A study of redundant metrics in defect prediction datasets

Jirayus Jiarpakdee; Chakkrit Tantithamthavorn; Akinori Ihara; Kenichi Matsumoto

doi:10.1109/ISSREW.2016.30

A study of redundant metrics in defect prediction datasets

Jirayus Jiarpakdee, Chakkrit Tantithamthavorn, Akinori Ihara, Kenichi Matsumoto

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Other

23 Citations (Scopus)

Abstract

Defect prediction models can help Software Quality Assurance (SQA) teams understand their past pitfalls that lead to defective modules. However, the conclusions that are derived from defect prediction models without mitigating redundant metrics issues may be misleading. In this paper, we set out to investigate if redundant metrics issues are affecting defect prediction studies, and its degree and causes of redundancy. Through a case study of 101 publicly-available defect datasets of systems that span both proprietary and open source domains, we observe that (1) 10%-67% of metrics of the studied defect datasets are redundant, and (2) the redundancy of metrics has to do with the aggregation functions of metrics. These findings suggest that researchers should be aware of redundant metrics prior to constructing a defect prediction model in order to maximize internal validity of their studies.

Original language	English
Title of host publication	Proceedings - 2016 IEEE 27th International Symposium on Software Reliability Engineering Workshops
Subtitle of host publication	ISSREW 2016
Editors	Jeremy Bradbury
Place of Publication	Piscataway NJ USA
Publisher	IEEE, Institute of Electrical and Electronics Engineers
Pages	51-52
Number of pages	2
ISBN (Electronic)	9781509036011
DOIs	https://doi.org/10.1109/ISSREW.2016.30
Publication status	Published - 2016
Externally published	Yes
Event	International Symposium on Software Reliability Engineering 2016 - Ottawa, Canada Duration: 23 Oct 2016 → 27 Oct 2016 Conference number: 27th http://2016.issre.net/

Conference

Conference	International Symposium on Software Reliability Engineering 2016
Abbreviated title	ISSRE 2016
Country/Territory	Canada
City	Ottawa
Period	23/10/16 → 27/10/16
Internet address	http://2016.issre.net/

Keywords

Defect prediction models
Redundant metrics
Software quality assurance

Access to Document

10.1109/ISSREW.2016.30

Cite this

Jiarpakdee, J., Tantithamthavorn, C., Ihara, A., & Matsumoto, K. (2016). A study of redundant metrics in defect prediction datasets. In J. Bradbury (Ed.), Proceedings - 2016 IEEE 27th International Symposium on Software Reliability Engineering Workshops: ISSREW 2016 (pp. 51-52). [7789378] IEEE, Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/ISSREW.2016.30

@inproceedings{f8e0649fce8549e99b5671ab721ab6fe,

title = "A study of redundant metrics in defect prediction datasets",

abstract = "Defect prediction models can help Software Quality Assurance (SQA) teams understand their past pitfalls that lead to defective modules. However, the conclusions that are derived from defect prediction models without mitigating redundant metrics issues may be misleading. In this paper, we set out to investigate if redundant metrics issues are affecting defect prediction studies, and its degree and causes of redundancy. Through a case study of 101 publicly-available defect datasets of systems that span both proprietary and open source domains, we observe that (1) 10%-67% of metrics of the studied defect datasets are redundant, and (2) the redundancy of metrics has to do with the aggregation functions of metrics. These findings suggest that researchers should be aware of redundant metrics prior to constructing a defect prediction model in order to maximize internal validity of their studies.",

keywords = "Defect prediction models, Redundant metrics, Software quality assurance",

author = "Jirayus Jiarpakdee and Chakkrit Tantithamthavorn and Akinori Ihara and Kenichi Matsumoto",

year = "2016",

doi = "10.1109/ISSREW.2016.30",

language = "English",

pages = "51--52",

editor = "Bradbury, {Jeremy }",

booktitle = "Proceedings - 2016 IEEE 27th International Symposium on Software Reliability Engineering Workshops",

publisher = "IEEE, Institute of Electrical and Electronics Engineers",

address = "United States of America",

note = "International Symposium on Software Reliability Engineering 2016, ISSRE 2016 ; Conference date: 23-10-2016 Through 27-10-2016",

