Predictability classification for software effort estimation

Naoki Kinoshita, Akito Monden, Masateru Tshunoda, Zeynep Yucel

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In this paper, focusing on the problem that estimation accuracy of software development effort greatly varies among software projects, we propose a predictability classification method for software projects before conducting effort estimation. In the proposed method, given a project to be estimated, we first evaluate whether the effort can be accurately estimated or not by identifying the project as 'predictable' or 'unpredictable'. In case of predictable projects, we conduct the effort estimation. Otherwise, estimation is avoided. As a result of an experiment to assess the effectiveness of the proposed method using six industry datasets, (i) the mean square residual and residual variance are shown to be suitable measures for recognition of predictability; and (ii) the average absolute error is significantly reduced in five datasets, by avoiding the estimation when a project belongs to the unpredictable class, which proves the effectiveness of the proposed method. By using the proposed method, practitioners become aware of cases when they can rely on the estimation and when they cannot.

Original languageEnglish
Title of host publicationProceedings - 2018 IEEE/ACIS 3rd International Conference on Big Data, Cloud Computing, Data Science and Engineering, BCD 2018
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages43-48
Number of pages6
ISBN (Electronic)9781538656051
DOIs
Publication statusPublished - Nov 9 2018
Event3rd IEEE/ACIS International Conference on Big Data, Cloud Computing, Data Science and Engineering, BCD 2018 - Yonago, Japan
Duration: Jul 10 2018Jul 12 2018

Other

Other3rd IEEE/ACIS International Conference on Big Data, Cloud Computing, Data Science and Engineering, BCD 2018
CountryJapan
CityYonago
Period7/10/187/12/18

Fingerprint

Software engineering
Industry
Experiments

Keywords

  • Data-mining
  • Empirical-study
  • Multivariate-regression
  • Software-effort-estimation
  • Software-project-management

ASJC Scopus subject areas

  • Information Systems
  • Software
  • Computer Networks and Communications
  • Computer Vision and Pattern Recognition

Cite this

Kinoshita, N., Monden, A., Tshunoda, M., & Yucel, Z. (2018). Predictability classification for software effort estimation. In Proceedings - 2018 IEEE/ACIS 3rd International Conference on Big Data, Cloud Computing, Data Science and Engineering, BCD 2018 (pp. 43-48). [8530690] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/BCD2018.2018.00015

Predictability classification for software effort estimation. / Kinoshita, Naoki; Monden, Akito; Tshunoda, Masateru; Yucel, Zeynep.

