Enhanced Intrusion Detection System Based on Deep Learning

Authors- Saddam Hussain, Prof. Santosh Nagar, Prof. Anurag Shrivastava

Abstract-– With the rapid growth of digital networks, cyber security threats have become increasingly sophisticated, making traditional Intrusion Detection Systems (IDS) less effective. To address this challenge, we propose an Enhanced Intrusion Detection System (IDS) based on deep learning techniques, significantly improving threat detection accuracy and response efficiency. The proposed system leverages Deep Neural Networks (DNNs), Convolutional Neural Networks (CNNs), and Long Short-Term Memory (LSTM) to analyze network traffic and detect malicious activities in real time. By integrating feature selection, the model optimizes classification performance, reducing false positives while maintaining high detection rates. The system is trained and evaluated on the NSL-KDD dataset, achieving 97.6% accuracy, 97.2% precision, 96.7% recall, and an F1-score of 96.9%, demonstrating its superior performance in identifying cyber threats. These results highlight the potential of deep learning in enhancing network security by providing intelligent, real-time IDS capable of effectively mitigating emerging cyber threats. The proposed approach sets a strong foundation for developing highly accurate and adaptive intrusion detection systems in modern cyber security environments.

DOI: /10.61463/ijset.vol.13.issue2.299