Evaluating the Applicability of Reliability Prediction Models between Different Software

Shin Ichi Sato, Akito Monden, Ken Ichi Matsumoto

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

2 Citations (Scopus)

Abstract

The prediction of fault-prone modules in a large software system is an important part in software evolution. Since prediction models in past studies have been constructed and used for individual systems, it has not been practically investigated whether a prediction model based on one system can also predict fault-prone modules accurately in other systems. Our expectation is that if we could build a model applicable to different systems, it would be extremely useful for software companies because they do not need to invest manpower and time for gathering data to construct a new model for every system. In this study, we evaluated the applicability of prediction models between two software systems through two experiments. In the first experiment, a prediction model using 19 module metrics as predictor variables was constructed in each system and was applied to the opposite system mutually. The result showed predictors were too fit to the base data and could not accurately predict fault-prone modules in the different system. On the basis of this result, we focused on a set of predictors showing great effectiveness in every model; and, in consequent, we identified two metrics (Lines of Code and Maximum Nesting Level) as commonly effective predictors in all the models. In the second experiment, by constructing prediction models using only these two metrics, prediction performance were dramatically improved. This result suggests that the commonly effective model applicable to more than two systems can be constructed by focusing on commonly effective predictors.

Original languageEnglish
Title of host publicationInternational Workshop on Principles of Software Evolution (IWPSE)
EditorsM. Aoyama, K. Inoue, V. Rajlich
Pages97-102
Number of pages6
Publication statusPublished - 2002
Externally publishedYes
EventFifth International Workshop on Principles of Software Evolution IWPSE 2002 - Orlando, FL, United States
Duration: May 19 2002May 20 2002

Other

OtherFifth International Workshop on Principles of Software Evolution IWPSE 2002
CountryUnited States
CityOrlando, FL
Period5/19/025/20/02

Fingerprint

Experiments
Industry

Keywords

  • Fault-prone module
  • Metrics
  • Prediction
  • Software measurement
  • Software reliability

ASJC Scopus subject areas

  • Software

Cite this

Sato, S. I., Monden, A., & Matsumoto, K. I. (2002). Evaluating the Applicability of Reliability Prediction Models between Different Software. In M. Aoyama, K. Inoue, & V. Rajlich (Eds.), International Workshop on Principles of Software Evolution (IWPSE) (pp. 97-102)

Evaluating the Applicability of Reliability Prediction Models between Different Software. / Sato, Shin Ichi; Monden, Akito; Matsumoto, Ken Ichi.

International Workshop on Principles of Software Evolution (IWPSE). ed. / M. Aoyama; K. Inoue; V. Rajlich. 2002. p. 97-102.

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

Sato, SI, Monden, A & Matsumoto, KI 2002, Evaluating the Applicability of Reliability Prediction Models between Different Software. in M Aoyama, K Inoue & V Rajlich (eds), International Workshop on Principles of Software Evolution (IWPSE). pp. 97-102, Fifth International Workshop on Principles of Software Evolution IWPSE 2002, Orlando, FL, United States, 5/19/02.
Sato SI, Monden A, Matsumoto KI. Evaluating the Applicability of Reliability Prediction Models between Different Software. In Aoyama M, Inoue K, Rajlich V, editors, International Workshop on Principles of Software Evolution (IWPSE). 2002. p. 97-102
Sato, Shin Ichi ; Monden, Akito ; Matsumoto, Ken Ichi. / Evaluating the Applicability of Reliability Prediction Models between Different Software. International Workshop on Principles of Software Evolution (IWPSE). editor / M. Aoyama ; K. Inoue ; V. Rajlich. 2002. pp. 97-102
@inproceedings{9491a1809843497499af1a8f781ea7b3,
title = "Evaluating the Applicability of Reliability Prediction Models between Different Software",
abstract = "The prediction of fault-prone modules in a large software system is an important part in software evolution. Since prediction models in past studies have been constructed and used for individual systems, it has not been practically investigated whether a prediction model based on one system can also predict fault-prone modules accurately in other systems. Our expectation is that if we could build a model applicable to different systems, it would be extremely useful for software companies because they do not need to invest manpower and time for gathering data to construct a new model for every system. In this study, we evaluated the applicability of prediction models between two software systems through two experiments. In the first experiment, a prediction model using 19 module metrics as predictor variables was constructed in each system and was applied to the opposite system mutually. The result showed predictors were too fit to the base data and could not accurately predict fault-prone modules in the different system. On the basis of this result, we focused on a set of predictors showing great effectiveness in every model; and, in consequent, we identified two metrics (Lines of Code and Maximum Nesting Level) as commonly effective predictors in all the models. In the second experiment, by constructing prediction models using only these two metrics, prediction performance were dramatically improved. This result suggests that the commonly effective model applicable to more than two systems can be constructed by focusing on commonly effective predictors.",
keywords = "Fault-prone module, Metrics, Prediction, Software measurement, Software reliability",
author = "Sato, {Shin Ichi} and Akito Monden and Matsumoto, {Ken Ichi}",
year = "2002",
language = "English",
pages = "97--102",
editor = "M. Aoyama and K. Inoue and V. Rajlich",
booktitle = "International Workshop on Principles of Software Evolution (IWPSE)",

