Explainable AI for zero-day attack detection in IoT networks using attention fusion model

Abstract The proposed research addresses the challenge of detecting malicious network traffic in IoT environments, focusing on enhancing detection accuracy while ensuring interpretability. The proposed attention fusion classification model utilizes both long-term and short-term attention mechanisms...

Full description

Saved in:
Bibliographic Details
Main Authors: Deepa Krishnan, Swapnil Singh, Vijayan Sugumaran
Format: Article
Language:English
Published: Springer 2025-07-01
Series:Discover Internet of Things
Subjects:
Online Access:https://doi.org/10.1007/s43926-025-00184-8
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849332359397113856
author Deepa Krishnan
Swapnil Singh
Vijayan Sugumaran
author_facet Deepa Krishnan
Swapnil Singh
Vijayan Sugumaran
author_sort Deepa Krishnan
collection DOAJ
description Abstract The proposed research addresses the challenge of detecting malicious network traffic in IoT environments, focusing on enhancing detection accuracy while ensuring interpretability. The proposed attention fusion classification model utilizes both long-term and short-term attention mechanisms to capture temporal patterns and protocol-specific features, which improves the differentiation between benign and malicious traffic. Empirical results indicate strong performance, with precision-recall scores of 0.9999 for both the DDoS_TCP and DDoS_UDP classes, and a perfect score of 1.0000 for the Normal class. The model also demonstrates solid performance for the DDoS_HTTP (0.9791), Password (0.9418), and SQL_Injection (0.9461) classes. Furthermore, it excels at identifying complex behaviors in upload-based attacks and network vulnerabilities, achieving precision-recall scores of 0.9333 for the Uploading class and 0.9963 for the Vulnerability Scanner class. The binary classification accuracy is 99.9966%, and the multiclass accuracy for Zero-day attacks is 71.0926%. The results suggest that the model offers significant potential for improving IoT security. This study introduces the novel use of attention mechanisms for interpretability, enhancing the detection of a broad range of attack types, and contributes to advancing intrusion detection system capabilities. Future research can focus on expanding datasets, refining interpretability techniques, and addressing adversarial vulnerabilities for further model enhancement.
format Article
id doaj-art-360046a9c8984098a2b97cd84f64c8f2
institution Kabale University
issn 2730-7239
language English
publishDate 2025-07-01
publisher Springer
record_format Article
series Discover Internet of Things
spelling doaj-art-360046a9c8984098a2b97cd84f64c8f22025-08-20T03:46:13ZengSpringerDiscover Internet of Things2730-72392025-07-015112510.1007/s43926-025-00184-8Explainable AI for zero-day attack detection in IoT networks using attention fusion modelDeepa Krishnan0Swapnil Singh1Vijayan Sugumaran2Department of Computer Engineering, Mukesh Patel School of Technology Management and Engineering, SVKM’s NMIMS UniversityComputer Science Department, Virginia TechDepartment of Decision and Information Sciences, Oakland UniversityAbstract The proposed research addresses the challenge of detecting malicious network traffic in IoT environments, focusing on enhancing detection accuracy while ensuring interpretability. The proposed attention fusion classification model utilizes both long-term and short-term attention mechanisms to capture temporal patterns and protocol-specific features, which improves the differentiation between benign and malicious traffic. Empirical results indicate strong performance, with precision-recall scores of 0.9999 for both the DDoS_TCP and DDoS_UDP classes, and a perfect score of 1.0000 for the Normal class. The model also demonstrates solid performance for the DDoS_HTTP (0.9791), Password (0.9418), and SQL_Injection (0.9461) classes. Furthermore, it excels at identifying complex behaviors in upload-based attacks and network vulnerabilities, achieving precision-recall scores of 0.9333 for the Uploading class and 0.9963 for the Vulnerability Scanner class. The binary classification accuracy is 99.9966%, and the multiclass accuracy for Zero-day attacks is 71.0926%. The results suggest that the model offers significant potential for improving IoT security. This study introduces the novel use of attention mechanisms for interpretability, enhancing the detection of a broad range of attack types, and contributes to advancing intrusion detection system capabilities. Future research can focus on expanding datasets, refining interpretability techniques, and addressing adversarial vulnerabilities for further model enhancement.https://doi.org/10.1007/s43926-025-00184-8Security attackExplainable AIDetectionZero dayIoT
spellingShingle Deepa Krishnan
Swapnil Singh
Vijayan Sugumaran
Explainable AI for zero-day attack detection in IoT networks using attention fusion model
Discover Internet of Things
Security attack
Explainable AI
Detection
Zero day
IoT
title Explainable AI for zero-day attack detection in IoT networks using attention fusion model
title_full Explainable AI for zero-day attack detection in IoT networks using attention fusion model
title_fullStr Explainable AI for zero-day attack detection in IoT networks using attention fusion model
title_full_unstemmed Explainable AI for zero-day attack detection in IoT networks using attention fusion model
title_short Explainable AI for zero-day attack detection in IoT networks using attention fusion model
title_sort explainable ai for zero day attack detection in iot networks using attention fusion model
topic Security attack
Explainable AI
Detection
Zero day
IoT
url https://doi.org/10.1007/s43926-025-00184-8
work_keys_str_mv AT deepakrishnan explainableaiforzerodayattackdetectioniniotnetworksusingattentionfusionmodel
AT swapnilsingh explainableaiforzerodayattackdetectioniniotnetworksusingattentionfusionmodel
AT vijayansugumaran explainableaiforzerodayattackdetectioniniotnetworksusingattentionfusionmodel