Title :
Type-2 fuzzy probabilistic system
Author_Institution :
Centre for Comput. Intell., De Montfort Univ., Leicester, UK
Abstract :
The paper proposes a new type-2 fuzzy logic system, Type-2 Fuzzy Probabilistic System, coupled with type-2 fuzzy probability, to deal with higher order fuzzy events with fuzzy sub-events together. The new concepts of Membership Probability Density Function (Mpdf) and ̃J-plane Representation for type-2 fuzzy set are also presented. A special Type-2 Fuzzy Probabilistic Model, Type-2 Fuzzy ARIMA, is proposed. As a proof-of-concept application, an experiment with a 36 month forecast of a real-valued problem is conducted showing some improvement over Box-Jenkins Model (SARIMA). As series of concepts and results prove, Type-2 Fuzzy Probabilistic System can be determined by fuzzy statistical evaluation of high order fuzzy events and process. This paper realises type-2 fuzzy logic systems evolving from rule-based fuzzy systems to the systems based on type-2 fuzzy probability.
Keywords :
fuzzy logic; fuzzy set theory; knowledge based systems; probability; Box-Jenkins model; J-plane representation; Mpdf; SARIMA; fuzzy statistical evaluation; fuzzy subevents; high order fuzzy events; membership probability density function; proof-of-concept application; real-valued problem; rule-based fuzzy systems; type-2 fuzzy ARIMA; type-2 fuzzy logic system; type-2 fuzzy probabilistic system; type-2 fuzzy set; Correlation; Fuzzy logic; Fuzzy sets; Probabilistic logic; Probability; Probability density function; Time series analysis; ARIMA; characteristic moments; fuzzy event; fuzzy logic system; fuzzy probability; fuzzy subset; real-valued random quantities; type-2 fuzzy sets;
Conference_Titel :
Fuzzy Systems and Knowledge Discovery (FSKD), 2012 9th International Conference on
Conference_Location :
Sichuan
Print_ISBN :
978-1-4673-0025-4
DOI :
10.1109/FSKD.2012.6234273