IJARCCE adheres to the suggestive parameters outlined by the University Grants Commission (UGC) for peer-reviewed journals, upholding high standards of research quality, ethical publishing, and academic excellence.
Network Weight Updating Method for Intrusion Detection Using Artificial Neural Networks
Dr.S.SARAVANAKUMAR, JERRIN SIMLA.A, L.MEGALAN LEO, Dr.A.REGARAJAN Professor, Dept of IT, Panimalar Institute of Technology ,Chennai Asst.Professor, Department of CSE, Panimalar Institute of Technology, Chennai Asst.Professor, Department of ECE, St.Joseph College of Technology, Chennai Associate Professor, Department of IT, Veltech Multitech SRS engineering college, Chennai
Abstract: Application of Artificial Neural Networks (ANN) to intrusion detection has been considered in this research work. Experimental data were collected using KDD database.Using the data collected, the training patterns and test patterns are obtained. An ANN has been used to train the data offline. The weight updating algorithms developed for the ANN are based on the back propagation algorithms, echo state neural network and the functional update method. The method of presenting the patterns to the input layer of the network has been analyzed. The different methods of presenting the input patterns, such as reducing the dimension of the input patterns by a transformation and preprocessing of the input patterns for non linear classifiers have been investigated.In order to find the optimum number of nodes required in the hidden layer of an ANN a method has been proposed, based on the change in the mean squared error dynamically, during the successive sets of iterations. Two classification and four classification problems for training an ANN at different levels have been studied. Optimal discriminant plane technique which is a classical method has been used to reduce the dimension of the input pattern and then used to train an ANN. This reduces the size of the network and the computational effort is reduced drastically.The training of an ANN has been considered by splitting the single configuration into two configurations. The convergence rates of the split network are faster than that of single configuration network.The input patterns are preprocessed and presented to the input layer of ANN. Various types of preprocessing of the patterns are investigated. Comparisons of the classification performance and computational effort of the different weight updating algorithms with different training methods have been given. Several algorithms developed in this research work may find other application areas and in security with little modifications.
Keywords: ANN, weight updating, input layer, network, KDD
[1] Dr.S.SARAVANAKUMAR, JERRIN SIMLA.A, L.MEGALAN LEO, Dr.A.REGARAJAN Professor, Dept of IT, Panimalar Institute of Technology ,Chennai Asst.Professor, Department of CSE, Panimalar Institute of Technology, Chennai Asst.Professor, Department of ECE, St.Joseph College of Technology, Chennai Associate Professor, Department of IT, Veltech Multitech SRS engineering college, Chennai, βNetwork Weight Updating Method for Intrusion Detection Using Artificial Neural Networks,β International Journal of Advanced Research in Computer and Communication Engineering (IJARCCE)
This work is licensed under a Creative Commons Attribution 4.0 International License.