Binary Grasshopper Optimization Based Feature Selection for Intrusion Detection System Using Feed Forward Neural Network Classifier

Abstract

Background: Due to the advanced improvement in internet and network technologies, significant number of intrusions and attacks takes place. An intrusion detection system (IDS) is employed to prevent distinct attacks. Several machine learning approaches has been presented for the classification of IDS. But, IDS suffer from the curse of dimensionality that results to increased complexity and decreased resource exploitation. Consequently, it becomes necessary that significant features of data must be investigated by the use of IDS for reducing the dimensionality. Aim: In this article, a new feature selection (FS) based classification system is presented which carries out the FS and classification processes. Methods: Here, the binary variants of the Grasshopper Optimization Algorithm called BGOA is applied as a FS model. The significant features are integrated using an effective model to extract the useful ones and discard the useless features. The chosen features are given to the feed forward neural network (FFNN) model to train and test the KDD99 dataset. Results: The validation of the presented model takes place using a benchmark KDD Cup 1999 dataset. By the inclusion of FS process, the classifier results gets increased by attaining FPR of 0.43, FNR of 0.45, sensitivity of 99.55, specificity of 99.57, accuracy of 99.56, Fscore of 99.59 and kappa value of 99.11. Conclusion: The experimental outcome ensured the superior performance of the presented model compared to diverse models under several aspects and is found to be an appropriate tool for detecting intrusions.