ObasCId : uma abordagem ontologicamente fundamentada para EROA

• Autor(a) principal: Parreira Júnior, Paulo Afonso
• Data de Publicação: 2015
• Tipo de documento: Tese
• Idioma: por
• Título da fonte: Repositório Institucional da UFSCAR
• Texto Completo: https://repositorio.ufscar.br/handle/ufscar/7625
• topic: Engenharia de software; Identificação de interesses; Classificação de interesses; Interesses transversais; Engenharia de requisitos orientada a aspectos; Ontologias de domínio; CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO

A software concern may be defined as a set of requirements related to the same purpose. A CrossCutting Concern (CCC) (or Early-Aspect) is a concern whose requirements cutacross other concern requirements and the inadequate modularization of this type of concern may hinder the software development and evolution. The Aspect-Oriented Requirements Engineering (AORE) area provides more appropriate strategies for CCC identification, representation and composition. A commonly reported issue on AORE approaches is the low effectiveness provided by them, regarding to the concern identification and classification. This is due to two main causes: (i) the lack of a broader understanding about software concerns domain; and (ii) the lack of appropriate resources to support software engineers during the identification and classification of software concerns. This work aims to mitigate the reported issue by dealing with its main causes. To do this, we present: (i) a reference ontology for software concerns domain, called O4C (Ontology for Concerns), that aims to make clear and precise the description of the elements of this domain; (ii) an ontologically-based AORE approach, called ObasCId (Ontologically-based Concern Identification); and (iii) a computational tool, called OBasCId-Tool, that automates some activities of the ObasCId approach. The evaluation of the proposed approach and its computational support was carried out through experimental studies, in order to verify the effectiveness and efficiency of these research products regarding to concern identification and classification. The results indicate the ObasCId approach may contribute positively to the recall of software concerns without harming the precision and the execution time of the approach. Regarding to the ObasCId- Tool, it has been considered approved by its users, regarding to its usefulness and easeof- use features.