Effects of Class Imbalance in Unsupervised Human Activity Recognition for Office Work Task Characterization

Sara Santos, Phillip Probst, Luís Silva, Hugo Gamboa

2025

Abstract

Office workers spend most of their time sitting, often with rigid postures, for prolonged periods of time. This has been recognized by the European Union as a risk factor for work-related musculoskeletal disorders. To study work activities and their distribution over time, Human Activity Recognition (HAR) techniques need to be implemented. Since supervised learning techniques require labeled data and large datasets for training, unsupervised learning is a viable alternative for HAR. However, these models may be affected by the highly imbalanced distribution of activities typically observed in office workers. Considering this, this work studied the impact of data imbalance on clustering performance when the dataset is comprised of 33 %, 50 %, 70 %, and 90 % of sitting activity. Office activities were collected from 19 subjects and three traditional clustering models were employed. KMeans and Gaussian Mixture Model were more affected than Agglomerative Clustering, which seems to be more robust to data imbalance. With 90 % of sitting time, all three models performed poorly, which emphasizes the need for clustering models that can handle highly imbalanced data.

Paper Citation

in Harvard Style

Santos S., Probst P., Silva L. and Gamboa H. (2025). Effects of Class Imbalance in Unsupervised Human Activity Recognition for Office Work Task Characterization. In Proceedings of the 18th International Joint Conference on Biomedical Engineering Systems and Technologies - Volume 1: BIOSIGNALS; ISBN 978-989-758-731-3, SciTePress, pages 988-995. DOI: 10.5220/0013266300003911

in Bibtex Style

@conference{biosignals25,
author={Sara Santos and Phillip Probst and Luís Silva and Hugo Gamboa},
title={Effects of Class Imbalance in Unsupervised Human Activity Recognition for Office Work Task Characterization},
booktitle={Proceedings of the 18th International Joint Conference on Biomedical Engineering Systems and Technologies - Volume 1: BIOSIGNALS},
year={2025},
pages={988-995},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013266300003911},
isbn={978-989-758-731-3},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 18th International Joint Conference on Biomedical Engineering Systems and Technologies - Volume 1: BIOSIGNALS
TI - Effects of Class Imbalance in Unsupervised Human Activity Recognition for Office Work Task Characterization
SN - 978-989-758-731-3
AU - Santos S.
AU - Probst P.
AU - Silva L.
AU - Gamboa H.
PY - 2025
SP - 988
EP - 995
DO - 10.5220/0013266300003911
PB - SciTePress