Physiological signals as predictors of mental workload: Evaluating single classifier and ensemble learning models
With a growing emphasis on cognitive processing in occupational tasks and the prevalence of wearable sensing devices, understanding and managing mental workload has broad implications for safety, efficiency, and well-being. This study aims to develop machine learning (ML) models for p...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Indonesian Institute of Science, Universitas Andalas
2023
|
| Subjects: | |
| Online Access: | https://umpir.ump.edu.my/id/eprint/44267/ |
| _version_ | 1848827321164759040 |
|---|---|
| author | Nailul, Izzah Sutarto, Auditya Purwandini Hendi, Ade Ainiyah, Maslakhatul Muhammad Nubli, Abdul Wahab |
| author_facet | Nailul, Izzah Sutarto, Auditya Purwandini Hendi, Ade Ainiyah, Maslakhatul Muhammad Nubli, Abdul Wahab |
| author_sort | Nailul, Izzah |
| building | UMP Institutional Repository |
| collection | Online Access |
| description | With a growing emphasis on cognitive processing in occupational tasks and the prevalence of wearable sensing devices, understanding and managing mental workload has broad implications for safety, efficiency, and well-being. This study aims to develop machine learning (ML) models for predicting mental workload using Heart Rate Variability (HRV) as a representation of the Autonomic Nervous System (ANS) physiological signals. A laboratory experiment, involving 34 participants, was conducted to collect datasets. All participants were measured during baseline, two cognitive tests, and recovery, which were further separated into binary classes (rest vs workload). A comprehensive evaluation was conducted on several ML algorithms, including both single (Support Vector Machine – SVM, and Naïve Bayes) and ensemble learning (Gradient Boost and AdaBoost) classiers and incorporating selected features and validation approaches. The ndings indicate that most HRV features differ signicantly during periods of mental workload compared to rest phases. The SVM classier with knowledge domain selection and leave-one-out cross-validation technique is the best model (68.385). These ndings highlight the potential to predict mental workload through interpretable features and individualized approaches even with a relatively simple model. The study contributes not only to the creation of a new dataset for specic populations (such as Indonesia) but also to the potential implications for maintaining human cognitive capabilities. It represents a further step toward the development of a mental workload recognition system, with the potential to improve decision-making where cognitive readiness is limited and human error is increased. |
| first_indexed | 2025-11-15T03:58:51Z |
| format | Article |
| id | ump-44267 |
| institution | Universiti Malaysia Pahang |
| institution_category | Local University |
| language | English |
| last_indexed | 2025-11-15T03:58:51Z |
| publishDate | 2023 |
| publisher | Indonesian Institute of Science, Universitas Andalas |
| recordtype | eprints |
| repository_type | Digital Repository |
| spelling | ump-442672025-09-22T08:21:40Z https://umpir.ump.edu.my/id/eprint/44267/ Physiological signals as predictors of mental workload: Evaluating single classifier and ensemble learning models Nailul, Izzah Sutarto, Auditya Purwandini Hendi, Ade Ainiyah, Maslakhatul Muhammad Nubli, Abdul Wahab Q Science (General) QP Physiology With a growing emphasis on cognitive processing in occupational tasks and the prevalence of wearable sensing devices, understanding and managing mental workload has broad implications for safety, efficiency, and well-being. This study aims to develop machine learning (ML) models for predicting mental workload using Heart Rate Variability (HRV) as a representation of the Autonomic Nervous System (ANS) physiological signals. A laboratory experiment, involving 34 participants, was conducted to collect datasets. All participants were measured during baseline, two cognitive tests, and recovery, which were further separated into binary classes (rest vs workload). A comprehensive evaluation was conducted on several ML algorithms, including both single (Support Vector Machine – SVM, and Naïve Bayes) and ensemble learning (Gradient Boost and AdaBoost) classiers and incorporating selected features and validation approaches. The ndings indicate that most HRV features differ signicantly during periods of mental workload compared to rest phases. The SVM classier with knowledge domain selection and leave-one-out cross-validation technique is the best model (68.385). These ndings highlight the potential to predict mental workload through interpretable features and individualized approaches even with a relatively simple model. The study contributes not only to the creation of a new dataset for specic populations (such as Indonesia) but also to the potential implications for maintaining human cognitive capabilities. It represents a further step toward the development of a mental workload recognition system, with the potential to improve decision-making where cognitive readiness is limited and human error is increased. Indonesian Institute of Science, Universitas Andalas 2023-12-20 Article PeerReviewed pdf en cc_by_nc_sa_4 https://umpir.ump.edu.my/id/eprint/44267/1/Physiological%20signals%20as%20predictors%20of%20mental%20workload.pdf Nailul, Izzah and Sutarto, Auditya Purwandini and Hendi, Ade and Ainiyah, Maslakhatul and Muhammad Nubli, Abdul Wahab (2023) Physiological signals as predictors of mental workload: Evaluating single classifier and ensemble learning models. Jurnal Optimasi Sistem Industri, 22 (2). pp. 81-98. ISSN 2088-4842. (Published) https://doi.org/10.25077/josi.v22.n2.p81-98.2023 https://doi.org/10.25077/josi.v22.n2.p81-98.2023 https://doi.org/10.25077/josi.v22.n2.p81-98.2023 |
| spellingShingle | Q Science (General) QP Physiology Nailul, Izzah Sutarto, Auditya Purwandini Hendi, Ade Ainiyah, Maslakhatul Muhammad Nubli, Abdul Wahab Physiological signals as predictors of mental workload: Evaluating single classifier and ensemble learning models |
| title | Physiological signals as predictors of mental workload: Evaluating single classifier and ensemble learning models |
| title_full | Physiological signals as predictors of mental workload: Evaluating single classifier and ensemble learning models |
| title_fullStr | Physiological signals as predictors of mental workload: Evaluating single classifier and ensemble learning models |
| title_full_unstemmed | Physiological signals as predictors of mental workload: Evaluating single classifier and ensemble learning models |
| title_short | Physiological signals as predictors of mental workload: Evaluating single classifier and ensemble learning models |
| title_sort | physiological signals as predictors of mental workload: evaluating single classifier and ensemble learning models |
| topic | Q Science (General) QP Physiology |
| url | https://umpir.ump.edu.my/id/eprint/44267/ https://umpir.ump.edu.my/id/eprint/44267/ https://umpir.ump.edu.my/id/eprint/44267/ |