Publicação
Mapping a Co-competition Network of the Portuguese Public Procurement Market
| Resumo: | Fraud in the public procurement sector is a persistent and challenging issue which can have harmful consequences in society, hindering economic and social growth, and corroding public trust in the government. This thesis aims to use network science methods to analyze and detect fraudulent activities by utilizing a Portuguese data set composed of thousands of public contracts between 2009 to 2023. The approach consists of utilizing the firms' co-occurring similarities to build a co-competition network and to detect communities of firms which are likely competitors and collaborators. With the aid of additional indicators such as activity diversity, regional diversity and single bidders, the characterization of the communities allows to emphasize groups of firms which are more probably engaging in illegal schemes such as collusion or market distortion. Furthermore, the work reinforces the efficiency of a network-based approach to unveil complex relationships, and in detecting and mitigating fraud. Finally, the findings and insights revealed in this study aim to aid in the development of fraud detection techniques and to provide a better understanding of the Portuguese public procurement environment. |
|---|---|
| Autores principais: | Silva, Gonçalo Baptista Perdigão |
| Assunto: | Public Procurement Network Analysis Co-competition Network Portugal Clustering Corruption Community Detection SDG 8 - Decent work and economic growth SDG 16 - Peace, justice and strong institutions SDG 17 - Partnerships for the goals |
| Ano: | 2024 |
| País: | Portugal |
| Tipo de documento: | dissertação de mestrado |
| Tipo de acesso: | acesso aberto |
| Instituição associada: | Universidade Nova de Lisboa |
| Idioma: | inglês |
| Origem: | Repositório Institucional da UNL |
| Resumo: | Fraud in the public procurement sector is a persistent and challenging issue which can have harmful consequences in society, hindering economic and social growth, and corroding public trust in the government. This thesis aims to use network science methods to analyze and detect fraudulent activities by utilizing a Portuguese data set composed of thousands of public contracts between 2009 to 2023. The approach consists of utilizing the firms' co-occurring similarities to build a co-competition network and to detect communities of firms which are likely competitors and collaborators. With the aid of additional indicators such as activity diversity, regional diversity and single bidders, the characterization of the communities allows to emphasize groups of firms which are more probably engaging in illegal schemes such as collusion or market distortion. Furthermore, the work reinforces the efficiency of a network-based approach to unveil complex relationships, and in detecting and mitigating fraud. Finally, the findings and insights revealed in this study aim to aid in the development of fraud detection techniques and to provide a better understanding of the Portuguese public procurement environment. |
|---|