VISAGE: Detection and automatic classification of urban violence through social media data

• Matheus Henrique C. T. de Souza UFRJ http://orcid.org/0009-0000-3900-8374
• Eliel Roger da Silva UFRJ https://orcid.org/0009-0003-6171-4402
• Tiago Cruz de França UFRRJ https://orcid.org/0000-0002-1682-4587
• Jonice Oliveira UFRJ https://orcid.org/0000-0002-2495-1463

DOI: https://doi.org/10.5753/brasnam.2025.8111

Resumo

Urban violence remains a challenge for cities, requiring innovative approaches to provide actionable insights for authorities. This study presents VISAGE, a framework designed to detect and classify urban violence in social media textual data using an extension of the Quantification of Violence Scale (QoVS). Ten Machine learning models, including BERT and Random Forest, were evaluated on Twitter datasets from violent events in Brazil, with BERT achieving an F1-score of 0.86. Results demonstrate the feasibility of automating violence assessment from text. Limitations like dataset imbalance and labeling process are discussed with future work targeting real-time and multimodal analysis.

Referências

Ashmore, R., Calinescu, R., and Paterson, C. (2019). Assuring the Machine Learning Lifecycle: Desiderata, Methods, and Challenges. arXiv:1905.04223 [cs].

Belghit, M. T. (2024). Urban violence and criminality: a socio-ethnographic study. Management Intercultural, (52):17–24. Publisher: Romanian Foundation for Business Intelligence, Editorial Department.

Devlin, J., Chang, M.-W., Lee, K., and Toutanova, K. (2019). BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding. arXiv:1810.04805 [cs].

França, T., Gomes, J., and Oliveira, J. (2017). A Twitter Opinion Mining Gold Standard for Brazilian Uprising in 2013.

Gresenz, C. R., Singh, L., Wang, Y., Haber, J., and Liu, Y. (2023). Development and Assessment of a Social Media–Based Construct of Firearm Ownership: Computational Derivation and Benchmark Comparison. Journal of Medical Internet Research, 25:e45187.

Hawkins, D. S. (2023). “When you Search a Hashtag, it Feels Like You’re Searching for Death:”Black Twitter and Communication About Police Brutality Within the Black Community. Social Media + Society, 9(2):20563051231179705. Publisher: SAGE Publications Ltd.

Kozhamkulova, Z., Kirgizbayeva, B., Sembina, G., Smailova, U., Suleimenova, M., Keneskanova, A., and Baizakova, Z. (2023). MoveNET Enabled Neural Network for Fast Detection of Physical Bullying in Educational Institutions. International Journal of Advanced Computer Science and Applications, 14(5).

Krug, E. G., Mercy, J. A., Dahlberg, L. L., and Zwi, A. B. (2002). The world report on violence and health. Lancet (London, England), 360(9339):1083–1088.

Llauradó, J. M., Pujol, F. A., Tomás, D., Visvizi, A., and Pujol, M. (2023). Study of image sensors for enhanced face recognition at a distance in the Smart City context. Scientific Reports, 13(1). Type: Article.

Mussiraliyeva, S., Bagitova, K., and Sultan, D. (2023). Social Media Mining to Detect Online Violent Extremism using Machine Learning Techniques. International Journal of Advanced Computer Science and Applications, 14(6).

Patton, D. U., Lane, J., Leonard, P., Macbeth, J., and Smith Lee, J. R. (2017). Gang violence on the digital street: Case study of a South Side Chicago gang member’s Twitter communication. New Media & Society, 19(7):1000–1018. Publisher: SAGE Publications.

Pedersen, S. (2002). What is Political History Now? pages 36–56, London. Palgrave Macmillan UK. Book Title: What is History Now?

Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., Prettenhofer, P., Weiss, R., Dubourg, V., Vanderplas, J., Passos, A., Cournapeau, D., Brucher, M., Perrot, M., and Duchesnay, (2011). Scikit-learn: Machine Learning in Python. Journal of Machine Learning Research, 12(85):2825–2830.

Peixoto, B. M., Lavi, B., Dias, Z., and Rocha, A. (2021). Harnessing high-level concepts, visual, and auditory features for violence detection in videos. Journal of Visual Communication and Image Representation, 78:103174.

Phillips, D. (2018). Brazil military’s growing role in crime crackdown fuels fears among poor. The Guardian.

Ponce-León, E. and López-Nava, I. H. CICESE at DA-VINCIS 2023: Violent Events Detection in Twitter using Data Augmentation Techniques.

Publications, U. N. (2023). Global Study on Homicide 2023. United Nations Fund for Population Activities. Google-Books-ID: nACJ0AEACAAJ.

Pujol, F. A., Mora, H., and Pertegal, M. L. (2020). A soft computing approach to violence detection in social media for smart cities. Soft Computing, 24(15):11007–11017.

Souza, F., Nogueira, R., and Lotufo, R. (2020). BERTimbau: Pretrained BERT Models for Brazilian Portuguese. In Cerri, R. and Prati, R. C., editors, Intelligent Systems, pages 403–417, Cham. Springer International Publishing.

Steven Bird, Ewan Klein, and Edward Loper (2009). Natural Language Processing with Python[Book]. ISBN: 9780596516499.

Suryaningrum, K. M. (2023). Comparison of the TF-IDF Method with the Count Vectorizer to Classify Hate Speech. Engineering, MAthematics and Computer Science Journal (EMACS), 5(2):79–83. Number: 2.

Toktarova, A., Syrlybay, D., Myrzakhmetova, B., Anuarbekova, G., Rakhimbayeva, G., Zhylanbaeva, B., Suieuova, N., and Kerimbekov, M. (2023). Hate Speech Detection in Social Networks using Machine Learning and Deep Learning Methods. International Journal of Advanced Computer Science and Applications, 14(5).

Tyrer, P., Cooper, S., Herbert, E., Duggan, C., Crawford, M., Joyce, E., Rutter, D., Seivewright, H., O’Sullivan, S., Rao, B., Cicchetti, D., and Maden, T. (2007). The Quantification of Violence Scale: a Simple Method of Recording Significant Violence. International Journal of Social Psychiatry, 53(6):485–497. Publisher: SAGE Publications Ltd.