Using beneficial ownership data for risk assessment in public procurement

Volkotrub, Antonina

Домівка
→
Students Research & Project Works | Роботи студентів
→
Факультет прикладних наук
→
Освітня програма наук про дані
→
2024
→
Перегляд матеріалів

dc.contributor.author	Volkotrub, Antonina
dc.date.accessioned	2024-08-23T11:38:46Z
dc.date.available	2024-08-23T11:38:46Z
dc.date.issued	2024
dc.identifier.citation	Volkotrub Antonina. Using beneficial ownership data for risk assessment in public procurement. Ukrainian Catholic University, Faculty of Applied Sciences, Department of Computer Sciences. Lviv 2024, 38 p.	uk
dc.identifier.uri	https://er.ucu.edu.ua/handle/1/4680
dc.language.iso	en	uk
dc.subject	beneficial ownership data	uk
dc.subject	risk assessment	uk
dc.subject	public procurement	uk
dc.title	Using beneficial ownership data for risk assessment in public procurement	uk
dc.type	Preprint	uk
dc.status	Публікується вперше	uk
dc.description.abstracten	Considering the importance of public procurement spending, we contend that pre- diction of corruption risk is essential and allows a decrease in the amount of money laundered from procurements. While several endeavours have been undertaken in this regard, they predominantly rely exclusively on procurement data overlooking externally available data. One such valuable source is beneficial ownership data that can provide valuable insights. Currently, the EU countries have to provide public ac- cess to this data and show who the real owner of the company is even when he or she is hiding behind offshores, which gives extensive sets of verified data. Conse- quently, this study seeks to explore the creation of features from publicly accessible beneficial ownership registers and assess whether these features enhance the pre- dictive accuracy of corruption risk models. It is also important to acknowledge even with the incorporation of the red flags identificators, it remains challenging to defini- tively label a specific tender as corrupt. Hence, we opt for a regression approach and predict the corruption risk index with the usage of non-linear models. Notably, prior research mostly relied on linear models for regression task. Therefore, this research is one of a few attempts to utilise supervised ML algorithms to identify corruption risk in public procurement.	uk