Investigations on Classification Algorithms for Intrusion Detection System in MANETS

Abstract

Intrusion Detection System is software based monitoring mechanism for a computer network that detects presence of malevolent activity in the network. Intrusion detection is an eminent upcoming area in relevance as more and more complex data is being stored and processed in networked systems. This paper focuses on investigations of well-known machine learning techniques to address the security issues in the MANET networks which are used for detection and classification of attacks: Intuitionistic fuzzy, genetic algorithm RVM (Relevance Vector Machine), and neural network algorithm. Machine Learning techniques can learn normal and anomalous patterns from training data and generate classifiers that then are used to detect attacks on computer systems. The selected attributes were applied to Data Mining Classification Algorithms which helps in bringing out the best and effective Algorithm by making use of the error rates, false positive and packet drop rates.