Detection and Classification of Intrusions using Fusion Probability of HMM

detection system is a technique of identifying unwanted packets that creates harm in the network; hence various IDS are implemented for the security of network traffic flow. Here in this paper an efficient technique of identifying intrusions is implemented using hidden markov model and then classification of these intrusions is done. The methodology sis applied on KDDCup 99 dataset where the dataset is first clustered using K-means algorithms and then a number of attributes is selected which are used for the detection of intrusion is passed to the HMM, after calculating probability from each of the states, these probabilities are fused to get the resultant final probability and also overall probability is calculated from dataset on the basis of which intrusions are classified as low, medium or high. KeywordsAnomaly, HMM, Behavioral Distance