Explainable AI-Driven Intrusion Detection System Using Hybrid Deep Learning Models for Secure Network Infrastructure

View at publisher Summarise with VIRAT
Authors: K. Sathish, M. Ramkumar Prabhu, V. Suresh Kumar, S. Dinakar Raj
Cited by: 0
References: 16

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Threats to network security developments are still in existence and system advanced intrusion detection systems are required which can attain high precision and interpretable decision-making procedures. Conventional IDS solutions tend not to be very transparent and also security analysts may find it hard to make out patterns of attack and have a feel of how the systems react to them. The research studied a novel hybrid deep learning model composed of Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) networks along with the explainable AI methods such as SHAP (SHapley Additive exPlanations) and LIME (Local Interpretable Model-agnostic Explanations) that can be used to maximize the model interpretability. The performance of the system was tested on NSL-KDD, CICIDS2017 datasets by extending the preprocessing, feature selection, and cross-validation elements to 98.7% detection accuracy, 97.3% precision, 96.8% recall, and 97.05 F1-score which was 12-15% better than any conventional machine learning algorithms with 89% explainability confidence scores. Concluding Remarks: The explainable AI and hybrid deep learning models have a great effect on the performance of detection and bring interpretability to the next level, offering security professionals understandable information they can use to mitigate the threat ahead of time.

Fields of Study & Subjects

Network Security and Intrusion Detection Network Packet Processing and Optimization Internet Traffic Analysis and Secure E-voting

Publication details

Journal / source
International Conference on Communication, Computing and Electronics Systems (ICCCES)
Pages
372–377
Publication date
21 January 2026
Language
EN