University of Twente Student Theses
Predicting Job Flourishing through Algorithmic Management: a study of Machine Learning techniques
Kazaferi, Atis (2024) Predicting Job Flourishing through Algorithmic Management: a study of Machine Learning techniques.
PDF
4MB |
Abstract: | Nurturing a positive state of mind and engaging employees is vital for organizations to ensure, among others, employee retention and organizational performance. Drawing upon insights from Algorithmic Management (AM), this thesis aimed to define the best methods and criteria to predict job flourishing in employees of an educational institution in the Netherlands, utilizing secondary data. Job flourishing was defined as a broader construct of work engagement, and we based a binary classification on highly engaged employees. A few challenges were found in terms of defining the best criteria in the case of job flourishing, depending on organizational HR goals. Additionally, it was challenging to optimize the defined models due to the hidden underlying structure of the dataset. Results indicated the importance in flourishing prediction of item level constructs of structural job resources, job characteristics, and supervisor relations, supporting theoretical evidence. Moreover, of interest was the contribution of tenure and contract type, possibly due to organizational context, which led to a negative correlation of these socio-demographics to job flourishing prediction. Interestingly, the feature engineered logistic Regression model with Genetic Algorithm application was the best at predicting flourishing employees, while multiple models performed well at prediction of non-flourishing employees. We summarize our findings in a usage map for practitioners and researchers that want to bridge the gap between well-being and artificial intelligence tools, by developing AM tools contributing to the HR and well-being domain. |
Item Type: | Essay (Master) |
Faculty: | BMS: Behavioural, Management and Social Sciences |
Subject: | 70 social sciences in general |
Programme: | Business Administration MSc (60644) |
Link to this item: | https://purl.utwente.nl/essays/100593 |
Export this item as: | BibTeX EndNote HTML Citation Reference Manager |
Repository Staff Only: item control page