In 2024, the highest amount of crimes recorded in the state of São Paulo was for thievery, excluding vehicle thievery, with more than half a million cases. The second most common crime was robberies, with nearly 189,000 incidences.

Abstract This article analyzes the relationships between crime, violence and cities, a traditional object of the social sciences and urban studies. With the city of São Paulo as the background and reported murders as the parameter, the article registers the evolution of vital and criminal records and questions the hypothesis that analyzes and explains crime distribution, in urban territories, based on the dualistic and binary logic between non-violent and violent neighborhoods, between center and periphery. Evaluating the stock of knowledge available in this field of study and discussing methodological issues related to the quality of the data that support analyses, the article proposes some methodological approaches that, today, can better explain crime distribution in a diversified and complex metropolitan fabric.

In the state of São Paulo, Brazil, the population in 2021 was composed by 63.7 percent of people who identified as white. However, this ethnic group only represented 31 percent of the civilians killed by security agents. Meanwhile, 69 percent of civilian deaths caused by the police were black people, who constituted a little more than a third of the state's population. Moreover, the share of people of black ethnicity killed by the police in the state's capital reached nearly 70 percent of the total that year.

In 2024, six of the eight Brazilian cities with the highest homicide rates were in the Northeast. Feira da Santana led the ranking of the most violent city in Brazil, with a murder rate of ***** per 100,000 inhabitants. It was followed followed by Recife, with a homicide rate of more than ** per 100,000 inhabitants. In Latin America and the Caribbean, Feira da Santana was the **** most deadly city.

Although Brazil remains severely affected by civil violence, the state of São Paulo has made significant inroads into fighting criminality. In the last decade, São Paulo has witnessed a 70% decline in homicide rates, a result that policy-makers attribute to a series of crime-reducing measures implemented by the state government. While recent academic studies seem to confirm this downward trend, no estimation of the total impact of state policies on homicide rates currently exists. The present article fills this gap by employing the synthetic control method to compare these measures against an artificial São Paulo. The results indicate a large drop in homicide rates in actual São Paulo when contrasted with the synthetic counterfactual, with about 20,000 lives saved during the period. The theoretical usefulness of the synthetic control method for public policy analysis, the role of the Primeiro Comando da Capital as a causal mediator, and the practical implications of the security measures taken by the São Paulo state government are also discussed.

In 2024, São Paulo was the Brazilian state that had the highest number of car thefts and break-ins, totaling approximately 125,690. It was followed by Rio de Janeiro state with 48,270 cases reported. Over 344,000 vehicle thefts and break-ins in the South American country during 2024 were recorded.

The Brazilian Crimes Dataset (BCD) contains crime records and pre-processing procedures used in our experiments on crime analysis and prediction [1]. In particular, we proposed an approach to predict crimes and evaluated it by using crime records crawled from the brazilian web site Onde Fui Roubado.

Please consider citing the following references if you found this dataset useful:

[1] Úrsula Rosa Monteiro de Castro, Marcos Wander Rodrigues, Wladmir Cardoso Brandão. Predicting crime by exploiting supervised learning on heterogeneous data. In: Proceedings of the 22nd International Conference on Enterprise Information Systems (ICEIS'20), 2020.

[2] Úrsula Rosa Monteiro de Castro, Wladmir Cardoso Brandão. (2020). BCD: Brazilian Crimes Dataset (Version 1.0) [Data set]. Zenodo.

SPSafe Dataset

  About the Dataset

SPSafe is a standardized dataset of crime incident reports registered in the state of São Paulo, Brazil, from 2003 to 2022. This dataset was created to address and resolve issues of standardization, consistency, and heterogeneity found in the original data provided by the Secretariat of Public Security of the State of São Paulo (SSP/SP). The primary goal of SPSafe is to provide a high-quality, easy-to-use resource for public safety… See the full description on the dataset page: https://huggingface.co/datasets/joaomjr/spsafe.

How do socially relevant attributes influence juvenile criminal sentencing? While judicial decisions should, in principle, be fully based on legally relevant factors such as the seriousness of the offense and the defendant’s criminal record, I ask whether and how extralegal characteristics related to the adolescent’s position in structural relations affect the decision-making process. I propose a mixed-methods design to study mechanisms of criminal sentencing. Using data from a representative sample of the São Paulo juvenile justice system records, I estimate mixed-effects logistic models to assess the probability of being sentenced to confinement given certain extralegal attributes, while controlling for legally relevant variables. Interaction effects show that adolescents registered as full-time students and classified as drug users are more likely to be sentenced to confinement than their counterparts, even when the arraignment is the same. The second step involved weekly visits to the juvenile courthouse in São Paulo over four months to observe judicial hearings. Prosecutors are central to the decision-making process. The standard decision-making mechanism is based on police documents and legally relevant information. When there is a rupture in the definition of the situation (usually when non-minority defendants enter the courtroom), a new mechanism emerges and more lenient decisions are made.

Social, demographic, clinical, and criminal characteristics of the prison population of São Paulo state, Brazil (N = 1809).

