Addressing big data analytics for classification intrusion detection system

Abstract

Currently, with the rapid developments communication technologies, large number of trustworthy online systems and facilities has been introduced. The cybersecurity is quiet on the rise threat from unauthorized; such security threats can be detected by an intrusion detection system. Thus, enhancing the intrusion detection system is main object of numbers of research and developers for monitoring the network security. Addressing challenges of big data in intrusion detection is one issue faced the researchers and developers due to dimensionality reduction in network data. In this paper, hybrid model is proposed to handle the dimensionality reduction in intrusion detection system. The genetic algorithm was applied as preprocessing steps for selecting most significant features from entire big network dataset. The genetic algorithm was applied to generate subset of relevant features from network data set for handling dimensionality reduction. The Support Vector Machine (SVM) algorithm was processed the relevant features for detecting intrusion. The NSL-KDD standard data was considered to test the performance of the hybrid model. Standard evaluation metrics were employed to presents the results of hybrid model. It is concluded that the empirical results of hybrid outperformed the performance of existing systems.