A hybrid cheetah and grey wolf optimization algorithm for network intrusion detection
Deepshikha Kumari, Prashant Pranav, Abhinav Sinha, Sandip Dutta · Engineering Research Express · 2025
Abstract In the rapidly evolving field of cybersecurity, anomaly detection continues to be a crucial challenge for identifying and preventing potential threats. This study presents an innovative hybrid approach to enhance intrusion detection systems by combining the Cheetah Optimizer Algorithm (COA) and the Grey Wolf Optimizer (GWO) with Convolutional Neural Networks (CNN). To evaluate the performance of this hybrid COA-GWO algorithm, the research utilizes five cutting-edge multiclass datasets: TII-SSRC, WSN-DS, KITSUNE, MSCAD, and Edge-IIOT. By integrating these optimization algorithms, the proposed method significantly improves the CNN-based intrusion detection models’ ability to identify network anomalies, achieving remarkable detection accuracies of 99%, 97%, 96%, 97%, and 97% on the respective datasets. These results highlight the powerful potential of hybrid optimization techniques combined with deep learning to enhance the accuracy and efficiency of intrusion detection systems. Ultimately, this research advances intrusion detection strategies by showcasing the superior performance of the COA-GWO optimized CNN in addressing a range of real-world cybersecurity challenges.