accessible to secure information from existed vulnerability. In this way we can analyze the information and to decide several algorithms have been proposed which might be emerged to mitigate vulnerabilities. This paper surveys machine learning algorithm SVM that has been implemented for classification and regression. Related to networking, Lot of researches are done that proves SVM performs efficient performance in simplification of network issues when compare to other network classifiers. Our research paper shows a speculative part of SVM with its concepts and applications.