Attention-Based Deep Learning Frameworks for Network Intrusion Detection: An Empirical Study

This paper explores the application of deep learning models for network intrusion detection, comparing Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), and hybrid CNN-RNN architectures with and without attention mechanisms. Using the NSL-KDD dataset, we evaluate the performance of each model in terms of test loss and accuracy. The results demonstrate that RNNs with attention mechanisms achieve the lowest test loss, while pure RNNs provide the highest accuracy. The hybrid CNN-RNN model with attention offers a balanced approach, leveraging both spatial and temporal features of network traffic data. This study highlights the potential of attention-enhanced deep learning models for robust and efficient network intrusion detection.