A COMPREHENSIVE ANALYSIS OF NETWORK INTRUSION DETECTION BASED ON CONVOLUTIONAL NEURAL NETWORK AND HYBRID ENSEMBLE ALGORITHM

Abstract

Network intrusion detection has become a critical component in safeguarding digital infrastructure against increasingly sophisticated cyberattacks. Traditional machine learning methods often struggle with high-dimensional data and limited feature extraction capabilities. This paper proposes a novel hybrid framework that integrates Convolutional Neural Networks (CNNs) with decision tree for deep feature extraction with an ensemble of classical machine learning algorithms— Stochastic Gradient Descent (SGD), and eXtreme Gradient Boosting (XGBoost)—for robust classification. The CNN automatically learns hierarchical patterns from raw network traffic data, which are subsequently fed into a hybrid ensemble classifier. Experimental evaluations conducted on benchmark intrusion detection datasets demonstrate that the proposed model significantly improves detection accuracy, precision, and recall while reducing false positives. The hybrid approach leverages the strengths of both deep learning and ensemble methods, offering a scalable and adaptive solution for modern intrusion detection systems.

Author

Mrs.POORNIMA K, Mrs.SAVITHRI M C, Dr.SATHYABALAJI N