Ambiente inteligente de análise de riscos de fraudes em contratos públicos

• Ano de defesa: 2023
• Autor(a) principal: Costa, Luan Bruno Barbosa de Souza
• Orientador(a): Rodrigues Júnior, Methanias Colaço
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Programa de Pós-Graduação: Pós-Graduação em Ciência da Computação
• Palavras-chave em Português: Inteligência artificial; Crimes contra a administração pública; Corrupção; Tipologia criminal
• Palavras-chave em Inglês: Data science; Artificial intelligence; Corruption; Criminal typology; Redflag
• Área do conhecimento CNPq: CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO
• Link de acesso: https://ri.ufs.br/jspui/handle/riufs/19468

Context: The management of public resources is subject to illegal acts, which deprive society of the full use of these resources. The identification of such acts, in turn, depends on the analysis of a large amount of data. Objective: The objective of this work is the creation and evaluation of a tool capable of reducing the scope of search for irregularities, by means of signs of possible fraud (redflags) in contracts signed by the state, in the context of the servers of the Special Action Group to Combat Organized Crime (GAECO) of the Prosecutor of the State of Sergipe (MPSE). Methodology: Using the research-action methodology, which included the conduct of a systematic mapping, a list of fraud risk types was identified and automated in the Galactus Snitch application, which was evaluated through a qualitative survey. Results: The systematic mapping carried out identified 19 works according to the criteria informed, in which about 80% of these sought to indicate some type of fraud in bidding processes, obtaining accuracy between 72% and 99%, with different techniques to address the problem. Since then, the Galactus Snitch was developed and the survey applied revealed a high acceptability of the tool, for which 66.7% of the servers stated fully agree that the system is able to reduce by more than 90% the scope of search for irregularities, by pointing out possible fraudulent schemes. In addition, 91.7% of servers also stated that they fully agree that the types identified are of great relevance for the system to the expected goal. Conclusions: Using data analysis from different sources of information, the work presents an important contribution to the process of identifying fraudulent schemes in public procurement, as, according to the respondents, it is able to reduce the scope of the search for fraud.