University of Twente Student Theses
Corruption Risk Detection in Dutch Healthcare: A Red-Flag Approach
Peeters, I.A.C. (2025) Corruption Risk Detection in Dutch Healthcare: A Red-Flag Approach.
|
PDF
1MB |
| Abstract: | Public procurement is the process by which public authorities purchase goods and/ or services from the private sector. Procurement accounts for 42% of the Dutch GDP and healthcare has the biggest share. Unfortunately, around €4.4 billion of the Dutch GDP is lost to corruption and has serious consequences. Governmental institutions have developed indices to detect corrupt practices using red flags. Red flags are indicators of corruption but not a guarantee. The research paper aimed to predict the level of possible corruption in Dutch healthcare tenders through machine learning with past-defined red flags. The Random Forest algorithm made predictions of whether a tender would be considered corrupt based on levels. Four different types of predictive models were created, but two were analysed in detail. The recall, precision, accuracy, and F1-score measures were calculated to assess and evaluate the models. It is possible to predict the level of potential corruption of tenders in Dutch healthcare through machine learning with well-defined levels. The predictions are more accurate when predictors are coded numerically rather than dichotomously. Furthermore, a SHAP beeswarm plot was used to examine the influence of the predictors. The advertisement period and contract execution period were the most influential. |
| Item Type: | Essay (Master) |
| Faculty: | BMS: Behavioural, Management and Social Sciences |
| Subject: | 85 business administration, organizational science |
| Programme: | Business Administration MSc (60644) |
| Link to this item: | https://purl.utwente.nl/essays/104916 |
| Export this item as: | BibTeX EndNote HTML Citation Reference Manager |
Show download statistics for this publication
Show download statistics for this publicationRepository Staff Only: item control page