A Survey on Deep Learning-Based Intrusion Detection Systems with SHAP Explainability: Techniques, Challenges, and Future Directions

Radhika S K; Rashmi V

doi:10.36948/ijfmr.2025.v07i05.59221

A Survey on Deep Learning-Based Intrusion Detection Systems with SHAP Explainability: Techniques, Challenges, and Future Directions

Author(s)	Ms. Radhika S K, Ms. Rashmi V
Country	India
Abstract	The rapid growth of networked systems and the increasing sophistication of cyberattacks have made intrusion detection a critical component of modern cybersecurity. Traditional machine learning–based Intrusion Detection Systems (IDS) rely heavily on manual feature engineering and often fail to generalize to novel attack patterns. In contrast, deep learning–based IDSshave proven to deliver enhanced performance via automatic feature extraction and end-to-end learning. However, the high complexity and opaque nature of these models have led to a significant lack of interpretability, often described as the black-box problem. This absence of transparency hinders trust, accountability, and adoption in real-world security environments. To mitigate this issue, explainable artificial intelligence (XAI) techniques—particularly the SHapley Additive exPlanations (SHAP) method—have emerged as powerful tools for interpreting deep learning predictions. SHAP provides both local and global explanations, identifying the most influential features contributing to attack or benign classifications, thereby enhancing the transparency and reliability of IDS decisions. This study provides a structured examination of recent advancements in deep learning-based network intrusion detection mechanisms integrated with SHAP explainability. It categorizes existing methods based on model architecture, detection approach, datasets, and interpretability strategies. Furthermore, it compares key studies to highlight their contributions, strengths, and limitations. The study further highlights existing research gaps challenges such as computational overhead, lack of standardized benchmarks for explainability, and limited evaluation of multiclass and zero-day attack scenarios. Finally, it outlines future directions aimed at developing efficient, interpretable, and deployable IDS frameworks that combine high detection performance with human-understandable decision reasoning.
Keywords	Intrusion Detection System (IDS), Deep Learning, Explainable AI (XAI), SHAP, Model Interpretability, Cybersecurity
Field	Computer > Network / Security
Published In	Volume 7, Issue 5, September-October 2025
Published On	2025-10-31
DOI	https://doi.org/10.36948/ijfmr.2025.v07i05.59221

View / Download PDF File

E-ISSN 2582-2160

doi

CrossRef DOI is assigned to each research paper published in our journal.

IJFMR DOI prefix is
10.36948/ijfmr

Downloads

Research Paper Format Copyright Permission Form and Undertaking Form Cover Page Vol 8 Isu 2 Cover Page Vol 8 Isu 1 Cover Page Vol 7 Isu 6

All research papers published on this website are licensed under Creative Commons Attribution-ShareAlike 4.0 International License, and all rights belong to their respective authors/researchers.

CC-BY-SA

About IJFMR Fees & Payment Current Issue Publication Archive	Submit Research Paper Track Submission Status Publication Guidelines Publication Ethics Peer Review & Plagiarism	Join as a Reviewer Editors & Reviewers Reviewer Referral Program Get Reviewer Membership Certi.	Website/Journal Policies Usage Policy Content Policies Privacy Policy

Contact Us		+91-9687-828-838	editor@ijfmr.com

International Journal For Multidisciplinary Research

A Widely Indexed Open Access Peer Reviewed Multidisciplinary Bi-monthly Scholarly International Journal

A Survey on Deep Learning-Based Intrusion Detection Systems with SHAP Explainability: Techniques, Challenges, and Future Directions

Share this