A Hybrid Scheme based on Big Data Analytics using Intrusion Detection System

Objective: Network security plays a key role for many organizations. Host based and network based Intrusion Detection Systems are available in the market depending upon the detection technology used by them. The objective of this research paper is maintaining security across the heterogeneous data from homogeneous sources and co-relating the heterogeneous data from different sources using hybrid strategy. Methods/Statistical Analysis: A real time detection Intrusion Prevention Systems (IPS), prevents security intrusions by gathering and composing with technologies. Findings: Heterogeneous data from different sources has been collected from KDD Cup Dataset and segregated into learning phase and detection phase. In the learning phase, known attacks will be identified. Similarly detection phase also will consider the same. Applications/Improvements: The proposed system specifies a set of rules and high DoS, R2L, U2R, Probe. One may attempt to get good results by improving the efficiency and reducing the complexity present in the model. In future several reduction techniques may be studied to get more features.