Localização evolucionária de defeitos em software baseada na singularidade de escores de suspeita

• Ano de defesa: 2022
• Autor(a) principal: Ferreira, Willian de Jesus
• Orientador(a): Leitão Júnior, Plinio de Sá
• Banca de defesa: Leitão Júnior, Plinio de Sá, Bulcão Neto, Renato De Freitas, Chaim, Marcos Lordello
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Goiás
• Programa de Pós-Graduação: Programa de Pós-graduação em Ciência da Computação (INF)
• Departamento: Instituto de Informática - INF (RG)
• País: Brasil
• Palavras-chave em Português: Localização de defeitos; SBSE; Depuração de software; Programação genética
• Palavras-chave em Inglês: Fault Localization; Software debug; Genetic programing
• Área do conhecimento CNPq: CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO
• Link de acesso: http://repositorio.bc.ufg.br/tede/handle/tede/12436

Context. Software is subject to the presence of faults, which impacts its quality as well as production and maintenance costs. Evolutionary fault localization has used data from the test activity (test spectra) as a source of information about defects, and its automation aims to obtain better accuracy and lower software repair cost. Motivation. Our analysis identified that test spectra commonly used in the research field have a high ratio of sample repetition, which impairs the training and evolution of models (heuristics). Problem. We investigate whether the uniqueness of suspiciousness scores can boost the ability to find software faults, aiming to deal with samples repetition, that is, if an exploration based on how distinguishable program elements are about being defective can generate competitive models. Methodology. The investigation formalized hypotheses, introduced three training strategies to guide the proposal and carried out an experimental evaluation, aiming to reach conclusions regarding the assessment of research questions and hypotheses. Analysis. The results have shown the competitiveness of all the proposed training strategies through evaluation metrics commonly used in the research field. Conclusion. Statistical analyses confirmed that the uniqueness of suspiciousness scores guides the generation of superior heuristics for fault localization.