The number of femicides, murders committed against women because of their gender, in Brazil slightly increased in 2024 in comparison to the previous year. Throughout 2024, ***** femicides were registered in Brazil, up from ***** recorded one year before. That year, the Brazilian state with the highest number of femicides was São Paulo.

During the Brazilian Carnival season of 2020, more than 2.4 thousand were arrested throughout the state of São Paulo. Nearly one quarter of those arrests corresponded to people who had already been convicted but were fleeing from justice.

Abstract In this paper we analyzed data from police investigations and trials of intentional homicides shelved in Belo Horizonte between 2003 and 2013 in order to understand the clearance’s determinants. As independent variables, we used the characteristics of victim, crimes’ attributes and police procedures. The results inform that the homicides features and police truth-finding methods (centered on the flagrante delicto and the presence of eyewitnesses) were the variables that explained the clearance likelihood. However, the length of time may affect this result, since police inquiries that are not completed within five year are not likely to be trialed.

In 2023, São Paulo was the Brazilian state with the highest number of violent deaths of LGBT+ people. Out of the *** deaths reported that year, 27were registered in that state. Ceará ranked second, with ** deaths. Further, most cause of these deaths were homicides.

Data about homicide rate and the relationship with drug market and socioeconomic factors

In 2024, Brazil's homicide rate reached **** incidents per 100,000 people. This is the lowest figure recorded in the country since 2012.

In 2024, there were a total of 2,532 shootings registered in Rio de Janeiro, Brazil. March was the month with the most shootings in each of the given years, except for 2024 - in that year, there were 254 occurrences, or about 10 percent of the year's total. The lowest number of shootings in the given period occurred in August 2023 when 153 cases were recorded. Police violence in Rio In 2022, the Supreme Court ordered the state government of Rio de Janeiro to come up with a plan to reduce police lethality, as the level of violence in police actions was deemed unacceptable, due to high numbers of casualties and human rights violations. The number of civilians killed as a result of police intervention more than quadrupled between 2013 and 2019, reaching a record number of 1,814 that year. Despite the decrease in comparison to 2019, every year from 2020 to 2022 saw more than 1,200 civilians being killed. Furthermore, it is deemed that there is structural racism in the actions of security forces. For instance, 80 percent of the deaths caused by police interventions in the state during 2023 were of people of color. Shootings and massacres in Rio Civil society and public institutions have made proposals to alleviate this situation. One of them is the ADPF 635 (Allegation of Violation of a Fundamental Precept), also known as ADPF Favelas Case, presented by the Brazilian Socialist Party, and whose preliminary approval took place in June 2020. The measure restricted unplanned police operations in the favelas during the pandemic. Despite its frequent violations, it showed evident results. Shootings fell from 7,368 in 2019 to less than 3,000 in 2024. Over one third of documented shootings in 2024 were due to police operations, while 288 were motivated by murder or attempted murder, the second most common reason. In March 2022, the government of Rio de Janeiro published a plan to reduce deaths during police operations. That year, the State of Rio de Janeiro recorded 92 fewer deaths than the previous year, and the number has fallen every year since.

Download the 17K-Graffiti dataset and its pre-trained weights for detecting Graffiti. The dataset provides larger graffiti instances containing a variety of graffiti types and annotated boundary boxes.

[NOTE] To access the dataset, which is only available for academic use, please send us your complete name, a brief description of your project, your advisor's name, and an academic email with a link to your university page.

For additional material regarding Code and data processing, please see the following GitHub repository at

https://github.com/visual-ds/17K-Graffiti

Please cite the published paper, if you find this dataset helpful on your research work:

@conference{visapp22,
author={Bahram Lavi and Eric K. Tokuda and Felipe Moreno-Vera and Luis Gustavo Nonato and Claudio T. Silva and Jorge Poco},
title={17K-Graffiti: Spatial and Crime Data Assessments in São Paulo City},
booktitle={Proceedings of the 17th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2022) - Volume 4: VISAPP},
year={2022},
pages={968-975},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0010883300003124},
isbn={978-989-758-555-5},
}

Abstract Macrosociological theories of criminal violence predict that the rate of violent crimes, especially intentional homicide, increases in response to social structures and processes that strengthen violent motivations or weaken social controls on violence. To test these hypotheses, we used several bivariate and multivariate regression models with panel data and variables constructed with demographic and mortality data, according to theoretical relevance, to verify whether the use of psychoactive substances, access to firearms, sociodemographic structures (population growth and density and proportion of young men), and the prevalence of socioeconomic exclusion increased the rate of intentional homicides in Brazilian microregions between 1996 and 2019. Most of the results significantly support the hypotheses. But the most powerful factor was the previous year’s homicide rate. This reveals an endogenous feedback tendency of violence in the short and medium terms, which can lead to the accumulation of the effects of the structural factors of intentional homicides.

The number of black and Pardo Brazilian people murdered in Brazil had been increasing throughout the years up until 2017. In 2022, around 35,500 black and Pardo citizens were killed in the South American country, down from almost 37,000 recorded one year before.