The impact of changes mislabeled by SZZ on Just-in-Time defect prediction

Yuanrui Fan; Xin Xia; Daniel Alencar da Costa; David Lo; Ahmed E. Hassan; Shanping Li

doi:10.1109/TSE.2019.2929761

The impact of changes mislabeled by SZZ on Just-in-Time defect prediction

Yuanrui Fan, Xin Xia, Daniel Alencar da Costa, David Lo, Ahmed E. Hassan, Shanping Li

Department of Software Systems & Cybersecurity

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

56 Citations (Scopus)

Abstract

Just-in-Time (JIT) defect prediction---a technique which aims to predict bugs at change level---has been paid more attention. JIT defect prediction leverages the SZZ approach to identify bug-introducing changes. Recently, researchers found that the performance of SZZ (including its variants) is impacted by many noises. SZZ may considerably mislabel changes that are used to train a JIT defect prediction model, and thus impact the prediction accuracy.

Original language	English
Pages (from-to)	1559-1586
Number of pages	26
Journal	IEEE Transactions on Software Engineering
Volume	47
Issue number	8
DOIs	https://doi.org/10.1109/TSE.2019.2929761
Publication status	Published - 1 Aug 2021

Keywords

Just-in-Time Defect Prediction
Mining Software Repositories
Noisy Data
SZZ

Access to Document

10.1109/TSE.2019.2929761

Cite this

@article{410cf10496cb4ad09ffc4e2e22b11872,

title = "The impact of changes mislabeled by SZZ on Just-in-Time defect prediction",

abstract = "Just-in-Time (JIT) defect prediction---a technique which aims to predict bugs at change level---has been paid more attention. JIT defect prediction leverages the SZZ approach to identify bug-introducing changes. Recently, researchers found that the performance of SZZ (including its variants) is impacted by many noises. SZZ may considerably mislabel changes that are used to train a JIT defect prediction model, and thus impact the prediction accuracy.",

keywords = "Just-in-Time Defect Prediction, Mining Software Repositories, Noisy Data, SZZ",

author = "Yuanrui Fan and Xin Xia and {Alencar da Costa}, Daniel and David Lo and Hassan, {Ahmed E.} and Shanping Li",

year = "2021",

month = aug,

day = "1",

doi = "10.1109/TSE.2019.2929761",

language = "English",

volume = "47",

pages = "1559--1586",

journal = "IEEE Transactions on Software Engineering",

issn = "0098-5589",

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

number = "8",

}

TY - JOUR

T1 - The impact of changes mislabeled by SZZ on Just-in-Time defect prediction

AU - Fan, Yuanrui

AU - Xia, Xin

AU - Alencar da Costa, Daniel

AU - Lo, David

AU - Hassan, Ahmed E.

AU - Li, Shanping

PY - 2021/8/1

Y1 - 2021/8/1

N2 - Just-in-Time (JIT) defect prediction---a technique which aims to predict bugs at change level---has been paid more attention. JIT defect prediction leverages the SZZ approach to identify bug-introducing changes. Recently, researchers found that the performance of SZZ (including its variants) is impacted by many noises. SZZ may considerably mislabel changes that are used to train a JIT defect prediction model, and thus impact the prediction accuracy.

AB - Just-in-Time (JIT) defect prediction---a technique which aims to predict bugs at change level---has been paid more attention. JIT defect prediction leverages the SZZ approach to identify bug-introducing changes. Recently, researchers found that the performance of SZZ (including its variants) is impacted by many noises. SZZ may considerably mislabel changes that are used to train a JIT defect prediction model, and thus impact the prediction accuracy.

KW - Just-in-Time Defect Prediction

KW - Mining Software Repositories

KW - Noisy Data

KW - SZZ

UR - http://www.scopus.com/inward/record.url?scp=85069935031&partnerID=8YFLogxK

U2 - 10.1109/TSE.2019.2929761

DO - 10.1109/TSE.2019.2929761

M3 - Article

AN - SCOPUS:85069935031

SN - 0098-5589

VL - 47

SP - 1559

EP - 1586

JO - IEEE Transactions on Software Engineering

JF - IEEE Transactions on Software Engineering

IS - 8

ER -

The impact of changes mislabeled by SZZ on Just-in-Time defect prediction

Abstract

Keywords

Access to Document

Other files and links

Cite this