A hybrid faulty module prediction using association rule mining and logistic regression analysis

Yasutaka Kamei, Akito Monden, Shuji Morisaki, Ken Ichi Matsumoto

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

20 Citations (Scopus)

Abstract

This paper proposes a fault-prone module prediction method that combines association rule mining with logistic regression analysis. In the proposed method, we focus on three key measures of interestingness of an association rule (support, confidence and lift) to select useful rules for the prediction. If a module satisfies the premise (i.e. the condition in the antecedent part) of one of the selected rules, the module is classified by the rule as either fault-prone or not. Otherwise, the module is classified by the logistic model. We experimentally evaluated the prediction performance of the proposed method with different thresholds of each rule interestingness measure (support, confidence and lift) using a module set in the Eclipse project, and compared it with three well-known fault-proneness models (logistic regression model, linear discriminant model and classification tree). The result showed that the improvement of the Fl-value of the proposed method was 0.163 at maximum compared to conventional models.

Original languageEnglish
Title of host publicationESEM'08: Proceedings of the 2008 ACM-IEEE International Symposium on Empirical Software Engineering and Measurement
Pages279-281
Number of pages3
DOIs
Publication statusPublished - 2008
Externally publishedYes
Event2nd International Symposium on Empirical Software Engineering and Measurement, ESEM 2008 - Kaiserslautern, Germany
Duration: Oct 9 2008Oct 10 2008

Other

Other2nd International Symposium on Empirical Software Engineering and Measurement, ESEM 2008
CountryGermany
CityKaiserslautern
Period10/9/0810/10/08

Fingerprint

Association rules
Regression analysis
Logistics

Keywords

  • Association rule mining
  • Empirical study
  • Fault-prone module prediction
  • Logistic regression analysis

ASJC Scopus subject areas

  • Computer Science Applications
  • Software
  • Electrical and Electronic Engineering

Cite this

Kamei, Y., Monden, A., Morisaki, S., & Matsumoto, K. I. (2008). A hybrid faulty module prediction using association rule mining and logistic regression analysis. In ESEM'08: Proceedings of the 2008 ACM-IEEE International Symposium on Empirical Software Engineering and Measurement (pp. 279-281) https://doi.org/10.1145/1414004.1414051

A hybrid faulty module prediction using association rule mining and logistic regression analysis. / Kamei, Yasutaka; Monden, Akito; Morisaki, Shuji; Matsumoto, Ken Ichi.

ESEM'08: Proceedings of the 2008 ACM-IEEE International Symposium on Empirical Software Engineering and Measurement. 2008. p. 279-281.

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

Kamei, Y, Monden, A, Morisaki, S & Matsumoto, KI 2008, A hybrid faulty module prediction using association rule mining and logistic regression analysis. in ESEM'08: Proceedings of the 2008 ACM-IEEE International Symposium on Empirical Software Engineering and Measurement. pp. 279-281, 2nd International Symposium on Empirical Software Engineering and Measurement, ESEM 2008, Kaiserslautern, Germany, 10/9/08. https://doi.org/10.1145/1414004.1414051
Kamei Y, Monden A, Morisaki S, Matsumoto KI. A hybrid faulty module prediction using association rule mining and logistic regression analysis. In ESEM'08: Proceedings of the 2008 ACM-IEEE International Symposium on Empirical Software Engineering and Measurement. 2008. p. 279-281 https://doi.org/10.1145/1414004.1414051
Kamei, Yasutaka ; Monden, Akito ; Morisaki, Shuji ; Matsumoto, Ken Ichi. / A hybrid faulty module prediction using association rule mining and logistic regression analysis. ESEM'08: Proceedings of the 2008 ACM-IEEE International Symposium on Empirical Software Engineering and Measurement. 2008. pp. 279-281
@inproceedings{81de343fed8d49fb81964ba1ba81c802,
title = "A hybrid faulty module prediction using association rule mining and logistic regression analysis",
abstract = "This paper proposes a fault-prone module prediction method that combines association rule mining with logistic regression analysis. In the proposed method, we focus on three key measures of interestingness of an association rule (support, confidence and lift) to select useful rules for the prediction. If a module satisfies the premise (i.e. the condition in the antecedent part) of one of the selected rules, the module is classified by the rule as either fault-prone or not. Otherwise, the module is classified by the logistic model. We experimentally evaluated the prediction performance of the proposed method with different thresholds of each rule interestingness measure (support, confidence and lift) using a module set in the Eclipse project, and compared it with three well-known fault-proneness models (logistic regression model, linear discriminant model and classification tree). The result showed that the improvement of the Fl-value of the proposed method was 0.163 at maximum compared to conventional models.",
keywords = "Association rule mining, Empirical study, Fault-prone module prediction, Logistic regression analysis",
author = "Yasutaka Kamei and Akito Monden and Shuji Morisaki and Matsumoto, {Ken Ichi}",
year = "2008",
doi = "10.1145/1414004.1414051",
language = "English",
isbn = "9781595939715",
pages = "279--281",
booktitle = "ESEM'08: Proceedings of the 2008 ACM-IEEE International Symposium on Empirical Software Engineering and Measurement",

}

TY - GEN

T1 - A hybrid faulty module prediction using association rule mining and logistic regression analysis

AU - Kamei, Yasutaka

AU - Monden, Akito

AU - Morisaki, Shuji

AU - Matsumoto, Ken Ichi

PY - 2008

Y1 - 2008

N2 - This paper proposes a fault-prone module prediction method that combines association rule mining with logistic regression analysis. In the proposed method, we focus on three key measures of interestingness of an association rule (support, confidence and lift) to select useful rules for the prediction. If a module satisfies the premise (i.e. the condition in the antecedent part) of one of the selected rules, the module is classified by the rule as either fault-prone or not. Otherwise, the module is classified by the logistic model. We experimentally evaluated the prediction performance of the proposed method with different thresholds of each rule interestingness measure (support, confidence and lift) using a module set in the Eclipse project, and compared it with three well-known fault-proneness models (logistic regression model, linear discriminant model and classification tree). The result showed that the improvement of the Fl-value of the proposed method was 0.163 at maximum compared to conventional models.

AB - This paper proposes a fault-prone module prediction method that combines association rule mining with logistic regression analysis. In the proposed method, we focus on three key measures of interestingness of an association rule (support, confidence and lift) to select useful rules for the prediction. If a module satisfies the premise (i.e. the condition in the antecedent part) of one of the selected rules, the module is classified by the rule as either fault-prone or not. Otherwise, the module is classified by the logistic model. We experimentally evaluated the prediction performance of the proposed method with different thresholds of each rule interestingness measure (support, confidence and lift) using a module set in the Eclipse project, and compared it with three well-known fault-proneness models (logistic regression model, linear discriminant model and classification tree). The result showed that the improvement of the Fl-value of the proposed method was 0.163 at maximum compared to conventional models.

KW - Association rule mining

KW - Empirical study

KW - Fault-prone module prediction

KW - Logistic regression analysis

UR - http://www.scopus.com/inward/record.url?scp=62949103673&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=62949103673&partnerID=8YFLogxK

U2 - 10.1145/1414004.1414051

DO - 10.1145/1414004.1414051

M3 - Conference contribution

AN - SCOPUS:62949103673

SN - 9781595939715

SP - 279

EP - 281

BT - ESEM'08: Proceedings of the 2008 ACM-IEEE International Symposium on Empirical Software Engineering and Measurement

ER -