The collapsing method of defuzzification for discretised interval type-2 fuzzy sets

  • Authors:

  • Sarah Greenfield;Francisco Chiclana;Simon Coupland;Robert John

  • Affiliations:

  • Centre for Computational Intelligence, School of Computing, De Montfort University, Leicester, LE1 9BH, United Kingdom;Centre for Computational Intelligence, School of Computing, De Montfort University, Leicester, LE1 9BH, United Kingdom;Centre for Computational Intelligence, School of Computing, De Montfort University, Leicester, LE1 9BH, United Kingdom;Centre for Computational Intelligence, School of Computing, De Montfort University, Leicester, LE1 9BH, United Kingdom

  • Venue:
  • Information Sciences: an International Journal
  • Year:
  • 2009

Quantified Score

Hi-index 0.07

Visualization

Abstract

This paper proposes a new approach for defuzzification of interval type-2 fuzzy sets. The collapsing method converts an interval type-2 fuzzy set into a type-1 representative embedded set (RES), whose defuzzified values closely approximates that of the type-2 set. As a type-1 set, the RES can then be defuzzified straightforwardly. The novel representative embedded set approximation (RESA), to which the method is inextricably linked, is expounded, stated and proved within this paper. It is presented in two forms: Simple RESA: this approximation deals with the most simple interval FOU, in which a vertical slice is discretised into 2 points. Interval RESA: this approximation concerns the case in which a vertical slice is discretised into 2 or more points. The collapsing method (simple RESA version) was tested for accuracy and speed, with excellent results on both criteria. The collapsing method proved more accurate than the Karnik-Mendel iterative procedure (KMIP) for an asymmetric test set. For both a symmetric and an asymmetric test set, the collapsing method outperformed the KMIP in relation to speed.