The aim of this project is to ascertain whether the data in existing Police recording systems can be used by existing mature data mining techniques in an efficient manner to achieve results that are more accurate than those achieved by Police specialists when analysing crime. The Police Service has no formalised methodology of recording and analysing crime data and it is incumbent on each Force to train and develop appropriate personnel to provide operational analysis. Police data is inconsistent and, frequently, incomplete making the task of formal analysis far more difficult and current analytical practices are semi-manual and time consuming producing results of limited accuracy. These analytical processes would benefit from using data mining techniques within a structured approach as discussed within this thesis. The usage of supervised and unsupervised learning techniques within a structured methodology to mining Police data is evaluated. The research demonstrates that data mining techniques can be successfully used in operational policing. High volume crimes such as burglary that have been committed by one or more known offenders can be classified and the model used to attribute currently undetected crimes to one or more of those known offenders. Burglary crimes that previously had no overt relationship and the identity of the offender is unknown can be clustered with the ability to suggest one or more offenders who may be responsible for committing the crime. The same techniques used in analysing high volume crime can be used to link low volume major crimes such as serious sexual assaults. The recognised benefits include an improvement in the accuracy of results over current semi-manual processes and a reduction in the time taken to achieve those results.

The crime rate is among the highest risk factors and endangers society. By analyzing the development and subsequent predicting future values, you can take steps that would help to reduce crime and create the conditions conducive to the good development of society and security of citizens. Therefore, the State should control crime by one of the most important priorities. This thesis deals with the analysis of crime and number of people prosecuted in the Czech Republic, constructed using time series. Development into the future is estimated in the next three years. Longer forecast should not make much sense because of the rapidly changing laws. The conclusion is devoted to crime in Czech regions. Analysis is performed using cluster analysis, which examines the number of crimes in different regions. The result is a decision that the region together in terms of crime the most similar.

This study presents a brief analysis of criminal occurrences plus incidents in the city of Santa Maria RS, during the year of 2010, attended by the Military Brigade (Brigada Militar). This is a survey of the empirical-analytic features with a quantitative approach and of descriptive base. Violence is a complex, multidisciplinary phenomena, which has no single cause, but rather a multitude of factors. The literature suggests that the analysis of social, economic and demographic patterns are crucial for understanding the criminal dynamics. For the analysis it was used the Georeferencing of Activity of the Military Brigade (GeoAtBM), a software of geographical information associated with a spatial database, developed through a cooperation agreement between the Federal University of Santa Maria (UFSM) and the Military Brigade (BM) of the State of Rio Grande do Sul. The system enables a vast array of analysis, using maps, graphs, tables, and reports, allowing for correlations about crime. The results obtained through this innovative tool, GeoAtBM contribute to the managers of Public Safety conduct operational planning and decision-making in resource management, with emphasis on prevention of crimes. It emphasizes the importance of an integrated database for the scenario analysis and indicators of public safety in order to avoid inconsistency in the statistical results due to the existence of several databases, lack of standardization and constant corrections in the information recorded. / O presente estudo apresenta uma análise sucinta das ocorrências criminais mais incidentes na cidade de Santa Maria, Rio Grande do Sul, durante o ano de 2010, atendidas pela Brigada Militar. Trata-se de uma pesquisa com características empírico-analíticas, com abordagem quantitativa e de base descritiva. A violência é um fenômeno complexo e multidisciplinar, que não é gerada por uma causa única, mas por uma multiplicidade de fatores. A literatura aponta que a análise de padrões sociais, econômicos e demográficos são primordiais para o entendimento da dinâmica criminal. Para a análise, utilizou-se o Georreferenciamento da Atividade da Brigada Militar (GeoAtBM), um software de informações geográficas associado a um banco de dados espacial, desenvolvido através de um Termo de Cooperação entre a Universidade Federal de Santa Maria (UFSM) e a Brigada Militar (BM) do Estado do Rio Grande do Sul. O sistema possibilita uma vasta gama de análise, através de mapeamentos, gráficos, tabelas e relatórios, permitindo estabelecer correlações entre os elementos da criminalidade. Os resultados obtidos através desta ferramenta inovadora contribuem para que os gestores de Segurança Pública realizem o planejamento operacional e a tomada de decisão na administração de recursos, com ênfase na prevenção dos delitos. Pontua-se a importância de um banco de dados integrados para a análise de cenários e indicadores de segurança pública, a fim de evitar inconsistência nos resultados estatísticos devido à existência de diversos bancos de dados, à falta de padronização e às retificações constantes das informações registradas.

