Title :
EM algorithm for discrete software reliability models: a unified parameter estimation method
Author :
Okamura, Hiroyuki ; Murayama, Atsushi ; Dohi, Tadashi
Author_Institution :
Dept. of Inf. Eng., Hiroshima Univ., Japan
Abstract :
In this paper, we consider the discrete software reliability models to assess the software reliability in discrete time circumstance, and develop a unified parameter estimation method based on the EM (Expectation-Maximization) principle. In numerical examples, the effectiveness of the EM algorithm to estimate the model parameters in the discrete software reliability models is investigated with real software error data. It is shown that the proposed method can provide the accurate estimates of model parameters effectively.
Keywords :
program testing; software reliability; EM algorithm; discrete software reliability models; discrete time circumstance; expectation-maximization algorithm; unified parameter estimation; Maximum likelihood estimation; Nonlinear equations; Parameter estimation; Phase estimation; Reliability engineering; Software algorithms; Software debugging; Software reliability; Software testing; Stochastic processes;
Conference_Titel :
High Assurance Systems Engineering, 2004. Proceedings. Eighth IEEE International Symposium on
Print_ISBN :
0-7695-2094-4
DOI :
10.1109/HASE.2004.1281746