Table of Contents

Prerequisites on Probability Theory.- Prerequisites on Probability Theory.- Probabilistic Graphical Models.- Causal and Bayesian Networks.- Building Models.- Belief Updating in Bayesian Networks.- Analysis Tools for Bayesian Networks.- Parameter estimation.- Learning the Structure of Bayesian Networks.- Bayesian Networks as Classifiers.- Decision Graphs.- Graphical Languages for Specification of Decision Problems.- Solution Methods for Decision Graphs.- Methods for Analyzing Decision Problems.

Reviews

From the reviews: MATHEMATICAL REVIEWS "This is indeed an invaluable text for students in information technology, engineering, and statistics. It is also very helpful for researchers in these fields and for those working in industry. The book is self-contained...The book has enough illustrative examples and exercises for the reader. All the illustrations are motivated by real applications. Moreover, the book provides a good balance between pure mathematical treatment and the applied aspects of the subject." "The Bayesian network (BN), or probabilistic expert system, is technology for automating human-life reasoning under uncertainty in specific contexts. ... the book does an admirable job of concisely explaining a great range of concepts and techniques. ... the book is very well written and to my knowledge nothing else meets its specific goal of quickly equipping the reader with both practical skills and sufficient theoretical background. ... I certainly would not want to try to implement a BN application without reading this book." (David Tritchler, Sankhya: Indian Journal of Statistics, Vol. 64 (B Part 3), 2002) "Professor Jensen is certainly one of the most influential researchers in the field of Bayesian networks and it is not surprising that this book represents a very clear and useful presentation of the main properties and use of graphical models. ... I think that the present volume represents a useful integration of other material and a compact guide for either a student who wants an introduction to the field or a teacher who needs a reference for a course on probabilistic reasoning in AI." (Luigi Portinale, The Computer Journal, Vol. 46 (3), 2003) "This book is an introduction to Bayesian networks at an accessible level for first-year graduate or advanced undergraduate students. ... I found this book to be an excellent introduction to the topic. It is well written, provides broad topic coverage, and is quite accessible to the non-expert. ... I think Bayesian Networks and Decision Graphs would make a fine text for an introductory class in Bayesian networks or a useful reference for anyone interested in learning about the field." (David J. Marchette, Technometrics, Vol. 45 (2), 2003) "I can comfortably recommend this book as a primary source for topics related to Bayesian networks and decision graphs. This would be an excellent edition to any personal library." (Technometrics, Feburary 2008) From the reviews of the second edition: "The present book provides a very readable but also rigorous and comprehensive introduction to the subject. It would make a very good text for a graduate or an advanced undergraduate course. ... Altogether, this is a very useful book for anyone interested in learning Bayesian networks without tears." (Jayanta K. Ghosh, International Statistical Reviews, Vol. 76 (2), 2008) "This book is the second edition of Jensen's Bayesian Networks and Decision Graphs ... . Each chapter ends with a summary section, bibliographic notes, and exercises. ... provides a readable, self-contained, and above all, practical introduction to Bayesian networks and decision graphs. Its treatment is appropriate not just for statisticians, but also for computer scientists, engineers, and others researchers with appropriate mathematical background. ... highly recommend it as a text or a useful reference for anyone interested in probabilistic graphical models or decision graphs." (Alyson G. Wilson, Journal of the American Statistical Association, Vol. 104 (485), March, 2009) "Devoted to Bayesian Networks or Graphical Models and Influence Diagrams, covering a full course with nice exercises ... . It is useful as a reference for special topics. ... strongly recommended for readers or user of BNs who are interested in specifying dependency models. ... great importance to practitioners who try to find causality behind call-backs of products or crashes. ... the book can be recommended to anybody working on the interface of operations research, AI, statistics and computer science." (Hans-J. Lenz, Statistical Papers, Vol. 52, 2011)