Trust-ME: Trustworthy workplace well-being and productivity improvement
Weave research project. Collaboration between Jožef Stefan Institute and Università della Svizzera italiana.
Project description
By combining expertise in sensing technology, privacy, user experience design, and explainable AI, this project will harness the recent trends in these fields as well as psychology and physiology, and develop a new approach for monitoring job satisfaction, wellbeing, and productivity that will (1) offer insights into their complex relations to support interventions to improve them, and (2) employ privacy-aware development methods for creating powerful yet explainable AI that provides productivity insights.
Work package 1: Problem definition
Review of psycho-physiological constructs of interest and their measurement.
[TODO add realization]
Work package 2: Data collection and validation
Definition of the data-collection protocol, implementation of the data-collection hardware and software setup, auxiliary and main data collection.
[TODO add realization]
Work package 3: Federated machine learning
Baseline ML, ML for interrelated psychological constructs, federated learning, multi-modal federated learning.
[TODO add realization]
Work package 4: Trustworthy and explainable AI
Privacy and security, explainability methods, preventing bias, from explanation to intervention.
[TODO add realization]
Work package 5: Project management and dissemination
Project management, dissemination, other activities.
[TODO add realization]
Project partners
The TRUST-ME project will utilize the expertise from two separate institutions, USI (Switzerland) – experts in security, privacy and trustworthy AI – and JSI (Slovenia) – experts in machine learning for sensing and modelling psycho-physiological constructs including employee wellbeing.
The proposed project will thus explore the problem domain via two key strands of investigation: (1) multimodal monitoring and modelling of job satisfaction and productivity (led by JSI) and (2) multimodal, secure, private, and explainable AI for productivity assessment (led by USI).
Team USI:
prof. dr. Marc Langheinrich (lead)
dr. Martin Gjoreski
dr. Pietro Barbiero
Mohan Li
Daniil Kirilenko
Team JSI:
prof. dr. Mitja Luštrek (lead)
Gašper Slapničar
Zoja Anžur
Sebastijan Trojer
Tomi Božak
Project-related publications
- Z. Anžur, K. Žinkovič, J. Lukan, P. Barbiero, G. Slapničar, M. Li, M. Gjoreski, M. E. Debus, S. Trojer, M. Luštrek, and M. Langheinrich. “A Review of Methods for Unobtrusive Measurement of Work-Related Well-Being”. IEEE Transactions on Affective Computing, 2025 [under review]
- Trojer, S., Anžur, Z., Gjoreski, M., Slapničar, G., & Luštrek, M. (2024, October). “A Feature-Based Approach for Subtle Emotion Recognition in Realistic Scenarios”. In Companion of the 2024 on ACM International Joint Conference on Pervasive and Ubiquitous Computing (pp. 398-404).
- Trojer, S., Luštrek, M., Anžur, Z., Slapničar, G. “Comparison of feature- and embedding-based approaches for audio and visual emotion classification”. In Vol. A : proceedings of the 27th International Multiconference Information Society – IS 2024.
- Slapničar, G., Anžur, Z., Trojer, S., & Luštrek, M. (2024). Contact-Free Emotion Recognition for Monitoring of Well-Being: Early Prospects and Future Ideas. In Intelligent Environments 2024: Combined Proceedings of Workshops and Demos & Videos Session (pp. 58-67). IOS Press.
- Božak, T., Luštrek, M., Slapničar, G. “Feature-based emotion classification using eye-tracking data”. In Vol. A : proceedings of the 27th International Multiconference Information Society – IS 2024.
- Li, M., Gjoreski, M., Barbiero, P., Slapničar, G., Luštrek, M., Lane, N. D., & Langheinrich, M. (2025). A Survey on Federated Learning in Human Sensing. arXiv preprint arXiv:2501.04000.
Funding
Trust-ME project is a bilateral Weave project, funded by the Slovenian Agency of Research and Innovation (ARIS) under grant agreement N1-0319 and by the Swiss National Science Foundation (SNSF) under grant agreement 214991.
