Please use this identifier to cite or link to this item:
Title: EM algorithms for estimating software reliability based on masked data
Authors: Zhao, M.
Xie, M. 
Issue Date: Jun-1994
Source: Zhao, M.,Xie, M. (1994-06). EM algorithms for estimating software reliability based on masked data. Microelectronics Reliability 34 (6) : 1027-1038. ScholarBank@NUS Repository.
Abstract: In this paper, the software reliability estimation from masked data is considered based on superposition nonhomogeneous Poisson process models. The masked data are the system failure data when the exact causes of the failures, i.e., the components that have caused the system failure, may be unknown. The components of a software system may indicate its modules, testing strategies and the types of errors according to the practical situations. In general, the maximum likelihood estimates of parameters are difficult to find when there exist masked data, because the superposition process cannot be decomposed into the ordinary processes. In this study, the EM algorithm is investigated to solve the problem of maximum likelihood estimation. It is shown that the EM algorithm is powerful to deal with the masked data. By applying the EM algorithm, the masked data problem is simplified and is reduced to the common estimation problem without the masked data. This result makes it very easy to obtain maximum likelihood estimates of parameters. © 1994.
Source Title: Microelectronics Reliability
ISSN: 00262714
Appears in Collections:Staff Publications

Show full item record
Files in This Item:
There are no files associated with this item.

Page view(s)

checked on Feb 22, 2018

Google ScholarTM


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.