Causas de desaparecimento no Estado de São Paulo entre 2013 e 2014: uma análise automatizada de boletins de ocorrência
Carregando...
Orientador
Co-orientadores
Citações na Scopus
Tipo de documento
Artigo Científico
Data
2021
Arquivos
Título da Revista
ISSN da Revista
Título do Volume
Resumo
Este artigo estuda as causas de desaparecimento no estado de São Paulo por idade e sexo dos desaparecidos. Para tanto, usou o PLID, um banco de dados com os boletins de ocorrência dos desaparecimentos de 2013 e 2014. A metodologia consiste no uso de aprendizado de máquina para rotular automaticamente a causa de desaparecimento nos boletins, assim como métodos de quantificação para mensurar as proporções relativas de causa de desaparecimento estratificadas por idade e sexo. Os resultados indicam que grande parte dos boletins de ocorrência não possuem informações suficientes para inferir a causa do desaparecimento. Dentre os boletins com causa de desaparecimento clara, foi possível obter algumas classes de motivos mais frequentes para desaparecimento. A causa de desaparecimento mais frequente é a voluntária, sendo mais comum entre mulheres e homens menores de 18 anos. Também, cerca de 20% dos desaparecimentos ocorrem por usuários de drogas ou álcool, sendo mais comum entre menores de 50 anos. A partir de 50 anos, essa categoria torna-se menos frequente, sendo substituída pelo desaparecimento não intencional. Também se observou que menos de 5% dos indivíduos foram vítimas de crimes. Essas conclusões podem auxiliar na elaboração de políticas públicas mais assertivas.
This article studies the causes of disappearance in the state of São Paulo by age and sex of the disappeared. For this purpose, it used the PLID, a database with the disappearance reports of 2013 and 2014. The methodology consists in applying machine learning to automatically classify the cause of disappearance in the reports, as well as quantification methods to measure the relative proportions causes of disappearance stratified by age and sex. The results indicate that many of the reports do not have enough information to infer the cause of disappearance. Among the reports with clear cause of disappearance, it was possible to obtain some classes of more frequent reasons for disappearance. The most frequent cause is voluntary, being more common among women and men under 18 years. Also, about 20% of disappearances occur by drug or alcohol users, being more common among individuals under 50 years of age. From age 50 and up, this category becomes less frequent, being replaced by unintentional disappearance. It was also observed that less than 5% of individuals were victims of crimes. These conclusions can help with the development of more assertive public policies
This article studies the causes of disappearance in the state of São Paulo by age and sex of the disappeared. For this purpose, it used the PLID, a database with the disappearance reports of 2013 and 2014. The methodology consists in applying machine learning to automatically classify the cause of disappearance in the reports, as well as quantification methods to measure the relative proportions causes of disappearance stratified by age and sex. The results indicate that many of the reports do not have enough information to infer the cause of disappearance. Among the reports with clear cause of disappearance, it was possible to obtain some classes of more frequent reasons for disappearance. The most frequent cause is voluntary, being more common among women and men under 18 years. Also, about 20% of disappearances occur by drug or alcohol users, being more common among individuals under 50 years of age. From age 50 and up, this category becomes less frequent, being replaced by unintentional disappearance. It was also observed that less than 5% of individuals were victims of crimes. These conclusions can help with the development of more assertive public policies
Palavras-chave
Titulo de periódico
Revista da Defensoria Pública do Estado de São Paulo
Título de Livro
URL na Scopus
Idioma
pt
URL permanente
Notas
Membros da banca
Área do Conhecimento CNPQ
CIENCIAS SOCIAIS APLICADAS