Empirical model for predicting high, medium and low severity faults using object oriented metrics in Mozilla Firefox

  • Authors:

  • Satwinder Singh;Puneet Mittal;K. S. Kahlon

  • Affiliations:

  • Department of CSE & IT, BBSB Engineering College, Fatehgarh Sahib, India;Department of CSE & IT, BBSB Engineering College, Fatehgarh Sahib, India;Department of CSE, Guru Nanak Dev University, Amritsar, India

  • Venue:
  • International Journal of Computer Applications in Technology
  • Year:
  • 2013

Quantified Score

Hi-index 0.00

Visualization

Abstract

There have been numerous studies to predict the error proneness of class. If software testers have only a very limited amount of time left to conduct testing, knowing where the most severe errors are likely to occur in a system is more helpful than just knowing where errors are likely to occur. This paper describes how we calculated various object oriented metrics of three versions of Mozilla Firefox. And after that how we collected all the bugs along with their severity levels in these versions of Firefox using Bugzilla database and associated bugs with class. Logistic regression and neural network techniques are followed to predict the error proneness of class under error category. The findings suggest that various metrics can be used to predict error proneness of class under error category. Neural network approach can predict high and medium severity errors more accurately than the low severity errors.