Visualização de software baseada em uma metáfora do universo utilizando o conjunto de métricas CK

• Ano de defesa: 2014
• Autor(a) principal: Juliano, Renato Correa
• 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
• Palavras-chave em Português: Engenharia de software; Métricas de software; Anomalias; Visualização de software; Medição de software; Software engineering; Software metrics; Anomalies; Software visualization; CNPQ::CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO
• Link de acesso: https://repositorio.ufu.br/handle/123456789/12566; https://doi.org/10.14393/ufu.di.2014.224

The software development process is a complex, costly and difficult task, being subject of several studies over the years. In order to turn software more tangible, metrics calculation is used to provide useful informations to support developers in decision-making process. This present study used Chidamber and Kemerer suite metrics (CK metrics) to measure software developed under object oriented paradigm. A sistematic review was conducted and exposed the efficiency of CK metrics in several studies. Some metrics as CBO, RFC and WMC were successfully used in all studies analyzed, while other (LCOM , DIT and NOC) were successful in only a few. In addition, a table with values of these measures indicated possible software problems, such as predicting error prone, was created. A new visualization model is proposed, based on a simplistic metaphor of universe, which aims to facilitate the understanding of softwares transforming the classes of a system in celestial bodies have stipulated characteristics according to the value of metrics. This model was implemented in a software, called SUVsoft, that performs visualization software and calculate the CK metrics suite. Finally, five softwares of different sizes and contexts were visualized and analyzed, and it was observed that the application of gravitational force herewith color and radius allowed the identification of classes with discrepant values in CK metrics. The results of this dissertation can be used to guide future studies of CK metrics and also to assist Software Engineering activities through the visualization using the proposed model.