Skip to content
Conference Open access

Comparative Evaluation of Explainable AI-Driven Intrusion Detection Systems: Performance, Interpretability, and Scalability Analysis

2026 · EPJ Web of Conferences · 0 citations · 11 references

Abstract

Rapid advancements in cybersecurity have made it necessary for developing intelligent, as well as transparent, intrusion detection systems. Explainable Artificial Intelligence (XAI) has become an important concept in providing greater interpretability and accountability in machine learning-based IDSs. This paper conducts a thorough survey and evaluation of IDS models based on XAI, focusing on their architecture, detection performance, interpretability and scalability. This paper evaluates various IDS frameworks that use deep packet inspection, feature engineering methodologies, machine learning, deep learning approaches and explainability techniques such as SHAP and LIME. An overview of the architectural framework of XAI-based IDS systems is provided to demonstrate the interplay between traffic acquisition, detection engines, and explainers. In addition, a comparison of several recent studies is provided in terms of data sets, detection models, explanation approaches and performance metrics. Finally, the tradeoff between detection and interpretability, along with issues with regard to computational cost are discussed. In conclusion, some important gaps in research and areas for future research have been suggested in this paper.

Read PDF

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.