Inteligencia artificial para detectar corrupción en la administración pública municipal de Colombia
This research evaluates the application of machine learning algorithms for the early detection of corruption in the Colombian municipal administration. Two approaches are considered to achieve the objective: (i) An evaluation of supervised machine learning for the prediction of variables related to...
| Autor: | |
|---|---|
| Tipo de recurso: | tesis de maestría |
| Estado: | Versión publicada |
| Fecha de publicación: | 2021 |
| País: | Colombia |
| Institución: | Universidad de los Andes |
| Repositorio: | Séneca: repositorio Uniandes |
| Idioma: | español |
| OAI Identifier: | oai:repositorio.uniandes.edu.co:1992/50908 |
| Acceso en línea: | http://hdl.handle.net/1992/50908 |
| Access Level: | acceso abierto |
| Palabra clave: | Corrupción administrativa Aprendizaje automático (Inteligencia artificial) Administración pública Economía |
| Sumario: | This research evaluates the application of machine learning algorithms for the early detection of corruption in the Colombian municipal administration. Two approaches are considered to achieve the objective: (i) An evaluation of supervised machine learning for the prediction of variables related to corruption and, (ii) an evaluation of unsupervised learning for the segmentation of relative risk of corruption. The results show that, despite the acceptable results of predictions, unsupervised machine learning is emerging as the most useful tool for the early detection of municipal corruption in Colombia. Based on these findings, I created a Relative Risk of Municipal Corruption Index for the period 2020-2023. This index should be useful for the control bodies to target the investigation and prevention efforts in corruption. |
|---|