CONSTRUCTING COMPLETE FUZZY RULES OF FUZZY MODEL USING SINGULAR VALUE DECOMPOSITION

Abadi, Agus Maman and Subanar, and Widodo, - and Saleh, Samsubar (2011) CONSTRUCTING COMPLETE FUZZY RULES OF FUZZY MODEL USING SINGULAR VALUE DECOMPOSITION.

Full text not available from this repository.

Abstract

In the fuzzy model, there are many ways to design fuzzy rules from input-output data. Those are gradient descent training, table lookup scheme, recursive least squares and clustering. The aim in this paper is to construct complete fuzzy rules from input-output data using singular value decomposition. Designing fuzzy rules by singular value decomposition is based on minimizing the square of residual between overall output of the real system and identified model. Thus, the designed fuzzy rules are used to construct fuzzy model by choosing fuzzifier, defuzzifier and inference engine. Furthermore, the fuzzy model is applied to predict inflation rate in Indonesia. Key words: fuzzy rules, fuzzy model, singular value decomposition.

Item Type: Article
Subjects: Matematika dan Ilmu Pengetahuan Alam > Matematika
Depositing User: Agus Maman Abadi,S.Si.,M.Si.
Date Deposited: 20 Jul 2012 05:56
Last Modified: 20 Jul 2012 05:56
URI: http://eprints.uny.ac.id/id/eprint/2358

Actions (login required)

View Item View Item