Similarity and Compatibility in Fuzzy Set Theory

by Valerie V. Cross, Thomas A. Sudkamp

Assessing the degree to which two objects, an object and a query, or two concepts are similar or compatible is a fundamental component of human reasoning and consequently is critical in the development of automated diagnosis, classification, information retrieval and decision systems. The assessment of similarity has played an important role in such diverse disciplines such as taxonomy, psychology, and the social sciences. Each discipline has proposed methods for quantifying similarity judgments suitable for its particular applications. This book presents a unified approach to quantifying similarity and compatibility within the framework of fuzzy set theory and examines the primary importance of these concepts in approximate reasoning. Examples of the application of similarity measures in various areas including expert systems, information retrieval, and intelligent database systems are provided.

220 pages, Hardcover

First published April 8, 2002

Reviews

[Matthew MacLennan · Rating 5 out of 5]

Zadeh, what have you done to math!

This book focused on similarity measures (e.g. Jaccard, Euclidean, Hausdorff, Bhattacharya, norms, etc...) and sets in fuzzy set theory. Similarity is used in everything from webpage searching to aligning DNA sequences and fuzzifying it is not trivial. This book is only about 200 pages, but is packed with descriptions, definitions and comparisons of different similarity measures and their uses in knowledge systems (taxonomies etc...). A taxonomy of compatibility measures is also presented which is useful.

It is good to have this one on hand as a reference!