Meta-modelo funcional para recuperação de informação

Detalhes bibliográficos
Ano de defesa: 2006
Autor(a) principal: Oliveira, Luciene Chagas de
Orientador(a): Não Informado pela instituição
Banca de defesa: Não Informado pela instituição
Tipo de documento: Dissertação
Tipo de acesso: Acesso aberto
Idioma: por
Instituição de defesa: Universidade Federal de Uberlândia
BR
Programa de Pós-graduação em Ciência da Computação
Ciências Exatas e da Terra
UFU
Programa de Pós-Graduação: Não Informado pela instituição
Departamento: Não Informado pela instituição
País: Não Informado pela instituição
Palavras-chave em Português:
Link de acesso: https://repositorio.ufu.br/handle/123456789/12489
Resumo: Modelling is one of the central tasks in the development of information retrieval systems. A useful tool for developing a new information retrieval model is a generic framework. This frameworks can be seen as formal meta-models that make possible to describe and to investigate formally the semantics of the retrieval process and becomes possible to reason about features and properties of information retrieval models (IR). With the growth and the differences between the IR strategies and models formal modelling comes becoming more and more important. In this dissertation, we propose a generic and formal framework for defining IR models named Functional Framework. This framework is a meta-model for IR models, defining a level of abstraction that allows the representation, formulation and comparison of IR models. With this meta-model, IR models can be represented in a unique common language, which makes the study of characteristics and properties of the models and the combination of these models easier. The framework also provides a formalism that permits the comparison of models without the need to carry out experiments. Moreover, we show examples of how to represent the three classic IR models and we design a model based on distance equivalent to the classic vector model using the framework functional. We also analyze the combination of multiple evidence, presenting two case studies of the use of the framework to combine multiple evidence in contexts bayesian belief networks and in the vector space model. We show that the combination of multiple evidence in the bayesian belief network can be carried at in of several ways, being that each form corresponds to a similarity function in the vector model. The analysis of this correspondence is made through the functional framework. We show that the framework allows us to design new models and helps designers to modify these models to extend them with new evidence sources. As application of the functional meta-model, we also present the ideas of development of a meta-tool for experimental comparison between IR models.