OPTIMIZING MACHINE LEARNING MODELS FOR EFFICIENT PREVENTION OF CYBER ATTACKS IN NETWORK INTRUSION DETECTION SYSTEMS

This paper explores the optimization of machine learning (ML) models designed to improve cyber-attack prevention in network intrusion detection systems.The study emphasizes improving the performance and reliability of these models by fine-tuning their hyperparameters using advanced optimization techniques.Using the widely recognized CIC-IDS-2018 dataset, we conducted an exhaustive comparative analysis of various classification algorithms, including decision trees, support vector machines, logistic regression, and advanced ensemble approaches such as random forests and gradient boosting.Empirical evidence demonstrates substantial performance improvements from optimized configurations, especially in identifying complex and emerging cyber threats.Critical evaluation metrics, such as F1-score, accuracy, recall, and precision, confirm the superiority of optimized ML models in protecting vital infrastructures against ongoing cyber risks.

Hali tarjima qilinmagan