Uma avaliação sobre diferentes padrões de modelagens para índices e indicadores de poluição do ar em ontologias

• Ano de defesa: 2018
• Autor(a) principal: Taques, Ricardo Martins
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Tecnológica Federal do Paraná; Curitiba; Brasil; Programa de Pós-Graduação em Engenharia Elétrica e Informática Industrial; UTFPR
• Palavras-chave em Português: Ontologias (Recuperação da informação) - Avaliação; Representação do conhecimento (Teoria da informação); Sistemas de reconhecimento de padrões; Modelos matemáticos; Web semântica; Ar - Poluição - Índices; Indicadores ambientais; Métodos de simulação; Engenharia elétrica; Ontologies (Information retrieval) - Evaluation; Knowledge representation (Information theory); Pattern recognition systems; Mathematical models; Semantic web; Air - Pollution - Indexes; Environmental indicators; Simulation method; Electric engineering; CNPQ::CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO; Engenharia Elétrica
• Link de acesso: http://repositorio.utfpr.edu.br/jspui/handle/1/4188

Several projects have developed ontologies to represent indicators of environmental pollution. However, as seen in the correlated workers, some of the aspects regarding the representation of the concepts and relations involved in the index-indicator relationship were not widely explored. Considering the problems arising from these forms of representation, the present work aims to propose alternative forms for modeling the relation between index and indicators, using content patterns presented in a customized format, based on the features involved in scenarios extracted from the air pollution indexes’ specifications, provided by some of the most important agencies for environment control and monitoring. In addition, to evaluate the two models proposed in this dissertation, a set of criteria usually utilized to analyze the quality of ontologies are being used, i.e.: completeness, conciseness, expandability and the quality of the outputs. Along with the main objective, this work invites the subject matter experts to discuss about patterns related to indicators and indexes that aim to meet a particular perspective of the end-user needs. Finally, these proposed patterns are validated through case studies that guide an application to describe index and indicators in two ontologies. From these evaluations, a discussion is proposed by analyzing the results achieved prior to outline the conclusions about the work.