Submitted by Ricardo Cedraz Duque Moliterno (ricardo.moliterno@uefs.br) on 2015-11-17T00:44:34Z No. of bitstreams: 1 Disserta??o_Daniel Deivson Alves Portella.pdf: 880204 bytes, checksum: d2be49d558bf8b996a9ab715191cf142 (MD5) / Made available in DSpace on 2015-11-17T00:44:34Z (GMT). No. of bitstreams: 1 Disserta??o_Daniel Deivson Alves Portella.pdf: 880204 bytes, checksum: d2be49d558bf8b996a9ab715191cf142 (MD5) Previous issue date: 2013-03-08 / Coordena??o de Aperfei?oamento de Pessoal de N?vel Superior - CAPES / Funda??o de Amparo ? Pesquisa do Estado da Bahia - FAPEB / Introduction: Drug trafficking can be considered as one of the factors for the increase in homicides in large cities and metropolitan areas. The high frequency of crimes contributed to the population to make anonymous complaints about criminal acts through the Hotline program implemented in Bahia since 2005. Objectives: Analyze the Disk Denounce as a source of information for crime analysis in the Metropolitan Region of Salvador (MRS) from 2006 to 2011 and to assess the association between murder, drug trafficking and social indicators in Salvador in 2010. Methods: This was an ecological and descriptive study of multiple groups of explanatory type, taking the unit of analysis from the municipales of the MRS. It was used records of criminal complaints Disk Denounce, murders, people involved in drug trafficking registered by the Civil Police of Bahia and social indicators of the Brazilian Institute of Geography and Statistics (BIGS). Data were analyzed by calculating the absolute and relative frequency for complaints by class and type of subject. The Poisson regression model was used to investigate the association between the rate of murders (dependent variable), number of people involved in drug trafficking, the proportion of residents race / black (main independent variable) and social indicators (co variables). In the analysis we used the R statistical package software in the public domain. Results: Were registered in the period from 2006 to 2011, 65,575 complaints in RMS, distributed by the 13 municipalities of the RMS, drug trafficking was the most reported, representing a majority of all complaints in the class of narcotics complaints. In Salvador was registered 1,494 homicides in 2010, the murder rate of 70.4 per 100 thousand inhabitants. The multivariate analysis showed that there is a 10% increase in the proportion of people of race / black color in a neighborhood the average rate of murder increased 32.4%. This is an average increase of 17.4%, in neighborhoods that differ in 10 people involved in drug trafficking, both results showed a positive association with statistical significance. Final Considerations: The popular participation in reporting crimes and other offenses constitutes of an important aspect because primarily directs the actions of the police and cooperates with the State in fighting violence and secondly prevents the trivialization of violence. / Introdu??o: O tr?fico de drogas pode ser considerado como um dos fatores determinantes para o aumento de homic?dios nas metr?poles e nas regi?es metropolitanas. A grande frequ?ncia de crimes contribuiu para que a popula??o passasse a fazer den?ncias an?nimas sobre os fatos criminosos atrav?s do programa Disque Den?ncia,implantado na Bahia desde 2005. Objetivos: Analisar o Disque Den?ncia como fonte de informa??es para an?lises criminais na Regi?o Metropolitana de Salvador (RMS) de 2006-2011 e verificar associa??o entre homic?dio doloso, tr?fico de drogas e indicadores sociais em Salvador em 2010. M?todos: Trata-se de estudo descritivo e ecol?gico de m?ltiplos grupos do tipo explorat?rio tomando-se como unidade de an?lise os munic?pios da RMS. Foram utilizados registros de den?ncias criminais do Disque Den?ncia, homic?dios dolosos, pessoas envolvidas com o tr?fico de drogas registrados pela Pol?cia Civil da Bahia e indicadores sociais do Instituto Brasileiro de Geografia e Estat?stica (IBGE). Os dados foram analisados atrav?s do c?lculo de frequ?ncia absoluta e relativa para as den?ncias por classe e tipo de assunto. O modelo de regress?o de Poisson foi utilizado para verifica??o da associa??o entre a taxa de homic?dio doloso (vari?vel dependente), n?mero de pessoas envolvidas com o tr?fico de drogas, propor??o de residentes da ra?a/cor negra (vari?veis independentes principais) e indicadores sociais (covari?veis). Na an?lise foi utilizado o software R pacote estat?stco de dom?nio p?blico. Resultados: Foram registradas, no per?odo de 2006 a 2011, 65.575 den?ncias na RMS, distribu?das pelos 13 munic?pios da RMS, o tipo de assunto tr?fico de drogas foi o mais denunciado, representando a maioria de todos os assuntos denunciados na classe de den?ncias subst?ncias entorpecentes. Em Salvador foi registrado 1.494 homic?dios dolosos em 2010, tendo taxa de 70,4 homic?dio doloso por 100mil habitantes. A an?lise multivariada mostrou que a medida que h? aumento de 10% na propor??o de habitantes da ra?a/cor negra em um bairro a taxa m?dia de homic?dio doloso aumenta 32,4%. Esse aumento m?dio ? de 17,4%, nos bairros que diferem em 10 pessoas envolvidas com o tr?fico de drogas, ambos os resultados apresentaram associa??o positiva com signific?ncia estat?stica. Considera??es finais: A participa??o popular na den?ncia de crimes e outros delitos se constitui em aspecto importante porque de um lado direciona as a??es das autoridades policiais e colabora com o Estado no enfrentamento da viol?ncia e por outro lado evita a banaliza??o de atos violentos.