Show simple item record

Physical Activity Recognition Based on a Parallel Approach for an Ensemble of Machine Learning and Deep Learning Classifiers

dc.contributor.authorMariem Abid
dc.contributor.authorAmal Khabou
dc.contributor.authorYoussef Ouakrim
dc.contributor.authorHugo Watel
dc.contributor.authorSafouene Chemcki
dc.contributor.authorAmar Mitiche
dc.contributor.authorAmel Benazza-Benyahia
dc.contributor.authorNeila Mezghani
dc.contributor.otherLaboratoire LIO, Centre de Recherche du CHUM, Montreal, QC H2X 0A9, Canada
dc.contributor.otherLaboratoire LIO, Centre de Recherche du CHUM, Montreal, QC H2X 0A9, Canada
dc.contributor.otherLaboratoire LIO, Centre de Recherche du CHUM, Montreal, QC H2X 0A9, Canada
dc.contributor.otherLaboratoire LIO, Centre de Recherche du CHUM, Montreal, QC H2X 0A9, Canada
dc.contributor.otherLICEF Institute, TELUQ University, Montreal, QC G1K 9H6, Canada
dc.contributor.otherINRS, Centre Énergie, Matériaux et Télécommunications, Montreal, QC G1K 9A9, Canada
dc.contributor.otherLR11TIC01, COSIM Lab., University of Carthage SUP’COM, El Ghazala 2083, Tunisia
dc.contributor.otherLaboratoire LIO, Centre de Recherche du CHUM, Montreal, QC H2X 0A9, Canada
dc.date.accessioned2025-10-09T04:57:41Z
dc.date.available2025-10-09T04:57:41Z
dc.date.issued01-07-2021
dc.identifier.urihttps://www.mdpi.com/1424-8220/21/14/4713
dc.identifier.urihttp://digilib.fisipol.ugm.ac.id/repo/handle/15717717/40856
dc.description.abstractHuman activity recognition (HAR) by wearable sensor devices embedded in the Internet of things (IOT) can play a significant role in remote health monitoring and emergency notification to provide healthcare of higher standards. The purpose of this study is to investigate a human activity recognition method of accrued decision accuracy and speed of execution to be applicable in healthcare. This method classifies wearable sensor acceleration time series data of human movement using an efficient classifier combination of feature engineering-based and feature learning-based data representation. Leave-one-subject-out cross-validation of the method with data acquired from 44 subjects wearing a single waist-worn accelerometer on a smart textile, and engaged in a variety of 10 activities, yielded an average recognition rate of 90%, performing significantly better than individual classifiers. The method easily accommodates functional and computational parallelization to bring execution time significantly down.
dc.language.isoEN
dc.publisherMDPI AG
dc.subject.lccChemical technology
dc.titlePhysical Activity Recognition Based on a Parallel Approach for an Ensemble of Machine Learning and Deep Learning Classifiers
dc.typeArticle
dc.description.keywordsmachine learning
dc.description.keywordsdeep learning
dc.description.keywordsbig data
dc.description.keywordsdata streams
dc.description.keywordsInternet of things
dc.description.keywordssensor data
dc.description.doi10.3390/s21144713
dc.title.journalSensors
dc.identifier.e-issn1424-8220
dc.identifier.oaioai:doaj.org/journal:a509cce4427046a5b8586e1fa9ad81f3
dc.journal.infoVolume 21, Issue 14


This item appears in the following Collection(s)

Show simple item record