An Efficient Neural Network Technique for Misuse Detection in IDS

Machine learning methods for Intrusion detection system has been giving elevating accuracy and quality detection potential on novel strikes. At present the two primary techniques of recognition are signature-based and anomaly-based on their detection method. IDS occur in a collection of "spices" and approach the target of recognizing suspicious traffic in alternate methods. There are network based (NIDS) and host based (HIDS) intrusion detection systems. Artificial neural networks (ANN) have become very important and profitable in domains as pattern classification and regression, because using rule-based programming does not offer the right solution or any solution at all. The neural network contains both the supervised and as well as unsupervised learning. In this research work, a neural network based technique is proposed to prevent misuse detection in intrusion detection system. The technique will be implemented on labeled and non-labeled data and combination of both. The technique is implemented for accuracy of intrusion detection in each iterations of detection.