A type-2 data mining optimization for predicting pistachio global market

Author

Golsefid, S. MalekMohamadi ; Turksen, I.B. ; Zarandi, M.H.F.

Author_Institution

Dept. of Ind. Eng., Amirkabir Univ. of Technol., Tehran, Iran

fYear

2012

fDate

6-8 Aug. 2012

Firstpage

1

Lastpage

6

Abstract

In this paper, a type-2 fuzzy TSK expert system is developed for optimizing the global market prediction. Interval type-2 fuzzy logic system permits us to model rule uncertainties and every membership value of an element is interval itself. The proposed type-2 fuzzy model applies the variables which indicate the export trade trend during the specified period. The PCM algorithm is employed to partitions international market into a collection of clusters, and the clusters are converted into a rule base of type-2 fuzzy TSK IF-THEN rule. The proposed model is implemented for forecasting export value of international market segment of pistachio from Iran. This model can be used for selecting proper segment of international market by predicting export value for every product.

Keywords

agricultural products; data mining; forecasting theory; fuzzy logic; international trade; optimisation; pattern clustering; possibility theory; Iran; PCM algorithm; export trade trend; export value forecasting; export value prediction; global market prediction optimization; international market partitioning; international market segment; interval type-2 fuzzy logic system; membership value; pistachio global market prediction; possibilistic clustering algorithm; rule uncertainty modeling; type-2 data mining optimization; type-2 fuzzy TSK IF-THEN rule; type-2 fuzzy TSK expert system; Clustering algorithms; Engines; Fuzzy sets; Genetic algorithms; Input variables; Market research; Uncertainty; Prediction; Type-2 TSK; Type-2 fuzzy modeling; global market; international market segmentation;

fLanguage

English

Publisher

ieee

Conference_Titel

Fuzzy Information Processing Society (NAFIPS), 2012 Annual Meeting of the North American

Conference_Location

Berkeley, CA

ISSN

pending

Print_ISBN

978-1-4673-2336-9

Electronic_ISBN

pending

Type

conf

DOI

10.1109/NAFIPS.2012.6291065

Filename

6291065