Publication detail

Similarity Measures in Case-Based Reasoning

DVOŘÁK, J., HODÁL, J.

Original Title

Similarity Measures in Case-Based Reasoning

Type

conference paper

Language

English

Original Abstract

Case-based reasoning (CBR) is a recent approach to problem solving. CBR system solves a new problem by adapting known solutions of similar previously solved problems (cases). Therefore the concept of similarity is the key notion in CBR. A case is characterized by a structure (usually a vector) being composed of attributes. An important task in CBR is to accumulate local similarity measures of attributes into a global similarity measure of the whole object. In this paper we study various similarity measures and possibilities of the accumulation of local measures into the global similarity measure. The main attention is devoted to the fuzzy similarity and fuzzy set-based techniques for matching in the phase of retrieving similar cases.

Key words in English

Case-based reasoning, retrieval, similarity measures, fuzzy similarity

Authors

DVOŘÁK, J., HODÁL, J.

RIV year

2001

Released

20. 8. 2001

Publisher

Hochschule Zittau/Gorlitz

Location

Zittau

ISBN

3-9808089-0-4

Book

Proceedings of the 9th Zittau Fuzzy Colloquium

Pages from

21

Pages to

28

Pages count

8

BibTex

@inproceedings{BUT3373,
  author="Jiří {Dvořák} and Jaroslav {Hodál}",
  title="Similarity Measures in Case-Based Reasoning",
  booktitle="Proceedings of the 9th Zittau Fuzzy Colloquium",
  year="2001",
  pages="8",
  publisher="Hochschule Zittau/Gorlitz",
  address="Zittau",
  isbn="3-9808089-0-4"
}