Please use this identifier to cite or link to this item:
Title: A new approach to fuzzy rule generation : fuzzy extension matrix
Authors: Wang, XZ
Wang, YD
Xu, XF
Ling, WD
Yeung, DS
Keywords: Extension matrix
Fuzzy entropy
Heuristic algorithm
Knowledge acquisition
Learning from fuzzy examples
Issue Date: 2001
Publisher: Elsevier
Source: Fuzzy sets and systems, 2001, v. 123, no. 3, p. 291-306 How to cite?
Journal: Fuzzy Sets and Systems 
Abstract: This paper proposes a new approach to fuzzy rule generation from a set of examples with fuzzy representation. The new approach called fuzzy extension matrix incorporates the fuzzy entropy to search for paths and generalizes the concept of crisp extension matrix. By discussing paths of the fuzzy extension matrix, a new heuristic algorithm for generating fuzzy rules is introduced. Compared with the crisp extension matrix, the proposed method has the capability of handling fuzzy representation and tolerating noisy data or missing data. A case study shows that the proposed heuristic algorithm partially inherits the advantages from the crisp case such as simplicity of rules and high learning accuracy. The proposed approach offers a new, practical way to automatically acquire imprecise knowledge.
ISSN: 0165-0114
DOI: 10.1016/S0165-0114(01)00002-1
Appears in Collections:Journal/Magazine Article

View full-text via PolyU eLinks SFX Query
Show full item record

Google ScholarTM



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.