Collection and analysis of human trafficking news from Brazilian websites
Resumo
Human trafficking (HT) is a serious crime, yet structured datasets on the topic remain scarce in Brazil. This paper presents a dataset of 3,556 news articles mentioning HT, collected from three Brazilian online portals. We describe a Python-based pipeline for collecting data from heterogeneous websites and discuss the main challenges encountered during extraction. The corpus was analyzed using AI-assisted labeling, semantic similarity, dimensionality reduction, clustering, and supervised classification. The dataset and collection methodology aim to support future research on HT, natural language processing, open data analysis, and public safety in the Brazilian context.
Palavras-chave:
human trafficking, Brazilian news, web crawling, natural language processing, text classification
Referências
Chajia, M. and Nfaoui, E. H. (2024). Customer Churn Prediction Approach Based on LLM Embeddings and Logistic Regression. Future Internet, 16(12). DOI: 10.3390/fi16120453.
da Costa, L. S., Oliveira, I. L., and Fileto, R. (2023). Text classification using embeddings: a survey. Knowledge and Information Systems, 65(7):2761–2803. DOI: 10.1007/s10115-023-01856-z.
Kanumolu, G., Madasu, L., Surange, N., and Shrivastava, M. (2024). TeClass: A human-annotated relevance-based headline classification and generation dataset for Telugu. In Proceedings of the Joint Intl. Conf. on Computational Linguistics, Language Resources and Evaluation, pages 15711–15720, Torino, Italia. ELRA and ICCL.
Passos, T. S., Santana, M. F. S., Cordero-Ramos, N., and Almeida-Santos, M. A. (2022). Profile of Reported Trafficking in Persons in Brazil Between 2009 and 2017. Journal of Interpersonal Violence, 37(11-12):NP8257–NP8273. DOI: 10.1177/0886260520976219.
Reimers, N. and Gurevych, I. (2021a). all-MiniLM-L6-v2: Lightweight Sentence Transformer Model. [link]. Sentence-Transformers model on Hugging Face Model Hub.
Reimers, N. and Gurevych, I. (2021b). paraphrase-multilingual-MiniLM-L12-v2: Multilingual Sentence Embedding Model. [link]. Sentence-Transformers model on Hugging Face Model Hub.
Rollo, F., Po, L., and Bonisoli, G. (2022). Online News Event Extraction for Crime Analysis. In Proceedings of the 30th Italian Symposium on Advanced Database Systems (SEBD 2022), volume 3194 of CEUR Workshop Proceedings, pages 257–264. CEUR-WS.org. URL: [link].
Saxena, V. K., Ashpole, B., Van Dijck, G., and Spanakis, G. (2025). ”MATCHED: Multi-modal Authorship-Attribution To Combat Human Trafficking in Escort-Advertisement Data”. In Findings of the Association for Computational Linguistics, pages 4334–4373, Vienna, Austria. ACL 2025. DOI: 10.18653/v1/2025.findings-acl.225.
United Nations Office on Drugs and Crime (2024). Global Report on Trafficking in Persons 2024. United Nations, Vienna. United Nations publication E.24.XI.11.
da Costa, L. S., Oliveira, I. L., and Fileto, R. (2023). Text classification using embeddings: a survey. Knowledge and Information Systems, 65(7):2761–2803. DOI: 10.1007/s10115-023-01856-z.
Kanumolu, G., Madasu, L., Surange, N., and Shrivastava, M. (2024). TeClass: A human-annotated relevance-based headline classification and generation dataset for Telugu. In Proceedings of the Joint Intl. Conf. on Computational Linguistics, Language Resources and Evaluation, pages 15711–15720, Torino, Italia. ELRA and ICCL.
Passos, T. S., Santana, M. F. S., Cordero-Ramos, N., and Almeida-Santos, M. A. (2022). Profile of Reported Trafficking in Persons in Brazil Between 2009 and 2017. Journal of Interpersonal Violence, 37(11-12):NP8257–NP8273. DOI: 10.1177/0886260520976219.
Reimers, N. and Gurevych, I. (2021a). all-MiniLM-L6-v2: Lightweight Sentence Transformer Model. [link]. Sentence-Transformers model on Hugging Face Model Hub.
Reimers, N. and Gurevych, I. (2021b). paraphrase-multilingual-MiniLM-L12-v2: Multilingual Sentence Embedding Model. [link]. Sentence-Transformers model on Hugging Face Model Hub.
Rollo, F., Po, L., and Bonisoli, G. (2022). Online News Event Extraction for Crime Analysis. In Proceedings of the 30th Italian Symposium on Advanced Database Systems (SEBD 2022), volume 3194 of CEUR Workshop Proceedings, pages 257–264. CEUR-WS.org. URL: [link].
Saxena, V. K., Ashpole, B., Van Dijck, G., and Spanakis, G. (2025). ”MATCHED: Multi-modal Authorship-Attribution To Combat Human Trafficking in Escort-Advertisement Data”. In Findings of the Association for Computational Linguistics, pages 4334–4373, Vienna, Austria. ACL 2025. DOI: 10.18653/v1/2025.findings-acl.225.
United Nations Office on Drugs and Crime (2024). Global Report on Trafficking in Persons 2024. United Nations, Vienna. United Nations publication E.24.XI.11.
Publicado
08/09/2026
Como Citar
NASCIMENTO, Ana Rosa A.; RIGATO, Matheus; SANTOS, João V. L. P. dos; NERES DE SOUSA, João V. C.; TRAINA JR., Caetano; CAZZOLATO, Mirela T..
Collection and analysis of human trafficking news from Brazilian websites. In: WORKSHOP DE TRABALHOS DE ALUNOS DA GRADUAÇÃO (WTAG) - SIMPÓSIO BRASILEIRO DE BANCO DE DADOS (SBBD), 41. , 2026, São Carlos/SP.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2026
.
p. 64-70.
DOI: https://doi.org/10.5753/sbbd_estendido.2026.249585.
