Um framework de testes unitários para procedimentos de carga em ambientes de business intelligence

• Ano de defesa: 2016
• Autor(a) principal: Santos, Igor Peterson Oliveira
• Orientador(a): Rodrigues Júnior, Methanias Colaço
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Sergipe
• Programa de Pós-Graduação: Pós-Graduação em Ciência da Computação
• País: Brasil
• Palavras-chave em Português: Computação; Programas de computador; Framework; Engenharia de software; Software > Testes; Inteligência competitiva (Administração); Testes de software; Engenharia de software experimental
• Palavras-chave em Inglês: Business Intelligence (BI); Data Warehouse; Data Analytics; Software testing; Experimental software engineering
• Área do conhecimento CNPq: CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO
• Link de acesso: https://ri.ufs.br/handle/riufs/3390

Business Intelligence (BI) relies on Data Warehouse (DW), a historical data repository designed to support the decision making process. Despite the potential benefits of a DW, data quality issues prevent users from realizing the benefits of a BI environment and Data Analytics. Problems related to data quality can arise in any stage of the ETL (Extract, Transform and Load) process, especially in the loading phase. This thesis presents an approach to automate the selection and execution of previously identified test cases for loading procedures in BI environments and Data Analytics based on DW. To verify and validate the approach, a unit test framework was developed. The overall goal is achieve data quality improvement. The specific aim is reduce test effort and, consequently, promote test activities in DW process. The experimental evaluation was performed by two controlled experiments in the industry. The first one was carried out to investigate the adequacy of the proposed method for DW procedures development. The Second one was carried out to investigate the adequacy of the proposed method against a generic framework for DW procedures development. Both results showed that our approach clearly reduces test effort and coding errors during the testing phase in decision support environments.