Data mining and analysis of criminal occurrences: a study on the Municipality of Divinópolis – Minas Gerais

Authors

DOI:

https://doi.org/10.33448/rsd-v10i13.20744

Keywords:

Criminal Analysis; Data Mining in Criminal Occurrences; Extraction of association rules.

Abstract

Crime is a problem that governments and societies face, with several investments being made in public security and intelligence, to try to punish and prevent criminal actions. This project aims to assist in public safety, applying data mining techniques on databases containing criminal incident bulletins in the city of Divinópolis-MG. The databases have occurrences from January 2016 to May 2019, provided by the Military Police of the State of Minas Gerais (PMMG). The methodological procedures, data selection was initially performed, followed by data pre-processing and transformation. Next, data mining techniques were applied, such as: Clustering and Extraction of Association Rules. In addition, the stage was dedicated to statistical analysis related to crimes of “Theft” and “Robbery”, as well as crimes related to violence against women. Among the results, two Association Rules stand out, found using the Apriori algorithm, “Night, Robbery” => “Male Victim” and “Firearms” => “No apparent injury”, in addition to the statistical rules were performed on the data, such as “analysis of the age groups of victims” and “distribution of criminal occurrences in the week”. Thus, it is concluded that this work reached the desired goals, bringing knowledge that can be used by public security agencies. Finally, it is suggested as future works, the expansion of the database, as well as working with latitude and longitude data for each criminal occurrence.

References

Atlas da Violência: Divinópolis ocupa a 13ª posição entre as cidades mineiras com a maior taxa de homicídios. (2019). G1 Centro-Oeste. https://g1.globo.com/mg/centro-oeste/noticia/2019/08/06/atlas-da-violencia-divinopolis-ocupa-a-13a-posicao-entre-as-cidades-mineiras-com-a-maior-taxa-de-homicidios.ghtml.

de Amo, S. (2004). Técnicas de mineração de dados. Jornada de Atualização em Informática. JAI – CSBC. Salvador – BA.

Decreto-Lei 2.848, de 07 de dezembro de 1940. Código Penal. Rio de Janeiro, 1940. http://www.planalto.gov.br/ccivil_03/decreto-lei/del2848compilado.htm

Diniz, A. M. (2005). Migração, desorganização social e violência urbana em Minas Gerais. Raega-O Espaço Geográfico em Análise, 9.

Dornelles, J. R. W. (2017). O que é crime. Brasiliense.

Engel, C. L. (2020). A violência contra a mulher. Instituto de Pesquisa Econômica Aplicada (IPEA). http://repositorio.ipea.gov.br/handle/11058/10313

Fayyad, U. M. et al. (1996). Advances in Knowledge Discovery and Data Mining. AAAIPress, The Mit Press.

Hammound, H. J. (2013). The Value of Big Data Isn’t the Data. Harvard Business Review. https://hbr.org/2013/05/the-value-of-big-data-isnt-the.html

Kodinariya, T. M., & Makwana, P. R. (2013). Review on determining number of Cluster in K-Means Clustering. International Journal, 1(6), 90-95.

Machado, F. N. R. (2018). Big Data O Futuro dos Dados e Aplicações. Saraiva Educação SA.

Marzan, C. S., Baculo, M. J. C., de Dios Bulos, R., & Ruiz Jr, C. (2017). Time series analysis and crime pattern forecasting of city crime data. In Proceedings of the International Conference on Algorithms, Computing and Systems (pp. 113-118).

Monitor da Violência: Mesmo com queda recorde de mortes de mulheres, Brasil tem alta no número de feminicídios em 2019. G1. https://g1.globo.com/monitor-da-violencia/noticia/2020/03/05/mesmo-com-queda-recorde-de-mortes-de-mulheres-brasil-tem-alta-no-numero-de-feminicidios-em-2019.ghtml

Ochi, L. S., Dias, C. R., & Soares, S. S. F. (2004). Clusterização em mineração de dados. Instituto de Computação-Universidade Federal Fluminense-Niterói, 1, 46.

Pereira, B. L., & Brandão, W. C. (2014). ARCA: Mining Crime Patterns Using Association Rules. In 11th International Conference Applied Computing. Porto (pp. 159-165).

Prado, K. H. J. et al. (2020). Applied intelligent data analysis to government data related to criminal incident: A systematic review. Journal of Applied Security Research, 15(3), 297-331.

Prado, K. H. J., & Júnior, M. C. (2020). Data Science aplicada à análise criminal baseada nos dados abertos governamentais de Minas Gerais. Research, Society and Development, 9(11), e36391110044-e36391110044.

Prodanov, C. C., & De Freitas, E.C. (2013). Metodologia do trabalho científico: métodos e técnicas da pesquisa e do trabalho acadêmico (2a ed.), Editora Feevale.

Romão, W., Niederauer, C. A., Martins, A., Tcholakian, A., Pacheco, R. C., & Barcia, R. M. (1999). Extração de regras de associação em C&T: O algoritmo Apriori. XIX Encontro Nacional em Engenharia de Produção, 34, 37-39.

Scalco, P. R. (2007). Criminalidade violenta em Minas Gerais: Uma proposta de alocação de recursos em segurança pública. Viçosa – MG.

Sevri, M., Karacan, H., & Akcayol, M. A. (2017). Crime analysis based on association rules using apriori algorithm. International Journal of Information and Electronics Engineering, 7(3), 99-102.

Tayal, D. K. et al. (2015). Crime detection and criminal identification in India using data mining techniques. AI & society, 30(1), 117-127.

Published

09/10/2021

How to Cite

MAMEDES, F. A. S.; SALDANHA, M. A.; RODRIGUES, E. S. Data mining and analysis of criminal occurrences: a study on the Municipality of Divinópolis – Minas Gerais. Research, Society and Development, [S. l.], v. 10, n. 13, p. e177101320744, 2021. DOI: 10.33448/rsd-v10i13.20744. Disponível em: https://rsdjournal.org/index.php/rsd/article/view/20744. Acesso em: 8 dec. 2021.

Issue

Section

Exact and Earth Sciences