Cloud-enhanced network security system using explainable AI and lightweight deep learning for multi-threat detection

School of Technology, Woxsen University, Hyderabad - 502345, Telangana, India

Published online: 19 June 2026

Abstract

The rapid expansion of cloud computing has increased the complexity and volume of cyber threats, making traditional intrusion detection systems insufficient. This study proposes a cloud-enhanced hybrid intrusion detection system based on a CNN-LSTM architecture to detect multi-type network attacks. The model combines convolutional layers for spatial feature extraction with LSTM units for temporal sequence learning. Evaluated on the CIC-IDS 2017 dataset, the proposed system achieves high detection performance in both binary and multiclass classification tasks, with strong precision, recall, and F1-scores. To improve transparency, an Explainable AI (XAI) module is integrated to provide feature-level interpretation of predictions. Additionally, a lightweight system design with a real-time dashboard enables practical deployment in cloud environments. The results demonstrate that the proposed approach offers an effective balance between detection accuracy, interpretability, and computational efficiency.