A Comparative Study of Ordering and Classification of Fault-ProneSoftware Modules

  • Authors:

  • Taghi M. Khoshgoftaar;Edward B. Allen

  • Affiliations:

  • Dept. of Computer Science and Engineering, Florida Atlantic University, Boca Raton, Florida 33431 USA;Dept. of Computer Science and Engineering, Florida Atlantic University, Boca Raton, Florida 33431 USA

  • Venue:
  • Empirical Software Engineering
  • Year:
  • 1999

Quantified Score

Hi-index 0.00

Visualization

Abstract

Softwarequality models can predict the quality of modules early enoughfor cost-effective prevention of problems. For example, softwareproduct and process metrics can be the basis for predicting reliability.Predicting the exact number of faults is often not necessary;classification models can identify fault-prone modules. However,such models require that ’’fault-prone‘‘ be defined before modeling,usually via a threshold. This may not be practical due to uncertainlimits on the amount of reliability-improvement effort. In suchcases, predicting the rank-order of modules is more useful. A module-order model predicts the rank-orderof modules according to a quantitative quality factor, such asthe number of faults. This paper demonstrates how module-ordermodels can be used for classification, and compares them withstatistical classification models. Two case studies of full-scale industrial software systems comparednonparametric discriminant analysis with module-order models. One case studyexamined a military command, control, and communications system.The other studied a large legacy telecommunications system. Wefound that module-order models give management more flexiblereliability enhancement strategies than classification models,and in these case studies, yielded more accurate results thancorresponding discriminant models.