Services on Demand
Journal
Article
Indicators
- Cited by SciELO
- Access statistics
Related links
- Cited by Google
- Similars in SciELO
- Similars in Google
Share
Revista científica
Print version ISSN 0124-2253On-line version ISSN 2344-8350
Abstract
LUNA-ORTEGA, Julio-César; COBOS-LOZADA, Carlos-Alberto and MENDOZA-BECERRA, Martha-Eliana. Model to Define Corruption Indices in Contracting Announcements in Colombia Based on Big Data and Natural Language Processing. Rev. Cient. [online]. 2023, n.46, pp.77-92. Epub Apr 26, 2023. ISSN 0124-2253. https://doi.org/10.14483/23448350.19640.
This research work proposes a macro-model that allows detecting different probable crimes or anomalies related to corruption in public procurement processes in Colombia. To this effect, the proposed model consists of five main components: 1) specialized services that seek to identify specific situations of probable corruption (three services were proposed: the detection of similarity between technical proposals, the detection of offer manipulation, and the detection of cartels); 2) transversal services that support the transformation of the model into a software tool; 3) additional services where general situations of probable corruption are addressed, specifically the citizen alert service; 4) the explicit relationships between services; and 5) the global output of the model. In the practical experimentation stage, two of the services proposed in this research were put to the test in various scenarios. With the results obtained by some of the classical metrics of the area, the quality of the prediction obtained by the services was determined.
Keywords : contracting; corruption; indices; model; services..