Discriminating features-based cost-sensitive approach for software defect prediction

Aftab Ali; Naveed Khan; Mamun Abu-Tair; Joost Noppen; Sally I McClean; Ian McChesney

doi:10.1007/s10515-021-00289-8

Discriminating features-based cost-sensitive approach for software defect prediction

Research output: Contribution to journal › Article › peer-review

32 Citations (Scopus)

124 Downloads (Pure)

Abstract

Correlated quality metrics extracted from a source code repository
can be utilized to design a model to automatically predict defects in a software system. It is obvious that the extracted metrics will result in a highly
unbalanced data, since the number of defects in a good quality software system should be far less than the number of normal instances. It is also a fact
that the selection of the best discriminating features significantly improves
the robustness and accuracy of a prediction model. Therefore, the contribution of this paper is twofold, first it selects the best discriminating features
that help in accurately predicting a defect in a software component. Secondly,
a cost-sensitive logistic regression and decision tree ensemble-based prediction
models are applied to the best discriminating features for precisely predicting
a defect in a software component. The proposed models are compared with
the most recent schemes in the literature in terms of accuracy, area under thecurve (AUC), and recall. The models are evaluated using 11 datasets and it
is evident from the results and analysis that the performance of the proposed
prediction models outperforms the schemes in the literature.

Original language	English
Article number	11
Number of pages	18
Journal	Automated software engineering
Volume	28
Early online date	12 Jul 2021
DOIs	https://doi.org/10.1007/s10515-021-00289-8
Publication status	Published online - 12 Jul 2021

Bibliographical note

Funding Information:
This research is supported by the BTIIC (BT Ireland Innovation Centre) project, funded by BT and Invest Northern Ireland.

Publisher Copyright:
© 2021, The Author(s).

Funding

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 9 Industry, Innovation, and Infrastructure

Keywords

AUC
Cost-sensitivity
Discriminating features
Machine learning models
Recall
Software bugs/defects

Access to Document

10.1007/s10515-021-00289-8Licence: CC BY

Ali2021_Article_DiscriminatingFeatures-basedCoFinal published version, 1.25 MBLicence: CC BY

Cite this

@article{f32d15198ab34d1f9d3a4b49cb7598f9,

title = "Discriminating features-based cost-sensitive approach for software defect prediction",

abstract = "Correlated quality metrics extracted from a source code repository can be utilized to design a model to automatically predict defects in a software system. It is obvious that the extracted metrics will result in a highly unbalanced data, since the number of defects in a good quality software system should be far less than the number of normal instances. It is also a fact that the selection of the best discriminating features significantly improves the robustness and accuracy of a prediction model. Therefore, the contribution of this paper is twofold, first it selects the best discriminating features that help in accurately predicting a defect in a software component. Secondly, a cost-sensitive logistic regression and decision tree ensemble-based prediction models are applied to the best discriminating features for precisely predicting a defect in a software component. The proposed models are compared with the most recent schemes in the literature in terms of accuracy, area under thecurve (AUC), and recall. The models are evaluated using 11 datasets and it is evident from the results and analysis that the performance of the proposed prediction models outperforms the schemes in the literature. ",

keywords = "AUC, Cost-sensitivity, Discriminating features, Machine learning models, Recall, Software bugs/defects",

author = "Aftab Ali and Naveed Khan and Mamun Abu-Tair and Joost Noppen and McClean, \{Sally I\} and Ian McChesney",

note = "Funding Information: This research is supported by the BTIIC (BT Ireland Innovation Centre) project, funded by BT and Invest Northern Ireland. Publisher Copyright: {\textcopyright} 2021, The Author(s).",

year = "2021",

month = jul,

day = "12",

doi = "10.1007/s10515-021-00289-8",

language = "English",

volume = "28",

journal = "Automated software engineering",

issn = "0928-8910",

publisher = "Springer Netherlands",

}

TY - JOUR

T1 - Discriminating features-based cost-sensitive approach for software defect prediction

AU - Ali, Aftab

AU - Khan, Naveed

AU - Abu-Tair, Mamun

AU - Noppen, Joost

AU - McClean, Sally I

AU - McChesney, Ian

PY - 2021/7/12

Y1 - 2021/7/12

N2 - Correlated quality metrics extracted from a source code repository can be utilized to design a model to automatically predict defects in a software system. It is obvious that the extracted metrics will result in a highly unbalanced data, since the number of defects in a good quality software system should be far less than the number of normal instances. It is also a fact that the selection of the best discriminating features significantly improves the robustness and accuracy of a prediction model. Therefore, the contribution of this paper is twofold, first it selects the best discriminating features that help in accurately predicting a defect in a software component. Secondly, a cost-sensitive logistic regression and decision tree ensemble-based prediction models are applied to the best discriminating features for precisely predicting a defect in a software component. The proposed models are compared with the most recent schemes in the literature in terms of accuracy, area under thecurve (AUC), and recall. The models are evaluated using 11 datasets and it is evident from the results and analysis that the performance of the proposed prediction models outperforms the schemes in the literature.

AB - Correlated quality metrics extracted from a source code repository can be utilized to design a model to automatically predict defects in a software system. It is obvious that the extracted metrics will result in a highly unbalanced data, since the number of defects in a good quality software system should be far less than the number of normal instances. It is also a fact that the selection of the best discriminating features significantly improves the robustness and accuracy of a prediction model. Therefore, the contribution of this paper is twofold, first it selects the best discriminating features that help in accurately predicting a defect in a software component. Secondly, a cost-sensitive logistic regression and decision tree ensemble-based prediction models are applied to the best discriminating features for precisely predicting a defect in a software component. The proposed models are compared with the most recent schemes in the literature in terms of accuracy, area under thecurve (AUC), and recall. The models are evaluated using 11 datasets and it is evident from the results and analysis that the performance of the proposed prediction models outperforms the schemes in the literature.

KW - AUC

KW - Cost-sensitivity

KW - Discriminating features

KW - Machine learning models

KW - Recall

KW - Software bugs/defects

UR - https://www.scopus.com/pages/publications/85110462591

U2 - 10.1007/s10515-021-00289-8

DO - 10.1007/s10515-021-00289-8

M3 - Article

SN - 0928-8910

VL - 28

JO - Automated software engineering

JF - Automated software engineering

M1 - 11

ER -

Discriminating features-based cost-sensitive approach for software defect prediction

Abstract

Bibliographical note

Funding

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

An Event-level Clustering Framework for Process Mining Using Common Sequential Rules

Cite this