Learning Bayesian Networks

Richard E. Neapolitan

688 pages, parution le 25/04/2003

Ajouter à une liste

Indisponible

Résumé

For courses in Bayesian Networks or Advanced Networking focusing on Bayesian networks found in departments of Computer Science, Computer Engineering and Electrical Engineering. Also appropriate as a supplementary text in courses on Expert Systems, Machine Learning, and Artificial Intelligence where the topic of Bayesian Networks is covered.

This book provides an accessible and unified discussion of Bayesian networks. It includes discussions of topics related to the areas of artificial intelligence, expert systems and decision analysis, the fields in which Bayesian networks are frequently applied. The author discusses both methods for doing inference in Bayesian networks and influence diagrams. The book also covers the Bayesian method for learning the values of discrete and continuous parameters. Both the Bayesian and constraint-based methods for learning structure are discussed in detail.

Hundreds of examples and problems, allows students to test themselves.
Complex concepts are introduced with simple examples, eases the student into the text without being overwhelmed.
Separation of the statistical development of Bayesian networks from the application to causality, avoids confusion which can occur when the two are covered simultaneously.

Contents

Preface

Basics

Introduction to Bayesian Networks
More DAG/Probability Relationships

Inference

Inference: Discrete Variables
More Inference Algorithms
Influence Diagrams

Learning

Parameter Learning: Binary Variables
More Parameter Learning
Bayesian Structure Learning
Constraint-Based Learning
More Structure Learning

Applications

Bibliography
Index

Caractéristiques techniques

	PAPIER
Éditeur(s)	Prentice Hall
Auteur(s)	Richard E. Neapolitan
Parution	25/04/2003
Nb. de pages	688
Format	18 x 24
Couverture	Relié
Poids	1180g
Intérieur	Noir et Blanc
EAN13	9780130125347
ISBN13	978-0-13-012534-7