url = "http://2016.issre.net/",

}

Jiarpakdee, J, Tantithamthavorn, C, Ihara, A & Matsumoto, K 2016, A study of redundant metrics in defect prediction datasets. in J Bradbury (ed.), Proceedings - 2016 IEEE 27th International Symposium on Software Reliability Engineering Workshops: ISSREW 2016., 7789378, IEEE, Institute of Electrical and Electronics Engineers, Piscataway NJ USA, pp. 51-52, International Symposium on Software Reliability Engineering 2016, Ottawa, Ontario, Canada, 23/10/16. https://doi.org/10.1109/ISSREW.2016.30

A study of redundant metrics in defect prediction datasets. / Jiarpakdee, Jirayus; Tantithamthavorn, Chakkrit; Ihara, Akinori et al.

Proceedings - 2016 IEEE 27th International Symposium on Software Reliability Engineering Workshops: ISSREW 2016. ed. / Jeremy Bradbury. Piscataway NJ USA : IEEE, Institute of Electrical and Electronics Engineers, 2016. p. 51-52 7789378.

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Other

TY - GEN

T1 - A study of redundant metrics in defect prediction datasets

AU - Jiarpakdee, Jirayus

AU - Tantithamthavorn, Chakkrit

AU - Ihara, Akinori

AU - Matsumoto, Kenichi

N1 - Conference code: 27th

PY - 2016

Y1 - 2016

N2 - Defect prediction models can help Software Quality Assurance (SQA) teams understand their past pitfalls that lead to defective modules. However, the conclusions that are derived from defect prediction models without mitigating redundant metrics issues may be misleading. In this paper, we set out to investigate if redundant metrics issues are affecting defect prediction studies, and its degree and causes of redundancy. Through a case study of 101 publicly-available defect datasets of systems that span both proprietary and open source domains, we observe that (1) 10%-67% of metrics of the studied defect datasets are redundant, and (2) the redundancy of metrics has to do with the aggregation functions of metrics. These findings suggest that researchers should be aware of redundant metrics prior to constructing a defect prediction model in order to maximize internal validity of their studies.

AB - Defect prediction models can help Software Quality Assurance (SQA) teams understand their past pitfalls that lead to defective modules. However, the conclusions that are derived from defect prediction models without mitigating redundant metrics issues may be misleading. In this paper, we set out to investigate if redundant metrics issues are affecting defect prediction studies, and its degree and causes of redundancy. Through a case study of 101 publicly-available defect datasets of systems that span both proprietary and open source domains, we observe that (1) 10%-67% of metrics of the studied defect datasets are redundant, and (2) the redundancy of metrics has to do with the aggregation functions of metrics. These findings suggest that researchers should be aware of redundant metrics prior to constructing a defect prediction model in order to maximize internal validity of their studies.

KW - Defect prediction models

KW - Redundant metrics

KW - Software quality assurance

UR - http://www.scopus.com/inward/record.url?scp=85009723271&partnerID=8YFLogxK

U2 - 10.1109/ISSREW.2016.30

DO - 10.1109/ISSREW.2016.30

M3 - Conference Paper

AN - SCOPUS:85009723271

SP - 51

EP - 52

BT - Proceedings - 2016 IEEE 27th International Symposium on Software Reliability Engineering Workshops

A2 - Bradbury, Jeremy

PB - IEEE, Institute of Electrical and Electronics Engineers

CY - Piscataway NJ USA

T2 - International Symposium on Software Reliability Engineering 2016

Y2 - 23 October 2016 through 27 October 2016

ER -

Jiarpakdee J, Tantithamthavorn C, Ihara A, Matsumoto K. A study of redundant metrics in defect prediction datasets. In Bradbury J, editor, Proceedings - 2016 IEEE 27th International Symposium on Software Reliability Engineering Workshops: ISSREW 2016. Piscataway NJ USA: IEEE, Institute of Electrical and Electronics Engineers. 2016. p. 51-52. 7789378 https://doi.org/10.1109/ISSREW.2016.30

A study of redundant metrics in defect prediction datasets

Abstract

Conference

Keywords

Access to Document

Other files and links

Cite this