Please use this identifier to cite or link to this item: https://repositorio.ufba.br/handle/ri/17176
metadata.dc.type: Artigo de Periódico
Title: The bug report duplication problem: an exploratory study
Other Titles: Software Quality Journal
Authors: Cavalcanti, Yguaratã Cerqueira
Silveira Neto, Paulo Anselmo da Mota
Lucrédio, Daniel
Vale, Tassio
Almeida, Eduardo Santana de
Meira, Silvio Romero de Lemos
metadata.dc.creator: Cavalcanti, Yguaratã Cerqueira
Silveira Neto, Paulo Anselmo da Mota
Lucrédio, Daniel
Vale, Tassio
Almeida, Eduardo Santana de
Meira, Silvio Romero de Lemos
Abstract: Duplicate bug report entries in bug trackers have a negative impact on software maintenance and evolution. This is due, among other factors, to the increased time spent on report analysis and validation, which in some cases takes over 20 min. Therefore, a considerable amount of time is lost in duplicate bug report analysis. In order to understand the possible factors that cause bug report duplication and its impact on software development, this paper presents an exploratory study in which bug tracking data from private and open source projects were analyzed. The results show, for example, that all projects we investigated had duplicate bug reports and a considerable amount of time was wasted by this duplication. Furthermore, features such as project lifetime, staff size, and the number of bug reports do not seem to be significant factors for duplication, while others, such as the submitters’ profile and the number of submitters, do seem to influence the bug report duplication.
Keywords: Bug reports
Bug tracker
Bug reports duplication
Exploratory study
Software configuration management
metadata.dc.rights: Acesso Aberto
URI: http://repositorio.ufba.br/ri/handle/ri/17176
Issue Date: 2013
Appears in Collections:Artigo Publicado em Periódico (PGCOMP)

Files in This Item:
File Description SizeFormat 
Yguaratã Cerqueira Cavalcanti.pdf864,22 kBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.