An optimization model of computer network security based on GABP neural network algorithm

Abstract

Background: Digital connectivity drives global innovation, ensuring computer network security is more critical than ever. With increasing interconnectivity, cyber threats are becoming more sophisticated, highlighting the urgent need for advanced defense mechanisms. Objective: This research proposed the investigation of the Genetic Algorithm with Back Propagation Neural Network (GA-BPNN) model for computer network safety. Methodology: The investigation utilized the CIC-IDS-2017 dataset to evaluate the GA-BPNN model. Data preprocessing was performed using min–max normalization to standardize the dataset. Independent component analysis (ICA) was employed for feature extraction to improve the model’s efficiency. The GA-BPNN model was then implemented and analyzed for its performance in detecting network threats. Results: The GA-BPNN approach demonstrated strong performance metrics for computer network security, achieving a recall (95%), F1 score (96.5%), precision (98.5%), and accuracy (98%). Conclusion: The GA-BPNN model enhances network security by improving threat detection and incident response. Its high accuracy and precision indicate its potential for strengthening cybersecurity defenses in interconnected systems. Implementing GA-BPNN in network security frameworks could lead to more effective protection against cyber threats.