Crowdsourcing em revisões sistemáticas

• Ano de defesa: 2021
• Autor(a) principal: Lopes, Rafael Aparecido
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Tecnológica Federal do Paraná; Cornelio Procopio; Brasil; Programa de Pós-Graduação em Informática; UTFPR
• Palavras-chave em Português: Engenharia de software; Revisão sistemática da literatura; Mapeamento sistemático; Software engineering; Systematic review of the literature; Systematic mapping; CNPQ::CIENCIAS EXATAS E DA TERRA; Ciência da Computação
• Link de acesso: http://repositorio.utfpr.edu.br/jspui/handle/1/27055

Context: Systematic Reviews (SRs) have been adopted in Software Engineering (ES) for more than a decade to provide synthesis of evidence on various topics.However, the process for conducting an SR remains laborious and time-consuming.It is important to explore approaches to conduct SRs more quickly and with less effort.One of these approaches is crowdsourcing, as sharing SR activities among a large number of researchers is a promising alternative. Objective: The goal is twofold: first, to evaluate the use of crowdsourcing, mapping its advantages and challenges to support activities in the SR process in ES; and to define a process model to lead the use of crowdsourcing in SRs. Method: Three methodological steps were carried out: (1) firstly, a Systematic Mapping (MS) was conducted in order to collaborate with a vision of the use of crowdsourcing to be adopted in the SR process in ES; (2) Subsequently, through a survey, the opinion of ES researchers on the adoption of crowdsourcing in RS was collected. (3) Finally, a case study was carried out for the application of the crowdsourcing-based SR process for the studies selection activity. Results: This subject is still little explored; One of the main benefits of using crowdsourcing in RS in ES is the acceleration of the process; Quality control is one of the biggest challenges of this approach, as the work of the crowd was not reliable in the case study; There is no complete crowd-based SR process, nor is there a crowdsourcing platform that specifically supports SR activities.