A Multi-Step Adaptive Synthetic Oversampling and Random Forest Cascaded Model for Multi-Class Intrusion Detection

Abstract

Hackers have developed better and smart traditions to attack WSN in sequence when data are transfer in systems. The harm, hackers can carry out upon thorough a WSNs is well understood. A reasonable damage scenario can be envisaged where a state intercepting encrypted financial data gets hacked. Logical cyber security systems have become without doubt significantfor improved security against malicious threats. The proposed multi-step adaptive synthetic oversampling and random forest cascaded model for intrusion detection system (IDS) using big data, The NSL-KDD dataset used as a benchmark to evaluate the feasibility and effectiveness of the proposed architecture. Simulation results demonstrate the potential of our proposed IDS system, performance better compared to existing methods.