Application of Machine Learning in the Prediction of Employee Satisfaction with Support Provided in a National Park

Martha Chadyiwa (), Juliana Kagura and Aimee Stewart
Additional contact information
Martha Chadyiwa: Department of Environmental Health, Faculty of Health Sciences, University of Johannesburg
Juliana Kagura: University of the Witwatersrand
Aimee Stewart: School of Therapeutic Sciences, Faculty of Health Science, School of Public Health, University of the Witwatersrand

Chapter Chapter 6 in Tourism and Hospitality for Sustainable Development, 2024, pp 107-119 from Springer

Abstract: Abstract Employee satisfaction with support provided after an occupational injury in national parks has an impact on both the employee and the organisation. Machine learning (ML) classification provides techniques that can be useful in predicting employee satisfaction with the support provided after an occupational injury. The main objective of this study was to develop a model that can predict employee satisfaction with the support provided after an occupational injury in the Kruger National Park. The data were collected from 241 employees who experienced an occupational injury between 2006 and 2016 in the Kruger National Park in South Africa. Features used as inputs to the machine learning algorithm included employee characteristics such as job title, years of experience, salary, and performance evaluation. Finally, the nature of work, workload, job satisfaction, company culture, and opportunities for growth and development were also considered. The Support Vector Machines (SVM), k-Nearest Neighbours (k-NN), XGB Classifier, and the Deep Learning Neural Network (DNN) had a good performance in predicting employee satisfaction with the support provided. The k-NN, using training data with Random oversampling, had the highest overall accuracy of 71%. The model predicted employees who did not have adequate employee satisfaction with the support provided with a precision of 72%, a recall of 80%, and an f1-score of 76%. The XGB classifier with an f-score of 69% and the XGB classifier using data with random oversampling with an f1-score of 69% had the highest accuracy in predicting whether employees were adequately satisfied with the support provided. Discussion and conclusion: The findings suggest that both the XGB classifier without oversampling and the XGB classifier with oversampling techniques perform equally well in predicting whether employees are satisfied with the support provided. Therefore, management at the Kruger National Park can use these predictions to focus their interventions on the support provided to employees after an occupational injury to enhance employee satisfaction.

Keywords: Prediction; Support; Deep learning neural networks; Support vector machine; XG boost; k-nearest neighbours; Satisfaction; Accuracy (search for similar items in EconPapers)
Date: 2024
References: Add references at CitEc
Citations:

There are no downloads for this item, see the EconPapers FAQ for hints about obtaining it.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:sprchp:978-3-031-63077-4_6

Ordering information: This item can be ordered from
http://www.springer.com/9783031630774

DOI: 10.1007/978-3-031-63077-4_6

Access Statistics for this chapter

More chapters in Springer Books from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().