Detalhes bibliográficos
| Ano de defesa: |
2015 |
| Autor(a) principal: |
Silva, Suelyn Fernanda da
 |
| Orientador(a): |
Canteri, Marcelo Giovanetti
|
| Banca de defesa: |
Fonseca, Inês Cristina de Batista
,
Matos, Simone Nasser
|
| Tipo de documento: |
Dissertação
|
| Tipo de acesso: |
Acesso aberto |
| Idioma: |
por |
| Instituição de defesa: |
UNIVERSIDADE ESTADUAL DE PONTA GROSSA
|
| Programa de Pós-Graduação: |
Programa de Pós Graduação Computação Aplicada
|
| Departamento: |
Computação para Tecnologias em Agricultura
|
| País: |
BR
|
| Palavras-chave em Português: |
|
| Palavras-chave em Inglês: |
|
| Área do conhecimento CNPq: |
|
| Link de acesso: |
http://tede2.uepg.br/jspui/handle/prefix/128
|
Resumo: |
Meta-analysis is the analysis results of various studies, and is performed in order to synthesize these results statistically formal manner. One application is in the meta validation diagrammatic scales to assess severity of disease in plants. The aim of this study was a metaanalysis in diagrammatic scales of data validation, used to help determine the severity of disease in different cultures, using software that perform meta-analysis with data that correlate variables and have free or trial version, was developed a computer model to automate this task in a simplified manner compared to the software used, R and Comprehensive Meta-Analisys. The database for the meta-analysis comprised a total of 64 studies with the validation subject diagrammatic scales, these studies were split and led to 3 separate databases for reviews where scales were used, was not used and the evaluators They were trained to scale. The meta-analysis with the models of fixed effect and random effect on the R software was carried out in order to obtain the lowest value of the Akaike Information Criterion, which was used to determine the most appropriate model, and then performed a meta-analysis with the Comprehensive Meta-Analisys software. The computational model developed uses the structure of a wizard software, which performs the steps of meta-analysis found the form of literature this model performs the meta-analysis tasks on specific steps guiding the user through every step facilitating the realization of the meta-analysis, so the system It provides options to the user and according to the choice made the system identifies the next step to run until the end of the meta-analysis. It was concluded that the computational model helped to simplify the process of meta and also by the results obtained it is concluded that there are benefits in using scales diagrammatic having a higher overall correlation has been established to 0.97 where it was used. |
