Advertisement
Published in:
2021 | OriginalPaper | Chapter
Well-Being and Efficiency in Financial Sector Analyzed with Multiclass Classification Machine Learning
Authors : Gregor Molan, Marija Molan
Published in: Proceedings of the 21st Congress of the International Ergonomics Association (IEA 2021)
Publisher: Springer International Publishing
Activate our intelligent search to find suitable subject content or patents.
Select sections of text to find matching patents with Artificial Intelligence. powered by
Select sections of text to find additional relevant content using AI-assisted search. powered by
Abstract
The main research goal is the identification of the most important well-being parameters determining worker’s efficiency in the financial sector from 2005 until 2020 in the time of great changes in the socio-economic situation in Slovenia. Data were collected in 2005, 2010, 2015, and 2020 - key periods of important milestones in economic growths and declines in Slovenia for 2723 financial workers. All data are analyzed using the ML classification method to identify the most important attributes (well-being parameters) determining worker’s efficiency. To prevent possible overfitting and/or underfitting there is a prescribed adequate tree depth for each decision tree as an additional domain knowledge added to the presented ML solution. We chose a binary decision tree learning, that is simple to understand and interpret with an ability to handle both numerical and categorical data.
Decision trees generated with the ML classification show, that workers with low efficiency (estimated as “poor” or “inadequate”) were mostly out of work. In 2005 the most important influential factor was psychological fatigue. From 2010 to 2020, physical fatigue was the most important influential factor.
External socio-economic factors determine the level of well-being and efficiency. Adequate level of well-being is the basis of workers’ efficiency and their health. The conclusions are based on the data from 2005 to 2020. Presented approach based on implementation of ML is a tool for identification of gripping points for intervention at work to maintain adequate level of workers well-being in a new working reality.