Ensem-DeepHAR: Identification of human activity in smart environments using ensemble of deep learning methods and motion sensor data

Recognizing human activity plays a crucial role in many applications such as medical care services in smart healthcare environments. Inertial or motion sensors can measure physiognomies such as acceleration and angular velocity of body movement while performing the activities and we can use them to...

Full description

Saved in:
Bibliographic Details
Main Authors: S.M. Mohidul Islam, Kamrul Hasan Talukder
Format: Article
Language:English
Published: Elsevier 2024-12-01
Series:Measurement: Sensors
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S266591742400374X
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1846145946902069248
author S.M. Mohidul Islam
Kamrul Hasan Talukder
author_facet S.M. Mohidul Islam
Kamrul Hasan Talukder
author_sort S.M. Mohidul Islam
collection DOAJ
description Recognizing human activity plays a crucial role in many applications such as medical care services in smart healthcare environments. Inertial or motion sensors can measure physiognomies such as acceleration and angular velocity of body movement while performing the activities and we can use them to learn the models capable of activity recognition. Over the past decades, many state-of-the-art activity recognition systems have been developed but there is still room to improve. In this paper, we have proposed a novel approach to identify human activity from motion sensor data by employing an enormous analysis of sensor data. Based on data analysis, we yielded quality data by preprocessing using a preprocessing chain for human activity recognition (PC-HAR) which also includes the Synthetic Minority Over-sampling Technique to balance the data of the dataset. As a recognition model, we proposed an ensemble of three different deep learning algorithms, namely, modified DeepConvLSTM, modified InceptionTime, and modified ResNet which is named ‘Ensem-DeepHAR’. The outcome of the proposed model is carried out by stacking predictions from each of the mentioned models and then a Random Forest as a meta-model uses those predictions to recognize the final activity. We evaluated our method on both person-dependent and person-independent cases and achieved 99.31 %, 99.08 %, and 97.52 % accuracies for the former case and 97.95 %, 98.11 %, and 99.51 % accuracies for the latter case using three common benchmark datasets: WISDM_ar_v1.1, PAMAP2, and UCI-HAR respectively. The various performance metrics and measures of experimental results establish the supremacy of the proposed model over the state-of-the-arts.
format Article
id doaj-art-39e4999f1b2d441b85e8aaf50860a817
institution Kabale University
issn 2665-9174
language English
publishDate 2024-12-01
publisher Elsevier
record_format Article
series Measurement: Sensors
spelling doaj-art-39e4999f1b2d441b85e8aaf50860a8172024-12-02T05:05:24ZengElsevierMeasurement: Sensors2665-91742024-12-0136101398Ensem-DeepHAR: Identification of human activity in smart environments using ensemble of deep learning methods and motion sensor dataS.M. Mohidul Islam0Kamrul Hasan Talukder1Corresponding author.; Computer Science and Engineering Discipline, Khulna University, Khulna, 9208, BangladeshComputer Science and Engineering Discipline, Khulna University, Khulna, 9208, BangladeshRecognizing human activity plays a crucial role in many applications such as medical care services in smart healthcare environments. Inertial or motion sensors can measure physiognomies such as acceleration and angular velocity of body movement while performing the activities and we can use them to learn the models capable of activity recognition. Over the past decades, many state-of-the-art activity recognition systems have been developed but there is still room to improve. In this paper, we have proposed a novel approach to identify human activity from motion sensor data by employing an enormous analysis of sensor data. Based on data analysis, we yielded quality data by preprocessing using a preprocessing chain for human activity recognition (PC-HAR) which also includes the Synthetic Minority Over-sampling Technique to balance the data of the dataset. As a recognition model, we proposed an ensemble of three different deep learning algorithms, namely, modified DeepConvLSTM, modified InceptionTime, and modified ResNet which is named ‘Ensem-DeepHAR’. The outcome of the proposed model is carried out by stacking predictions from each of the mentioned models and then a Random Forest as a meta-model uses those predictions to recognize the final activity. We evaluated our method on both person-dependent and person-independent cases and achieved 99.31 %, 99.08 %, and 97.52 % accuracies for the former case and 97.95 %, 98.11 %, and 99.51 % accuracies for the latter case using three common benchmark datasets: WISDM_ar_v1.1, PAMAP2, and UCI-HAR respectively. The various performance metrics and measures of experimental results establish the supremacy of the proposed model over the state-of-the-arts.http://www.sciencedirect.com/science/article/pii/S266591742400374XHuman activityPC-HARStackingEnsem-DeepHARMotion sensor dataPerson-independent
spellingShingle S.M. Mohidul Islam
Kamrul Hasan Talukder
Ensem-DeepHAR: Identification of human activity in smart environments using ensemble of deep learning methods and motion sensor data
Measurement: Sensors
Human activity
PC-HAR
Stacking
Ensem-DeepHAR
Motion sensor data
Person-independent
title Ensem-DeepHAR: Identification of human activity in smart environments using ensemble of deep learning methods and motion sensor data
title_full Ensem-DeepHAR: Identification of human activity in smart environments using ensemble of deep learning methods and motion sensor data
title_fullStr Ensem-DeepHAR: Identification of human activity in smart environments using ensemble of deep learning methods and motion sensor data
title_full_unstemmed Ensem-DeepHAR: Identification of human activity in smart environments using ensemble of deep learning methods and motion sensor data
title_short Ensem-DeepHAR: Identification of human activity in smart environments using ensemble of deep learning methods and motion sensor data
title_sort ensem deephar identification of human activity in smart environments using ensemble of deep learning methods and motion sensor data
topic Human activity
PC-HAR
Stacking
Ensem-DeepHAR
Motion sensor data
Person-independent
url http://www.sciencedirect.com/science/article/pii/S266591742400374X
work_keys_str_mv AT smmohidulislam ensemdeepharidentificationofhumanactivityinsmartenvironmentsusingensembleofdeeplearningmethodsandmotionsensordata
AT kamrulhasantalukder ensemdeepharidentificationofhumanactivityinsmartenvironmentsusingensembleofdeeplearningmethodsandmotionsensordata