Multiple bug spectral fault localization using genetic programming

Lee Naish; Neelofar Neelofar; Kotagiri Ramamohanarao

doi:10.1109/ASWEC.2015.12

Multiple bug spectral fault localization using genetic programming

Lee Naish, Neelofar, Kotagiri Ramamohanarao

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research › peer-review

12 Citations (Scopus)

Abstract

Debugging is crucial for producing reliable software. One of the effective bug localization techniques is Spectral-Based Fault Localization (SBFL). It locates a buggy statement by applying an evaluation metric to program spectra and ranking program components on the basis of the score it computes. Recently, genetic programming has been proposed as a way to find good metrics. We have found that the huge search space for metrics can cause this approach to be slow and unreliable, even for relatively simple data sets. Here we propose a restricted class of “hyperbolic” metrics, with a small number of numeric parameters. This class of functions is based on past theoretical and empirical results. We show that genetic programming can reliably discover effective metrics over a wide range of data sets of program spectra. We evaluate the performance for both real programs and model programs with single bugs, multiple bugs, “deterministic” bugs and nondeterministic bugs.

Original language	English
Title of host publication	Proceedings - 2015 24th Australasian Software Engineering Conference, ASWEC 2015
Publisher	IEEE, Institute of Electrical and Electronics Engineers
Pages	11-17
Number of pages	7
ISBN (Electronic)	9781467393904
DOIs	https://doi.org/10.1109/ASWEC.2015.12
Publication status	Published - 2015
Externally published	Yes
Event	Australasian Software Engineering Conference 2015 - Adelaide, Australia Duration: 28 Sept 2015 → 1 Oct 2015 Conference number: 24th https://web.archive.org/web/20151213123955/http://www.aswec2015.org/ https://ieeexplore.ieee.org/xpl/conhome/7364551/proceeding (Proceedings)

Conference

Conference	Australasian Software Engineering Conference 2015
Abbreviated title	ASWEC 2015
Country/Territory	Australia
City	Adelaide
Period	28/09/15 → 1/10/15
Internet address	https://web.archive.org/web/20151213123955/http://www.aswec2015.org/ https://ieeexplore.ieee.org/xpl/conhome/7364551/proceeding (Proceedings)

Access to Document

10.1109/ASWEC.2015.12

Cite this

@inproceedings{6e65d85183ec4fc7bca348e3edc1f4f9,

title = "Multiple bug spectral fault localization using genetic programming",

abstract = "Debugging is crucial for producing reliable software. One of the effective bug localization techniques is Spectral-Based Fault Localization (SBFL). It locates a buggy statement by applying an evaluation metric to program spectra and ranking program components on the basis of the score it computes. Recently, genetic programming has been proposed as a way to find good metrics. We have found that the huge search space for metrics can cause this approach to be slow and unreliable, even for relatively simple data sets. Here we propose a restricted class of “hyperbolic” metrics, with a small number of numeric parameters. This class of functions is based on past theoretical and empirical results. We show that genetic programming can reliably discover effective metrics over a wide range of data sets of program spectra. We evaluate the performance for both real programs and model programs with single bugs, multiple bugs, “deterministic” bugs and nondeterministic bugs.",

author = "Lee Naish and Neelofar and Kotagiri Ramamohanarao",

note = "Funding Information: We would like to thank Jason (Hua Jie) Lee, who generated the real program data set we used and continues to be involved in discussions on this research area. Publisher Copyright: {\textcopyright} 2015 IEEE.; Australasian Software Engineering Conference 2015, ASWEC 2015 ; Conference date: 28-09-2015 Through 01-10-2015",

year = "2015",

doi = "10.1109/ASWEC.2015.12",

language = "English",

pages = "11--17",

booktitle = "Proceedings - 2015 24th Australasian Software Engineering Conference, ASWEC 2015",

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

address = "United States of America",

url = "https://web.archive.org/web/20151213123955/http://www.aswec2015.org/, https://ieeexplore.ieee.org/xpl/conhome/7364551/proceeding",

}

Naish, L, Neelofar & Ramamohanarao, K 2015, Multiple bug spectral fault localization using genetic programming. in Proceedings - 2015 24th Australasian Software Engineering Conference, ASWEC 2015. IEEE, Institute of Electrical and Electronics Engineers, pp. 11-17, Australasian Software Engineering Conference 2015, Adelaide, South Australia, Australia, 28/09/15. https://doi.org/10.1109/ASWEC.2015.12

Multiple bug spectral fault localization using genetic programming. / Naish, Lee; Neelofar; Ramamohanarao, Kotagiri.
Proceedings - 2015 24th Australasian Software Engineering Conference, ASWEC 2015. IEEE, Institute of Electrical and Electronics Engineers, 2015. p. 11-17.

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research › peer-review

TY - GEN

T1 - Multiple bug spectral fault localization using genetic programming

AU - Naish, Lee

AU - Neelofar, null

AU - Ramamohanarao, Kotagiri

N1 - Conference code: 24th

PY - 2015

Y1 - 2015

N2 - Debugging is crucial for producing reliable software. One of the effective bug localization techniques is Spectral-Based Fault Localization (SBFL). It locates a buggy statement by applying an evaluation metric to program spectra and ranking program components on the basis of the score it computes. Recently, genetic programming has been proposed as a way to find good metrics. We have found that the huge search space for metrics can cause this approach to be slow and unreliable, even for relatively simple data sets. Here we propose a restricted class of “hyperbolic” metrics, with a small number of numeric parameters. This class of functions is based on past theoretical and empirical results. We show that genetic programming can reliably discover effective metrics over a wide range of data sets of program spectra. We evaluate the performance for both real programs and model programs with single bugs, multiple bugs, “deterministic” bugs and nondeterministic bugs.

AB - Debugging is crucial for producing reliable software. One of the effective bug localization techniques is Spectral-Based Fault Localization (SBFL). It locates a buggy statement by applying an evaluation metric to program spectra and ranking program components on the basis of the score it computes. Recently, genetic programming has been proposed as a way to find good metrics. We have found that the huge search space for metrics can cause this approach to be slow and unreliable, even for relatively simple data sets. Here we propose a restricted class of “hyperbolic” metrics, with a small number of numeric parameters. This class of functions is based on past theoretical and empirical results. We show that genetic programming can reliably discover effective metrics over a wide range of data sets of program spectra. We evaluate the performance for both real programs and model programs with single bugs, multiple bugs, “deterministic” bugs and nondeterministic bugs.

UR - http://www.scopus.com/inward/record.url?scp=85082804976&partnerID=8YFLogxK

U2 - 10.1109/ASWEC.2015.12

DO - 10.1109/ASWEC.2015.12

M3 - Conference Paper

AN - SCOPUS:85082804976

SP - 11

EP - 17

BT - Proceedings - 2015 24th Australasian Software Engineering Conference, ASWEC 2015

PB - IEEE, Institute of Electrical and Electronics Engineers

T2 - Australasian Software Engineering Conference 2015

Y2 - 28 September 2015 through 1 October 2015

ER -

Multiple bug spectral fault localization using genetic programming

Abstract

Conference

Access to Document

Other files and links

Cite this