A Blockchain-Based Intrusion Detection And Mitigation System Using Optimized Stacked Convolutional Extreme Gradient Boosting Machine

Intrusion detection systems (IDSs) are essential for protecting networks from malicious attacks. Though the existing research provided several advancements in the research area, there are various issues, including high false positive rates, delayed detection times, tampering, and inefficiencies in managing dynamic threats that limit their performance. Advanced researches are required because conventional IDS frequently cannot keep up with the growing complexity of infiltration techniques. To overcome these drawbacks, and to present a novel model, Platypterus Elite Bubble-net Optimization enabled Stacked Convolutional Extreme Gradient Boosting coupled with Parallel Proof of Work (PEBO-CnXGB-PPOW) is presented in the research. PEBO-CnXGB-PPOW makes use of Stacked Convolutional XGBoost to improve intrusion detection speed and accuracy by utilizing strong feature extraction and classification capabilities. To elevate the research, Blockchain-based storage is incorporated as a security concern. By combining the best features of whale and sailfish optimization methods, the PEBO algorithm achieves higher performance by allowing for faster convergence and effective parameter adjustment. For the most part, PEBO-CnXGB-PPOW not only addresses the shortcomings of current IDS but also establishes new standards for security, adaptability, and efficacy in the face of constantly changing Cybersecurity threats. The numerical achievements are evaluated with accuracy, precision, and recall achieved 98.36%, 98.91%, and 97.57%, respectively.