NewsFraud-SC: A Dataset of News about Corruption Extracted from SC Multiple Portals

• Ana Clara Stupp de Souza Universidade Federal de Santa Catarina (UFSC)
• Carina F. Dorneles Universidade Federal de Santa Catarina (UFSC)

DOI: https://doi.org/10.5753/dsw.2026.249443

Resumo

This article presents NewsFraud-SC, a dataset composed of news articles about illicit activities in the public sector, originating from multiple news portals in the state of Santa Catarina. With a total volume of 70 MiB, the dataset was constructed using an automated collection pipeline that integrates web scraping techniques, content cleaning via readability algorithms, and Named Entity Recognition (NER). Each record in the dataset is structured in JSON format, containing detailed metadata such as source, URL, and date, as well as previously extracted entities, including people, organizations, locations, and monetary values involved in the reported incidents. The work describes the collection methodology, descriptive statistics of the corpus, highlighting the distribution of topics and the density of entities. With 10,913 extracted entities and well-defined metadata, this corpus allows researchers to bypass the traditional data engineering bottleneck typically associated with web scraping and manual content cleaning. As a scientific contribution, the dataset is made available to the academic community, aiming to foster research in the extraction of complex relationships and government transparency technologies.

Palavras-chave: fraud, corruption, NER, dataset, news

Referências

Jost, L. (2006). Entropy and diversity. Oikos, 113(2):363–375.

Jurafsky, D. and Martin, J. H. (2026). Speech and Language Processing: An Introduction to Natural Language Processing, Computational Linguistics, and Speech Recognition, with Language Models. 3rd edition. Online manuscript released January 6, 2026.

Lyra, M. S., Damásio, B., Pinheiro, F. L., and Bacao, F. (2022). Fraud, corruption, and collusion in public procurement activities, a systematic literature review on data-driven methods. Applied Network Science, 7(1):83.

Sizov, S., Graupmann, J., and Theobald, M. (2003). From focused crawling to expert information: An application framework for web exploration and portal generation. In Proceedings VLDB Conference, pages 1105–1108. Morgan Kaufmann, San Francisco.

Souza, A. and Dorneles, C. (2025). Cono: Um coletor automatizado de notícias sobre corrupção em santa catarina. In Anais da XX Escola Regional de Banco de Dados, pages 129–132, Porto Alegre, RS, Brasil. SBC.

Weichselbraun, A., Hörler, S., Hauser, C., and Havelka, A. (2020). Classifying news media coverage for corruption risks management with deep learning and web intelligence. WIMS 2020, page 54–62, New York, NY, USA. Association for Computing Machinery.