Title
Comparison of Lauritzen-Spiegelhalter and successive restrictions algorithms for computing probability distributions in Bayesian networks
Source of Publication
AIP Conference Proceedings
Abstract
© 2016 Author(s). The basic task of any probabilistic inference system in Bayesian networks is computing the posterior probability distribution for a subset or subsets of random variables, given values or evidence for some other variables from the same Bayesian network. Many methods and algorithms have been developed to exact and approximate inference in Bayesian networks. This work compares two exact inference methods in Bayesian networks-Lauritzen-Spiegelhalter and the successive restrictions algorithm-from the perspective of computational efficiency. The two methods were applied for comparison to a Chest Clinic Bayesian Network. Results indicate that the successive restrictions algorithm shows more computational efficiency than the Lauritzen-Spiegelhalter algorithm.
Document Type
Conference Proceeding
ISBN
['9780735413962']
Publication Date
6-2-2016
DOI
10.1063/1.4952522
Recommended Citation
Smail, Linda, "Comparison of Lauritzen-Spiegelhalter and successive restrictions algorithms for computing probability distributions in Bayesian networks" (2016). Scopus Indexed Articles. 1518.
https://zuscholars.zu.ac.ae/scopus-indexed-articles/1518