Building effective defect-prediction models in practice

A. G. Koru, Hongfang D Liu

Research output: Contribution to journalArticle

118 Citations (Scopus)

Abstract

Successfully predicting defect-prone software modules can help developers improve product quality by focusing quality assurance activities on those modules. Emerging repositories of publicly available software engineering data sets support research in this area by providing static measures and defect data that developers can use to build prediction models and test their effectiveness. Stratifying NASA data sets from the PROMISE repository according to module size showed improved prediction performance in the subsets that included larger modules.

Original languageEnglish (US)
Pages (from-to)23-29
Number of pages7
JournalIEEE Software
Volume22
Issue number6
DOIs
StatePublished - Nov 2005
Externally publishedYes

Fingerprint

Defects
Set theory
Quality assurance
NASA
Software engineering

ASJC Scopus subject areas

  • Computer Graphics and Computer-Aided Design
  • Software

Cite this

Building effective defect-prediction models in practice. / Koru, A. G.; Liu, Hongfang D.

In: IEEE Software, Vol. 22, No. 6, 11.2005, p. 23-29.

Research output: Contribution to journalArticle

@article{97e9fad4f2c34f89aeae638ddce7c658,
title = "Building effective defect-prediction models in practice",
abstract = "Successfully predicting defect-prone software modules can help developers improve product quality by focusing quality assurance activities on those modules. Emerging repositories of publicly available software engineering data sets support research in this area by providing static measures and defect data that developers can use to build prediction models and test their effectiveness. Stratifying NASA data sets from the PROMISE repository according to module size showed improved prediction performance in the subsets that included larger modules.",
author = "Koru, {A. G.} and Liu, {Hongfang D}",
year = "2005",
month = "11",
doi = "10.1109/MS.2005.149",
language = "English (US)",
volume = "22",
pages = "23--29",
journal = "IEEE Software",
issn = "0740-7459",
publisher = "IEEE Computer Society",
number = "6",

}

TY - JOUR

T1 - Building effective defect-prediction models in practice

AU - Koru, A. G.

AU - Liu, Hongfang D

PY - 2005/11

Y1 - 2005/11

N2 - Successfully predicting defect-prone software modules can help developers improve product quality by focusing quality assurance activities on those modules. Emerging repositories of publicly available software engineering data sets support research in this area by providing static measures and defect data that developers can use to build prediction models and test their effectiveness. Stratifying NASA data sets from the PROMISE repository according to module size showed improved prediction performance in the subsets that included larger modules.

AB - Successfully predicting defect-prone software modules can help developers improve product quality by focusing quality assurance activities on those modules. Emerging repositories of publicly available software engineering data sets support research in this area by providing static measures and defect data that developers can use to build prediction models and test their effectiveness. Stratifying NASA data sets from the PROMISE repository according to module size showed improved prediction performance in the subsets that included larger modules.

UR - http://www.scopus.com/inward/record.url?scp=28244461468&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=28244461468&partnerID=8YFLogxK

U2 - 10.1109/MS.2005.149

DO - 10.1109/MS.2005.149

M3 - Article

AN - SCOPUS:28244461468

VL - 22

SP - 23

EP - 29

JO - IEEE Software

JF - IEEE Software

SN - 0740-7459

IS - 6

ER -