INTEGRATING NETWORK INTRUSION DETECTION WITH MACHINE LEARNING TECHNIQUES FOR ENHANCED NETWORK SECURITY

Abstract

In today's increasingly interconnected world, cybersecurity threats are more prevalent than ever, making robust 
intrusion detection systems (IDS) a critical necessity. This research introduces a hybrid IDS model that integrates 
Fuzzy C-Means clustering with classification methods such as Logistic Regression (LR), K-Nearest Neighbor 
(KNN), Stochastic Gradient Descent (SGD), and Naïve Bayes (NB). Advanced feature selection techniques are 
applied to improve detection accuracy and resilience against evolving cyberattacks. Extensive experimentation is 
used to validate the efficacy of this approach using the NIDS dataset. This study addresses limitations in traditional 
IDS methods, particularly their vulnerability to novel and complex attacks, and provides insights into leveraging 
machine learning (ML) to strengthen network security