Uma abordagem baseada em ontologias para integração semântica de sistemas

• Ano de defesa: 2011
• Autor(a) principal: Calhau, Rodrigo Fernandes
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal do Espírito Santo; BR; Mestrado em Informática; Centro Tecnológico; UFES; Programa de Pós-Graduação em Informática
• Palavras-chave em Português: Semantic system integration; Ontologies; Semantic interoperability; Software configuration management; Interoperabilidade semântica; Modelagem conceitual; Integração semântica de sistemas; Ontologias; Gerência de configuração de software; Integração semântica (sistemas de computador); Ontologias (Recuperação da informação); Gerenciamento de configurações de software; Ciência da Computação; 004
• Link de acesso: http://repositorio.ufes.br/handle/10/4239

In order to properly support the whole organization’s business process, systems should be integrated. However, integration is a complex problem, especially for large and dynamic organizations. In general, each system runs independently and implements its own data and process models. These models are not shared between systems, leading to several conflicts, including technical, syntactical and, especially, semantic conflicts. This heterogeneity is considered one of the major difficulties of the integration problem. In this context, ontologies can be used as an interlingua to map concepts and services used by the different enterprise applications, in a scenario of access to data and services via a shared ontology. In this work we present an Ontology-Based Approach for Semantic Integration (OBA-SI) that concentrates efforts in the integration requirements analysis and modeling. In this approach, integration is addressed in a high level of abstraction, looking for getting semantic agreement between the systems at the conceptual level. A premise of the proposed approach is that the assignment of semantics and the use of ontologies must be independent of the integration solution itself. The proposed approach deals with integration in three layers: data, service and process. To accomplish this, conceptual and behavioral models of the systems to be integrated are compared in the light of a reference ontology, which is responsible for assigning semantics to the shared items of the systems. Thus, to illustrate our approach, we present a real case of its application in integrating the version control system Subversion and a system for controlling changes, in order to support a Software Configuration Management process.