Intrusion Detection System Using Back Propagation Algorithm And Compare Its Performance With Self Organizing Map

Abstract

Abstract- In recent years, internet and computers have been utilized by many people all over the world in several fields. On the other hand, network intrusion and information safety problems are ramifications of using internet. The growing network intrusions have put companies and organizations at a much greater risk of loss. This thesis proposes a new learning methodology towards developing a novel Intrusion Detection System (IDS) by Back Propagation Neural Networks (BPN) and Self Organizing Map (SOM) and compares the performance between them. The main function of Intrusion Detection System is to protect the resources from threats. It analyzes and predicts the behaviors of users and then these behaviors resemble either an attack or the normal behavior. There are several existing techniques that provide more security to the network, but most of these techniques are static. This thesis tests the proposed method by a benchmark intrusion dataset to verify its feasibility and effectiveness. Results show that choosing good hidden layers network and input data will not only have impact on the performance, but also on the overall execution efficiency. The proposed method can significantly reduce the training time and epoch time. It provides a powerful tool to help supervisors analyze, model and understand the complex attack behavior of electronic crime.The proposed methodology implemented in sampled data from KddCup99 data set so that intrusion detection attacks database is standard for the evaluation of intrusion detection systems.