New directions in fuzzy automata

  • Authors:

  • Mansoor Doostfatemeh;Stefan C. Kremer

  • Affiliations:

  • Department of Computing and Information Science, University of Guelph, Guelph, Ont., Canada N1G 2W1;Department of Computing and Information Science, University of Guelph, Guelph, Ont., Canada N1G 2W1

  • Venue:
  • International Journal of Approximate Reasoning
  • Year:
  • 2005

Quantified Score

Hi-index 0.00

Visualization

Abstract

Automata are the prime example of general computational systems over discrete spaces. The incorporation of fuzzy logic into automata theory resulted in fuzzy auotomata which can handle continuous spaces. Moreover, they are able to model uncertainty which is inherent in many applications. Deterministic Finite-state Automata (DFA) have been the architecture, most used in many applications, but, the increasing interest in using fuzzy logic for many new areas necessitates that the formalism of fuzzy automata be more developed and better established to fulfill implementational requirements in a well-defined manner. This need is due to the fact that despite the long history of fuzzy automata and lots of research being done on that, there are still some issues which have not been well-established and issues which need some kind of revision. In particular, we focus on membership assignment, output mapping, multi-membership resolution, and the concept of acceptance for fuzzy automata. We develop a new general definition for fuzzy automata, and based on that, develop well-defined and application-driven methodologies to establish a better ground for fuzzy automata and pave the way for forthcoming applications.