Automated Public Transparency Auditing Using Web Scraping and Large Language Models for PNTP Compliance

• Iasmim Maria Freire da Silva Torres UFCG
• Cláudio de Souza Baptista UFCG
• André Luiz Firmino Alves IFPB
• Livia Aniely de Oliveira Almeida UFCG
• Eniedson Fabiano Pereira da Silva Júnior UFCG

DOI: https://doi.org/10.5753/lasdigov.2026.22521

Resumo

Public transparency, ensured by the Access to Information Law (Law No. 12.527/2011) and the National Public Transparency Program (Programa Nacional de Transparência Pública – PNTP), is still predominantly evaluated through manual processes, making it costly and difficult to scale. This study proposes an automated pipeline for auditing transparency portals of the Legislative Branch, combining web scraping and Large Language Models (LLMs) to assess compliance with PNTP criteria. Validation against human evaluations showed an average accuracy of 85.81%. The approach contributes to making transparency oversight more scalable, standardized, and traceable.

Referências

ATRICON (2025). Cartilha do programa nacional de transparência pública.

Hsu, E. and Anex-Ries, Q. (2025). A framework for assessing ai transparency in the public sector.

Iversen, O. and Huang, L. (2026). Leveraging large language models for bim-based automated compliance checking. Automation in Construction, 182:106707.

Jácome Filho, E. A. and Macêdo, J. M. A. (2022). Analysis of responsiveness and usability in websites serving public transparency in a mobile environment: Case study in the state of paraíba through heuristic evaluation. In International Conference on Human-Computer Interaction, Cham. Springer International Publishing.

Kingsman, N. et al. (2022). Public sector ai transparency standard: Uk government seeks to lead by example. Discover Artificial Intelligence, 2(1).

Lourenço, R. P. (2023). A framework for public eservices transparency. International Journal of Electronic Government Research (IJEGR), 19(1):1–19.

Martins, C. P. P. and Silva, G. S. (2024). Método e aplicação de automatização para download de dados abertos em portais de transparência. In Simpósio Brasileiro de Sistemas de Informação (SBSI). Sociedade Brasileira de Computação (SBC).

Odilla, F. (2023). Bots against corruption: Exploring the benefits and limitations of ai-based anti-corruption technology. Crime, Law and Social Change, 80(4):353–396.

Paiva, M. E. R., Ribeiro, L. L., and Gomes, J. W. F. (2021). O tamanho do governo aumenta a corrupção? uma análise para os municípios brasileiros. Revista de Administração Pública, 55(2):272–291.

Saldanha, D. M. F., Dias, C. N., and Guillaumon, S. (2022). Transparency and accountability in digital public services: Learning from the brazilian cases. Government Information Quarterly, 39(2):101680.

Santos, M. S. X. and Diniz, M. B. (2024). Determinantes da corrupção municipal no setor educacional brasileiro entre os anos de 2011 e 2014. Semestre Econômico, 27(62).