Improved Cyber-Security Mechanism Using Integration of OD-DRUNN and XAI-BASED Proactive Defense for Organizations

Today, cyber-attacks have become more severe and frequent, which calls for a new line of security defenses to protect against them. So, to preserve the organizations' details from cyber threats, many research works use different proactive defense mechanisms. But the attacks that are present in the machine learning (ML) model provide misclassification results. So, this research methodology proposed an integration of Explainable Artificial Intelligence (XAI) with Ordinary Differential Deep Recurrent Unit Neural Network (OD-DRUNN) for a proactive defense mechanism in cyber-security for organizations. Initially, the input data is pre-processed and the features are extracted. From the extracted features, the traffic pattern and user behaviors are analyzed using the Minimum Parameterized Muller Spanning Tree (MPMST) approach. Then, the obtained traffic pattern, user behavior, and the extracted features are given as input to the OD-DRUNN classifier. Then, the output is displayed with XAI. With the help of XAI output, the attack in the ML model is neglected. After that, the severity is assessed for the abnormal types using the potential level score. If the severity level is high and moderate, then security is ensured by the Cycloid Curved Optimized Cryptography (2COC) algorithm and stored in the blockchain. Otherwise, it follows normal security procedures for storage. The proposed methodology achieves 99.2% accuracy in potential vulnerability detection.