@inproceedings{84886c0fe89340e68478aaab13930dff,
title = "An Explainable AI Framework for Intrusion Detection Using SHAP and LIME",
abstract = "As cyber threats grow more complex, traditional intrusion detection systems struggle to adapt while remaining interpretable. Though deep learning offers high detection accuracy, its lack of transparency limits use in critical environments. This paper proposes a deep learning-based IDS enhanced with explainable AI techniques SHAP for global feature insights and LIME for instance-level explanations. Using the CICIDS-2017 dataset, the framework combines neural feature extraction with temporal analysis, embedding interpretability into the detection process. This approach balances accuracy and transparency, supporting actionable alerts and regulatory compliance, and paving the way for trustworthy, AI-driven cybersecurity.",
keywords = "CICIDS-2017, Cybersecurity, Deep Learning, Explainable AI, Intrusion Detection, LIME, SHAP",
author = "Deepika Goyal and Singh, \{Sunil K.\} and Sudhakar Kumar and Vanshika Chilkoti and Chui, \{Kwok Tai\} and Hsu, \{Ching Hsien\} and Gupta, \{Brij B.\}",
note = "Publisher Copyright: {\textcopyright} 2025 IEEE.; 14th IEEE Global Conference on Consumer Electronics, GCCE 2025 ; Conference date: 23-09-2025 Through 26-09-2025",
year = "2025",
doi = "10.1109/GCCE65946.2025.11275089",
language = "English",
series = "GCCE 2025 - 2025 IEEE 14th Global Conference on Consumer Electronics",
pages = "1315--1319",
booktitle = "GCCE 2025 - 2025 IEEE 14th Global Conference on Consumer Electronics",
}