Computing Reviews
Today's Issue Hot Topics Search Browse Recommended My Account Log In
Review Help
Probabilistic conditional independence structures (Information Science and Statistics)
Studeny M., Springer-Verlag Telos, 2004. Type: Book (9781852338916)
Date Reviewed: Oct 13 2005

Given a finite set of variables N, conditional independence (CI) structures establish a link between A, B, and C, which are subsets of N, in the form “A is conditionally independent of B, given C.” These types of structures are used in computational data modeling and analysis for a variety of applications, some of which are mentioned in chapter 1 of this book. This book looks at the probabilistic version of conditional independence structures, in which a probability measure P over N is introduced into the statement.

The book has a very good introductory chapter (chapter 2), in which the basic elements of conditional independence structures are explained. Chapter 2 establishes the basic formal notation to describe these structures and their statements. The chapter also covers the different classes of probability measures that are used in probabilistic CI structures.

CI structures are often represented in graphical formats. These may be undirected graphs, acyclic directed graphs, or classical chain graphs, to mention a few. Chapter 3 covers the graphical representation of CI in detail, with an extended section on advanced graphical models.

Chapter 4 provides the basics of the particular notion of imsets, introduced briefly at the end of chapter 2. Imsets are integer-valued functions over N. Chapter 4 focuses on structural imsets, which are effectively a combination of elementary imsets.

Chapters 5 and 6 may be of particular interest to language researchers, since they expand on the elements of the CI structures. Chapter 5 covers probabilistic models and how they can be described. Chapter 6 covers equivalence and implication.

Chapters 7 and 8 may be of particular interest to artificial intelligence researchers. Chapter 7 covers the problem with representative choice. Chapter 8 covers learning, classifying approaches to learning CI from sets of data into two groups: significance tests and quality criterion.

The book concludes with open problems (chapter 9), those that are still outstanding in the theory and implementation of probabilistic CI structures and models, and their operations.

Overall, the book is very well organized and presented. It reads very well, and introduces an interesting area of discrete mathematics that has potential in knowledge representation, modeling, and probabilistic reasoning, especially in the field of artificial intelligence.

Reviewer:  Aladdin Ayesh Review #: CR131868 (0608-0791)
Bookmark and Share
Probabilistic Algorithms (Including Monte Carlo) (G.3 ... )
Probabilistic Computation (F.1.2 ... )
Uncertainty, “Fuzzy,” And Probabilistic Reasoning (I.2.3 ... )
Deduction And Theorem Proving (I.2.3 )
Modes Of Computation (F.1.2 )
Would you recommend this review?
Other reviews under "Probabilistic Algorithms (Including Monte Carlo)": Date

Type: Article
Jul 1 1987
A probabilistic lower bound for checking disjointness of sets
Manber U. (ed) Information Processing Letters 19(1): 51-53, 1984. Type: Article
Feb 1 1985
On iterative methods in Markov modelling
Schatte P. Journal of Information Processing and Cybernetics 23(1): 49-51, 1987. Type: Article
Oct 1 1987

E-Mail This Printer-Friendly
Send Your Comments
Contact Us
Reproduction in whole or in part without permission is prohibited.   Copyright 1999-2023 ThinkLoud®
Terms of Use
| Privacy Policy