Compact TS-fuzzy models through clustering and OLS plus FIS model reduction

Research output: Contribution to journalArticle

24 Citations (Scopus)


Identification of uncertain and nonlinear systems is an important and challenging problem. Fuzzy models of the Takagi-Sugeno (TS) type may be a good choice to describe such systems, however in many cases these become soon complex. We noticed that generally too less effort is put into variable selection and the creation of suitable local rules. Moreover, in most cases no model reduction is applied, while this may simplify the model by removing redundant information. To handle these issues, we propose a three-step method to obtain compact TS-models that can be effectively used to represent complex systems. First, a new fuzzy clustering method is proposed for identification of compact TS-models. Second, the most relevant consequent variables of the TS-model are selected by an orthogonal least squares method based on the obtain clusters. Third, for selection or relevant antecedent variables, a new method is proposed based on Fisher's interclass separability criterion. The overall approach is demonstrated by means of the MPG (miles per gallon) nonlinear regression benchmark. Results are compared with those obtained by standard linear, neuro-fuzzy and advanced fuzzy clustering-based identification tools.

Original languageEnglish
Pages (from-to)1420-1423
Number of pages4
JournalIEEE International Conference on Fuzzy Systems
Publication statusPublished - Jan 1 2001


  • Compact TS-fuzzy model
  • Fischer's interclass separability
  • Gath-Geva fuzzy clustering
  • Model reduction
  • OLS

ASJC Scopus subject areas

  • Software
  • Theoretical Computer Science
  • Artificial Intelligence
  • Applied Mathematics

Fingerprint Dive into the research topics of 'Compact TS-fuzzy models through clustering and OLS plus FIS model reduction'. Together they form a unique fingerprint.

  • Cite this