DocumentCode :
3381676
Title :
J̃-plane - a new form of representation for Type-2 Fuzzy Probability
Author :
Yuying Wang
Author_Institution :
Centre for Comput. Intell., De Montfort Univ., Leicester, UK
fYear :
2013
fDate :
7-10 July 2013
Firstpage :
1
Lastpage :
8
Abstract :
The use of higher order fuzzy events to represent real-world problems is always an inhibitor of practical fuzzy probability. In this paper, a new form of representation J̃-plane is proposed for Type-2 Fuzzy Probability. Based on the J̃-plane representation, the type-2 fuzzy computation and type-2 fuzzy characteristic moments are investigated and applied in a case of type-2 fuzzy probabilistic computational model - Type-2 Fuzzy SARIMA. This model is implemented in Wireless Soft-Switch communication network CAPS (Call Attempt Per Second) prediction to show that Type-2 Fuzzy SARIMA is much better than the classic statistical model SARIMA with more accurately and timely forecast results, along with the infinite maximum supported seasonality and less errors. As the series of algorithms and results prove in this paper, J̃-plane representation is viable in real-world problems which facilitate bridging the gap between higher order fuzzy probability realizations and real-world problems.
Keywords :
autoregressive moving average processes; fuzzy logic; fuzzy set theory; J̃-plane representation; call attempt per second prediction; fuzzy characteristic moments; higher order fuzzy probability realizations; type-2 fuzzy SARIMA; type-2 fuzzy probability; wireless soft-switch communication network CAPS; Communication networks; Computational modeling; Correlation; Probability distribution; Random variables; Time series analysis; SARIMA; Type-2 fuzzy event; Type-2 fuzzy probabilistic computational model; real-valued random quantities; type-2 fuzzy probability theory; type-2 fuzzy sets;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Fuzzy Systems (FUZZ), 2013 IEEE International Conference on
Conference_Location :
Hyderabad
ISSN :
1098-7584
Print_ISBN :
978-1-4799-0020-6
Type :
conf
DOI :
10.1109/FUZZ-IEEE.2013.6622352
Filename :
6622352
Link To Document :
بازگشت