AI-POWERED NETWORK INTRUSION DETECTION SYSTEM USING CNN-LSTM

Abstract

The rapid growth of networked systems has significantly increased the risk of cyber-attacks, making intrusion detection a critical component of modern cybersecurity. This project proposes an AIpowered Network Intrusion Detection System (NIDS) that leverages a hybrid deep learning architecture combining Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) networks. The CNN component is utilized to automatically extract spatial features from network traffic data, while the LSTM captures temporal dependencies and sequential patterns essential for detecting sophisticated and evolving attacks. The proposed system is trained on benchmark intrusion detection datasets and is capable of identifying various types of network intrusions, including denial-of-service (DoS), probing, and unauthorized access attacks. By integrating CNN and LSTM, the model achieves improved detection accuracy and reduced false positive rates compared to traditional machine learning methods. Additionally, the system supports real-time monitoring and classification of network traffic, making it suitable for deployment in dynamic and high-volume network environments.