Proceedings - 2018 IEEE/ACIS 3rd International Conference on Big Data, Cloud Computing, Data Science and Engineering, BCD 2018. Institute of Electrical and Electronics Engineers Inc., 2018. p. 43-48 8530690.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Kinoshita, N, Monden, A, Tshunoda, M & Yucel, Z 2018, Predictability classification for software effort estimation. in Proceedings - 2018 IEEE/ACIS 3rd International Conference on Big Data, Cloud Computing, Data Science and Engineering, BCD 2018., 8530690, Institute of Electrical and Electronics Engineers Inc., pp. 43-48, 3rd IEEE/ACIS International Conference on Big Data, Cloud Computing, Data Science and Engineering, BCD 2018, Yonago, Japan, 7/10/18. https://doi.org/10.1109/BCD2018.2018.00015
Kinoshita N, Monden A, Tshunoda M, Yucel Z. Predictability classification for software effort estimation. In Proceedings - 2018 IEEE/ACIS 3rd International Conference on Big Data, Cloud Computing, Data Science and Engineering, BCD 2018. Institute of Electrical and Electronics Engineers Inc. 2018. p. 43-48. 8530690 https://doi.org/10.1109/BCD2018.2018.00015
Kinoshita, Naoki ; Monden, Akito ; Tshunoda, Masateru ; Yucel, Zeynep. / Predictability classification for software effort estimation. Proceedings - 2018 IEEE/ACIS 3rd International Conference on Big Data, Cloud Computing, Data Science and Engineering, BCD 2018. Institute of Electrical and Electronics Engineers Inc., 2018. pp. 43-48
@inproceedings{b3cb965ac18449fe83a4ce3e4482ae95,
title = "Predictability classification for software effort estimation",
abstract = "In this paper, focusing on the problem that estimation accuracy of software development effort greatly varies among software projects, we propose a predictability classification method for software projects before conducting effort estimation. In the proposed method, given a project to be estimated, we first evaluate whether the effort can be accurately estimated or not by identifying the project as 'predictable' or 'unpredictable'. In case of predictable projects, we conduct the effort estimation. Otherwise, estimation is avoided. As a result of an experiment to assess the effectiveness of the proposed method using six industry datasets, (i) the mean square residual and residual variance are shown to be suitable measures for recognition of predictability; and (ii) the average absolute error is significantly reduced in five datasets, by avoiding the estimation when a project belongs to the unpredictable class, which proves the effectiveness of the proposed method. By using the proposed method, practitioners become aware of cases when they can rely on the estimation and when they cannot.",
keywords = "Data-mining, Empirical-study, Multivariate-regression, Software-effort-estimation, Software-project-management",
author = "Naoki Kinoshita and Akito Monden and Masateru Tshunoda and Zeynep Yucel",
year = "2018",
month = "11",
day = "9",
doi = "10.1109/BCD2018.2018.00015",
language = "English",
pages = "43--48",
booktitle = "Proceedings - 2018 IEEE/ACIS 3rd International Conference on Big Data, Cloud Computing, Data Science and Engineering, BCD 2018",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - Predictability classification for software effort estimation

AU - Kinoshita, Naoki

AU - Monden, Akito

AU - Tshunoda, Masateru

AU - Yucel, Zeynep

PY - 2018/11/9

Y1 - 2018/11/9

N2 - In this paper, focusing on the problem that estimation accuracy of software development effort greatly varies among software projects, we propose a predictability classification method for software projects before conducting effort estimation. In the proposed method, given a project to be estimated, we first evaluate whether the effort can be accurately estimated or not by identifying the project as 'predictable' or 'unpredictable'. In case of predictable projects, we conduct the effort estimation. Otherwise, estimation is avoided. As a result of an experiment to assess the effectiveness of the proposed method using six industry datasets, (i) the mean square residual and residual variance are shown to be suitable measures for recognition of predictability; and (ii) the average absolute error is significantly reduced in five datasets, by avoiding the estimation when a project belongs to the unpredictable class, which proves the effectiveness of the proposed method. By using the proposed method, practitioners become aware of cases when they can rely on the estimation and when they cannot.

AB - In this paper, focusing on the problem that estimation accuracy of software development effort greatly varies among software projects, we propose a predictability classification method for software projects before conducting effort estimation. In the proposed method, given a project to be estimated, we first evaluate whether the effort can be accurately estimated or not by identifying the project as 'predictable' or 'unpredictable'. In case of predictable projects, we conduct the effort estimation. Otherwise, estimation is avoided. As a result of an experiment to assess the effectiveness of the proposed method using six industry datasets, (i) the mean square residual and residual variance are shown to be suitable measures for recognition of predictability; and (ii) the average absolute error is significantly reduced in five datasets, by avoiding the estimation when a project belongs to the unpredictable class, which proves the effectiveness of the proposed method. By using the proposed method, practitioners become aware of cases when they can rely on the estimation and when they cannot.

KW - Data-mining

KW - Empirical-study

KW - Multivariate-regression

KW - Software-effort-estimation

KW - Software-project-management

UR - http://www.scopus.com/inward/record.url?scp=85058492809&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85058492809&partnerID=8YFLogxK

U2 - 10.1109/BCD2018.2018.00015

DO - 10.1109/BCD2018.2018.00015

M3 - Conference contribution

SP - 43

EP - 48

BT - Proceedings - 2018 IEEE/ACIS 3rd International Conference on Big Data, Cloud Computing, Data Science and Engineering, BCD 2018

PB - Institute of Electrical and Electronics Engineers Inc.

ER -