Efficient and Explainable Human Activity Recognition Using Deep Residual Network with Squeeze-and-Excitation Mechanism

Wearable sensors for human activity recognition (HAR) have gained significant attention across multiple domains, such as personal health monitoring and intelligent home systems. Despite notable advancements in deep learning for HAR, understanding the decision-making process of complex models remains...

Full description

Saved in:
Bibliographic Details
Main Authors: Sakorn Mekruksavanich, Anuchit Jitpattanakul
Format: Article
Language:English
Published: MDPI AG 2025-04-01
Series:Applied System Innovation
Subjects:
Online Access:https://www.mdpi.com/2571-5577/8/3/57
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850157001799106560
author Sakorn Mekruksavanich
Anuchit Jitpattanakul
author_facet Sakorn Mekruksavanich
Anuchit Jitpattanakul
author_sort Sakorn Mekruksavanich
collection DOAJ
description Wearable sensors for human activity recognition (HAR) have gained significant attention across multiple domains, such as personal health monitoring and intelligent home systems. Despite notable advancements in deep learning for HAR, understanding the decision-making process of complex models remains challenging. This study introduces an advanced deep residual network integrated with a squeeze-and-excitation (SE) mechanism to improve recognition accuracy and model interpretability. The proposed model, ConvResBiGRU-SE, was tested using the UCI-HAR and WISDM datasets. It achieved remarkable accuracies of 99.18% and 98.78%, respectively, surpassing existing state-of-the-art methods. The SE mechanism enhanced the model’s ability to focus on essential features, while gradient-weighted class activation mapping (Grad-CAM) increased interpretability by highlighting essential sensory data influencing predictions. Additionally, ablation experiments validated the contribution of each component to the model’s overall performance. This research advances HAR technology by offering a more transparent and efficient recognition system. The enhanced transparency and predictive accuracy may increase user trust and facilitate smoother integration into real-world applications.
format Article
id doaj-art-cfb778137b4345c394704cffb69cf0f3
institution OA Journals
issn 2571-5577
language English
publishDate 2025-04-01
publisher MDPI AG
record_format Article
series Applied System Innovation
spelling doaj-art-cfb778137b4345c394704cffb69cf0f32025-08-20T02:24:18ZengMDPI AGApplied System Innovation2571-55772025-04-01835710.3390/asi8030057Efficient and Explainable Human Activity Recognition Using Deep Residual Network with Squeeze-and-Excitation MechanismSakorn Mekruksavanich0Anuchit Jitpattanakul1Department of Computer Engineering, School of Information and Communication Technology, University of Phayao, Phayao 56000, ThailandDepartment of Mathematics, Faculty of Applied Science, King Mongkut’s University of Technology North Bangkok, Bangkok 10800, ThailandWearable sensors for human activity recognition (HAR) have gained significant attention across multiple domains, such as personal health monitoring and intelligent home systems. Despite notable advancements in deep learning for HAR, understanding the decision-making process of complex models remains challenging. This study introduces an advanced deep residual network integrated with a squeeze-and-excitation (SE) mechanism to improve recognition accuracy and model interpretability. The proposed model, ConvResBiGRU-SE, was tested using the UCI-HAR and WISDM datasets. It achieved remarkable accuracies of 99.18% and 98.78%, respectively, surpassing existing state-of-the-art methods. The SE mechanism enhanced the model’s ability to focus on essential features, while gradient-weighted class activation mapping (Grad-CAM) increased interpretability by highlighting essential sensory data influencing predictions. Additionally, ablation experiments validated the contribution of each component to the model’s overall performance. This research advances HAR technology by offering a more transparent and efficient recognition system. The enhanced transparency and predictive accuracy may increase user trust and facilitate smoother integration into real-world applications.https://www.mdpi.com/2571-5577/8/3/57human activity recognition (HAR)explainable AI (XAI)wearable sensorssqueeze-and-excitation mechanismdeep residual network
spellingShingle Sakorn Mekruksavanich
Anuchit Jitpattanakul
Efficient and Explainable Human Activity Recognition Using Deep Residual Network with Squeeze-and-Excitation Mechanism
Applied System Innovation
human activity recognition (HAR)
explainable AI (XAI)
wearable sensors
squeeze-and-excitation mechanism
deep residual network
title Efficient and Explainable Human Activity Recognition Using Deep Residual Network with Squeeze-and-Excitation Mechanism
title_full Efficient and Explainable Human Activity Recognition Using Deep Residual Network with Squeeze-and-Excitation Mechanism
title_fullStr Efficient and Explainable Human Activity Recognition Using Deep Residual Network with Squeeze-and-Excitation Mechanism
title_full_unstemmed Efficient and Explainable Human Activity Recognition Using Deep Residual Network with Squeeze-and-Excitation Mechanism
title_short Efficient and Explainable Human Activity Recognition Using Deep Residual Network with Squeeze-and-Excitation Mechanism
title_sort efficient and explainable human activity recognition using deep residual network with squeeze and excitation mechanism
topic human activity recognition (HAR)
explainable AI (XAI)
wearable sensors
squeeze-and-excitation mechanism
deep residual network
url https://www.mdpi.com/2571-5577/8/3/57
work_keys_str_mv AT sakornmekruksavanich efficientandexplainablehumanactivityrecognitionusingdeepresidualnetworkwithsqueezeandexcitationmechanism
AT anuchitjitpattanakul efficientandexplainablehumanactivityrecognitionusingdeepresidualnetworkwithsqueezeandexcitationmechanism