Title :
Optimised Generalised Type-2 Join and Meet Operations
Author :
Greenfield, Sarah ; John, Robert
Author_Institution :
De Montfort Univ., Leicester
Abstract :
The inferencing stage of a type-2 fuzzy inferencing system is driven by join and meet operations. As conventionally implemented these algorithms are computationally complex. This article introduces optimised implementations for these operations. These alternative procedures pre-suppose that the grid method of discretisation is adopted. Time comparisons between the alternative and conventional implementations have been undertaken, for join and meet operations coded firstly in isolation, and secondly, within a full prototype type-2 fuzzy inferencing system. The experimental results show that the optimisation affords marked time reduction, particularly under finer discretisation.
Keywords :
decision making; fuzzy set theory; inference mechanisms; decision making program; grid method; type-2 fuzzy inferencing system; Computational complexity; Computational intelligence; Fuzzy logic; Fuzzy sets; Fuzzy systems; Geometry; Inference algorithms; Prototypes; Sampling methods; Uncertainty;
Conference_Titel :
Fuzzy Systems Conference, 2007. FUZZ-IEEE 2007. IEEE International
Conference_Location :
London
Print_ISBN :
1-4244-1209-9
Electronic_ISBN :
1098-7584
DOI :
10.1109/FUZZY.2007.4295355