Innovative Empirical Approach for Intrusion Detection Using ANN

Intrusion detection system based on Artificial Neural Network (ANN) is a very active field that detects normal or attack connection on the network and can improve the performance of Intrusion detection system (IDS), the flash alarm rate in establishing intrusive activities can be reduced. At present computer network and cloud based computing technology is used by an increasing number of users. Computer and network security has received and will still receive much attention. Any unexpected intrusion will damage the network. The areas like business, finance, medical, security sectors have made us reliant on the computer networks. It is important to secure system for which we require strong intrusion detection system which is capable of monitoring network which carries huge amount of data packets as well as reports malicious activity that occurs in the system. Therefore some strategy is needed for best promising security to monitor the anomalous behavior in computer network. A discussion of the upcoming technology and various methodologies which promise to improve the capability of computer system to detect intrusions is offered. In the proposed approach the Artificial Neural Network (ANN) algorithms are used as classifiers for detecting the normal and attack records by training and testing the KDD CUP 99 dataset. It is proved that the FFNN with 10 neurons and 2 layers has performed better over FFNN with different number of neurons and 2 layers.