Bayesian Software Reliability Prediction Using Software Metrics Information
Michael Peter Wiper, Ana Paula Palacios, Juan Miguel Marín · Quality Technology & Quantitative Management · 2012
This paper analyzes software reliability models where covariate information in the form of software metrics is available. Our approach uses neural network regression to estimate failure rates in models based on inter failure times or numbers of failures. Inference is carried out using a Bayesian approach which is implemented using the free software packages Winbugs and R. A real data example is used to illustrate the paper.