Title :
A Gibbs-sampler approach to estimate the number of faults in a system using capture-recapture sampling [software reliability]
Author :
Hayakawa, Yu ; Yip, Paul S F
Author_Institution :
Sch. of Math. & Comput. Sci., Victoria Univ., Wellington, New Zealand
fDate :
12/1/2000 12:00:00 AM
Abstract :
A new recapture debugging model is suggested to estimate the number of faults in a system, ν, and the failure intensity of each fault, φ. The Gibbs sampler and the Metropolis algorithm are used in this inference procedure. A numerical illustration suggests a notable improvement on the estimation of ν and φ compared with that of a removal debugging model
Keywords :
estimation theory; failure analysis; inference mechanisms; software reliability; Gibbs-sampler approach; Metropolis algorithm; capture-recapture sampling; estimation performance; failure intensity; inference procedure; recapture debugging model; software reliability; system fault number estimation; Counting circuits; Debugging; Fault detection; Fault diagnosis; Inference algorithms; Maximum likelihood estimation; Predictive models; Sampling methods; Software reliability; Statistical distributions;
Journal_Title :
Reliability, IEEE Transactions on
