Title
Using Mutual Information to Determine Relevance in Bayesian Networks
Abstract
. The control of Bayesian network (BN) evaluation is importantin the development of real-time decision making systems. Techniqueswhich focus attention by considering the relevance of variables in a BNallow more efficient use of computational resources. The statistical conceptof mutual information (MI) between two related random variablescan be used to measure relevance. We extend this idea to present a newmeasure of arc weights in a BN, and show how these can be combined togive a measure ...
Year
DOI
Venue
1998
10.1007/BFb0095287
PRICAI
Keywords
Field
DocType
determine relevance,mutual information,bayesian networks,bayesian network
Random variable,Heuristic,Bhattacharyya distance,Computer science,Decision support system,Bayesian network,Mutual information,Artificial intelligence,Probabilistic logic,Machine learning,Computational complexity theory
Conference
Volume
ISSN
ISBN
1531
0302-9743
3-540-65271-X
Citations 
PageRank 
References 
14
1.04
11
Authors
2
Name
Order
Citations
PageRank
Ann E. Nicholson169288.01
Nathalie Jitnah2637.98