Regras de associação para análise de casos de dengue em municípios paraibanos

• Ano de defesa: 2017
• Autor(a) principal: Medeiros Filho, Fernando
• Orientador(a): Araujo, Wellington Candeia de
• Banca de defesa: Brito, Nadja Maria da Silva Oliveira , Nóbrega, Clarissa de Lucena
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Estadual da Paraíba
• Programa de Pós-Graduação: Programa de Pós-Graduação Profissional em Ciência e Tecnologia em Saúde - PPGCTS
• Departamento: Pró-Reitoria de Pós-Graduação e Pesquisa - PRPGP
• País: BR
• Palavras-chave em Português: Mineração de dados; Dengue; Epidemiologia
• Palavras-chave em Inglês: Dengue; Epidemiology; Data Mining
• Área do conhecimento CNPq: CIENCIAS DA SAUDE
• Link de acesso: https://repositorio.uepb.edu.br/handle/123456789/72780

The objective of this work is to apply the techniques of Data Mining in the information contained in epidemiological bulletins of the portal of the Ministry of Health that concern the information of Dengue of 18 municipalities of the State of Paraíba. Through this volume of data association rules were applied through analyzes with the APRIORI algorithm. The techniques were used through the free software WEKA and the algorithm implemented by it to generate association rules. Throughout the process the visualization, recognition and organization of the patterns discovered through maps were made, finding relationships among the thirteen analyzes that were carried out at the time of work. They involved the 52 epidemiological weeks and from that found rules that are applicable to programs to combat Dengue. The identified patterns were part of the great proposal of this dissertation - To direct the actions to combat Dengue in the analyzed municipalities. Leading to conclude that actions involving Sanitary and Environmental Surveillance of the municipalities in question can perform better if the actions are internalized, segmented and adapted to the reality found.