}

TY - GEN

T1 - Evaluating the Applicability of Reliability Prediction Models between Different Software

AU - Sato, Shin Ichi

AU - Monden, Akito

AU - Matsumoto, Ken Ichi

PY - 2002

Y1 - 2002

N2 - The prediction of fault-prone modules in a large software system is an important part in software evolution. Since prediction models in past studies have been constructed and used for individual systems, it has not been practically investigated whether a prediction model based on one system can also predict fault-prone modules accurately in other systems. Our expectation is that if we could build a model applicable to different systems, it would be extremely useful for software companies because they do not need to invest manpower and time for gathering data to construct a new model for every system. In this study, we evaluated the applicability of prediction models between two software systems through two experiments. In the first experiment, a prediction model using 19 module metrics as predictor variables was constructed in each system and was applied to the opposite system mutually. The result showed predictors were too fit to the base data and could not accurately predict fault-prone modules in the different system. On the basis of this result, we focused on a set of predictors showing great effectiveness in every model; and, in consequent, we identified two metrics (Lines of Code and Maximum Nesting Level) as commonly effective predictors in all the models. In the second experiment, by constructing prediction models using only these two metrics, prediction performance were dramatically improved. This result suggests that the commonly effective model applicable to more than two systems can be constructed by focusing on commonly effective predictors.

AB - The prediction of fault-prone modules in a large software system is an important part in software evolution. Since prediction models in past studies have been constructed and used for individual systems, it has not been practically investigated whether a prediction model based on one system can also predict fault-prone modules accurately in other systems. Our expectation is that if we could build a model applicable to different systems, it would be extremely useful for software companies because they do not need to invest manpower and time for gathering data to construct a new model for every system. In this study, we evaluated the applicability of prediction models between two software systems through two experiments. In the first experiment, a prediction model using 19 module metrics as predictor variables was constructed in each system and was applied to the opposite system mutually. The result showed predictors were too fit to the base data and could not accurately predict fault-prone modules in the different system. On the basis of this result, we focused on a set of predictors showing great effectiveness in every model; and, in consequent, we identified two metrics (Lines of Code and Maximum Nesting Level) as commonly effective predictors in all the models. In the second experiment, by constructing prediction models using only these two metrics, prediction performance were dramatically improved. This result suggests that the commonly effective model applicable to more than two systems can be constructed by focusing on commonly effective predictors.

KW - Fault-prone module

KW - Metrics

KW - Prediction

KW - Software measurement

KW - Software reliability

UR - http://www.scopus.com/inward/record.url?scp=1442327860&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=1442327860&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:1442327860

SP - 97

EP - 102

BT - International Workshop on Principles of Software Evolution (IWPSE)

A2 - Aoyama, M.

A2 - Inoue, K.

A2 - Rajlich, V.

ER -