Computing Reviews

Preferences and similarities (1st ed.)
Riccia G., Dubois D., Lenz H., Kruse R., Springer Publishing Company, Incorporated,2008. 320 pp.Type:Book
Date Reviewed: 04/02/09

This book is the result of the “International School for the Synthesis of Expert Knowledge (ISSEK)” workshop, held in Udine, Italy. Like the majority of books based on workshops, this volume struggles with identifying threads for its four chapters of collected papers. As a result, each chapter title reads as a list of keywords--for example, chapter 1 is “Similarity, Dominance, Fuzzy Logic and Efficiency.”

Chapter 1 deals with the operational rules approach to similarities, including fuzzy logic (FL) rules and other forms of decision rules. In fact, only the third paper covers bipolar decision rules; the first two cover FL representation of similarity and preferences. The first paper, “Similarity of Fuzzy Sets and Dominance of Random Variables: A Quest for Transitivity,” is worth reading, since it touches on a current FL topic.

Chapter 2 deals with all forms of possibility, though readers may wonder why it is separate from chapter 3, which deals with probability. The first paper in this chapter, “Logical Approaches to Similarity-based Reasoning: an Overview,” has the wrong title in the table of contents and the running headings. In fact, it focuses on fuzzy similarity rules. The first impression one gets is of the fuzzy form of model-based or case-based reasoning, depending on interpretation. Reading the paper, however, it is intriguing to find out why fuzzy similarity-based reasoning is used and what significance the FL gives the models presented. The second paper, “Logics of Similarity and Their Dual Tableaux,” seems to be a continuation of the first one, making the chapter more interesting. The last paper in this chapter, “Proximities in Statistics: Similarity and Distance,” covers distances, and is likely to be of particular interest to anyone who works in information systems.

Chapter 3 is on probability. The first paper, “Similarity Relations and Independence Concepts,” looks promising, but the content falls short. The second paper, “Imprecision and Structure in Modelling Subjective Similarity,” is quite interesting. A more extended version may prove to be of interest to a wider audience, with applications in other artificial intelligence (AI) areas. Similarly, the last paper, “Defensive Forecasting,” is an interesting read, with the specific application area of similarity-based reasoning.

Chapter 4 is the most disappointing. It consists of four papers--papers that did not fit elsewhere. The second paper, “A Snapshot on Reasoning with Qualitative Preference Statements in AI,” is interesting, since it extends the notion of independence-based preference.

In general, the book is a fine literature review and starting point for researchers in similarity-based reasoning, especially information systems. However, it lacks the depth and breadth expected from a workshop volume. For starters, it is FL dominated, with some glimpses of modal logic; alternative multi-valued logics aren’t presented and algorithmic approaches aren’t mentioned. As a workshop volume, it is disappointing. Still, I highly recommend a few of the papers, for beginning researchers in the field.

Reviewer:  Aladdin Ayesh Review #: CR136653 (1002-0148)

Reproduction in whole or in part without permission is prohibited.   Copyright 2024 ComputingReviews.com™
Terms of Use
| Privacy Policy