A Review of Supervised Learning for (Workplace) Mental Health and Wellbeing

Abstract

Mental health problems such as anxiety and loneliness have seen a dramatic increase, despite the tremendous growth in the healthcare industry in recent years. Traditional methods of diagnosing mental health and wellbeing issues can be effective, but they are often very time consuming and labour intensive and require active patient participation. Recent research has demonstrated the power of utilising artificial intelligence and physiological/psychological data to diagnose and predict the mental wellbeing of individuals. This paper systematically reviews the applications of supervised learning techniques to predict mental health and wellbeing constructs, such as stress and anxiety, and their potential to support workplace wellbeing. Given that data are an integral part of supervised learning approaches, this paper also reviews data collection practices and relevant considerations, such as bias implicitly expressed by data, especially in a workplace environment. Additionally, the paper investigates the ethical nature and aspects of explainability of wellbeing support systems, which are particularly sensitive in this subject area. Based on these research objectives, the gaps in the literature are identified and future research directions are recommended, including explainable AI, environmental factors in wellbeing prediction and the ethical deployment of such systems in workplace settings.

Received: 20 July 2024 | Revised: 30 June 2025 | Accepted: 9 October 2025

Conflicts of Interest

The authors declare that they have no conflicts of interest to this work.

Data Availability Statement

Data that support the findings of this study are openly available in MAHNOB HCI-Tagging at https://doi.org/10.1109/T-AF-FC.2011.25, reference number [10]; DEAP at https://doi.org/10.1109/T-AFFC.2011.15, reference number [66]; WESAD at https://doi.org/10.1145/3242969.3242985, reference number [82]; DAPPER at https://doi.org/10.6084/m9.figshare.13803185, reference number [12]; MMASH at https://doi.org/10.3390/data5040091, reference number [61]; UBFC-Phys at https://doi.org/10.1109/TAFFC.2021.3056960, reference number [80]; POPANE at https://doi.org/10.6084/m9.figshare.17061512, reference number [4].

Author Contribution Statement

Rohit Venugopal: Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Data curation, Writing – original draft, Writing – review & editing, Visualization, Project administration. Dan Roll: Conceptualization, Software, Formal analysis, Writing – original draft. Mark J. Flynn: Conceptualization, Methodology, Validation, Formal analysis, Investigation, Resources, Data curation, Writing – review & editing, Project administration. Phillip G. Bell: Conceptualization, Resources, Writing – review & editing, Supervision, Funding acquisition. Longzhi Yang: Conceptualization, Validation, Resources, Writing – review & editing, Visualization, Supervision, Funding acquisition.