POTENTIATE THE DETECTION-RATE OF NETWORK INTRUSION DETECTION USING ADABOOST ALGORITHM

Network intrusion detection aims at differentiate the intrusions on the Internet from normal use of Internet and is an essential part of the information security system. Network consists of nodes whose operation can be controlled by underlying network. KDDCUP’99 is the mostly widely used data set for the evaluation of signature-based IDSs. In this paper, first a conventional online Adaboost process is used where decision stumps are used as weak classifier. In the second algorithm, online Adaboost process is used and online Gaussian mixture models (GMMs) are used as weak classifier. In addition to the algorithm proposed particle swarm optimization (PSO) and support vector machine (SVM) is used. A distributed intrusion detection framework is proposed, in which a local parameterized detection model is constructed in individual node using the online Adaboost algorithm. The global detection model is constructed in each node by combining the local parametric models using a minimum number of samples in the node, which is used to detect intrusions. The algorithm integrates the local detection models global model in each node. This handles the intrusion category found in other nodes, without having to share samples of these intrusion types.