DocumentCode :
2403357
Title :
A Study on Defect Density of Open Source Software
Author :
Rahmani, Cobra ; Khazanchi, Deepak
Author_Institution :
Coll. of Inf. Sci. & Technol., Univ. of Nebraska-Omaha, Omaha, NE, USA
fYear :
2010
fDate :
18-20 Aug. 2010
Firstpage :
679
Lastpage :
683
Abstract :
Open source software (OSS) development is considered an effective approach to ensuring acceptable levels of software quality. One facet of quality improvement involves the detection of potential relationship between defect density and other open source software metrics. This paper presents an empirical study of the relationship between defect density and download number, software size and developer number as three popular repository metrics. This relationship is explored by examining forty-four randomly selected open source software projects retrieved from SourceForge.net. By applying simple and multiple linear regression analysis, the results reveal a statistically significant relationship between defect density and number of developers and software size jointly. However, despite theoretical expectations, no significant relationship was found between defect density and number of downloads in OSS projects.
Keywords :
public domain software; regression analysis; software metrics; software quality; SourceForge.net; defect density; multiple linear regression analysis; open source software development; open source software metrics; open source software projects; software quality; Open source software; Probability; Regression analysis; Software engineering; Software measurement; Bug repository; Defect density; Linear regression; Open source software; Software repository metric;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Computer and Information Science (ICIS), 2010 IEEE/ACIS 9th International Conference on
Conference_Location :
Yamagata
Print_ISBN :
978-1-4244-8198-9
Type :
conf
DOI :
10.1109/ICIS.2010.11
Filename :
5591023
Link To Document :
بازگشت