An attempt to line-wise bug prediction using moving window

Keigo Fukutani, Akito Monden, Zeynep Yucel, Hideaki Hata

Research output: Contribution to journalArticle

Abstract

This paper presents a feasibility study to predict software bugs in statement level. The proposed method uses a window to inspect a given source file containing a bug. By moving the window from the top to the end of a source file, a set of buggy windows and non-buggy windows are derived. Then, we construct a spam filter-based bug prediction model, whose input is a set of tokens in a window and output is the probability of containing a bug in the window. Finally, we integrate bug prediction results on these windows to compute the probability of containing a bug in each statement. We conducted a feasibility study using 593 source files each containing one bug, and found that, by using the size=3 window, we could identify 67.1% of bugs in top 10 buggy statements of each source file prioritized by our bug prediction, which suggests that the bug pinpointing is feasible.

Original languageEnglish
Pages (from-to)122-128
Number of pages7
JournalComputer Software
Volume35
Issue number4
Publication statusPublished - Jan 1 2018

ASJC Scopus subject areas

  • Software

Cite this

An attempt to line-wise bug prediction using moving window. / Fukutani, Keigo; Monden, Akito; Yucel, Zeynep; Hata, Hideaki.

In: Computer Software, Vol. 35, No. 4, 01.01.2018, p. 122-128.

Research output: Contribution to journalArticle

Fukutani, Keigo ; Monden, Akito ; Yucel, Zeynep ; Hata, Hideaki. / An attempt to line-wise bug prediction using moving window. In: Computer Software. 2018 ; Vol. 35, No. 4. pp. 122-128.
@article{56c76ce0410e41ca84db31f61257b7e0,
title = "An attempt to line-wise bug prediction using moving window",
abstract = "This paper presents a feasibility study to predict software bugs in statement level. The proposed method uses a window to inspect a given source file containing a bug. By moving the window from the top to the end of a source file, a set of buggy windows and non-buggy windows are derived. Then, we construct a spam filter-based bug prediction model, whose input is a set of tokens in a window and output is the probability of containing a bug in the window. Finally, we integrate bug prediction results on these windows to compute the probability of containing a bug in each statement. We conducted a feasibility study using 593 source files each containing one bug, and found that, by using the size=3 window, we could identify 67.1{\%} of bugs in top 10 buggy statements of each source file prioritized by our bug prediction, which suggests that the bug pinpointing is feasible.",
author = "Keigo Fukutani and Akito Monden and Zeynep Yucel and Hideaki Hata",
year = "2018",
month = "1",
day = "1",
language = "English",
volume = "35",
pages = "122--128",
journal = "Computer Software",
issn = "0289-6540",
publisher = "Japan Society for Software Science and Technology",
number = "4",

}

TY - JOUR

T1 - An attempt to line-wise bug prediction using moving window

AU - Fukutani, Keigo

AU - Monden, Akito

AU - Yucel, Zeynep

AU - Hata, Hideaki

PY - 2018/1/1

Y1 - 2018/1/1

N2 - This paper presents a feasibility study to predict software bugs in statement level. The proposed method uses a window to inspect a given source file containing a bug. By moving the window from the top to the end of a source file, a set of buggy windows and non-buggy windows are derived. Then, we construct a spam filter-based bug prediction model, whose input is a set of tokens in a window and output is the probability of containing a bug in the window. Finally, we integrate bug prediction results on these windows to compute the probability of containing a bug in each statement. We conducted a feasibility study using 593 source files each containing one bug, and found that, by using the size=3 window, we could identify 67.1% of bugs in top 10 buggy statements of each source file prioritized by our bug prediction, which suggests that the bug pinpointing is feasible.

AB - This paper presents a feasibility study to predict software bugs in statement level. The proposed method uses a window to inspect a given source file containing a bug. By moving the window from the top to the end of a source file, a set of buggy windows and non-buggy windows are derived. Then, we construct a spam filter-based bug prediction model, whose input is a set of tokens in a window and output is the probability of containing a bug in the window. Finally, we integrate bug prediction results on these windows to compute the probability of containing a bug in each statement. We conducted a feasibility study using 593 source files each containing one bug, and found that, by using the size=3 window, we could identify 67.1% of bugs in top 10 buggy statements of each source file prioritized by our bug prediction, which suggests that the bug pinpointing is feasible.

UR - http://www.scopus.com/inward/record.url?scp=85056903260&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85056903260&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:85056903260

VL - 35

SP - 122

EP - 128

JO - Computer Software

JF - Computer Software

SN - 0289-6540

IS - 4

ER -