Applying data mining algorithms to encourage mental health disclosure in the workplace. (24th March 2021)

• Record Type: Journal Article
• Authors: Singer, Gonen; Golan, Maya
• Is Part Of: International journal of business information systems. Volume 36:Number 4(2021)
• Page Start: 553
• Page End: 571
• Publication Date: 2021-03-24
• Subjects: data mining -- decision tree -- classification -- mental illness -- mental health disclosure; Information storage and retrieval systems -- Business -- Periodicals; Business -- Data processing -- Periodicals; Management information systems -- Periodicals; System design -- Periodicals; System analysis -- Periodicals; 651.8
• Languages: English
• ISSNs: 1746-0972

Abstract

The importance of sharing mental health issues with supervisors is well established. However, the decision to disclose such intimate information is complex and is influenced by many intrinsic and extrinsic variables. The purpose of this study is to use machine learning algorithms to develop a tool that supervisors may use to enhance disclosure of mental health issues among their employees. Several interpretable machine learning algorithms are established based on a Kaggle dataset of more than 1, 400 participants that measures attitudes towards mental health and prevalence of mental health disorders in the tech workplace. The C4.5 algorithm is chosen as the best classifier of willingness to disclose a mental health disorder to supervisors, based on a variety of classification performance measures. Tailored intervention programs are applied and are shown to have the potential to increase the probability of disclosure by between 20% and 60%.