A NEW METHOD FOR GENERATJNG FUZZY RULES FROM TRAINING DATA AND ITS APPLICATION TO FORECASTING INFLATION RATE AND INTEREST RATE OF BANK INDONESIA CERTIFICATE
Table lookup scheme is a simple method to construct fuzzy rules of fuzzy model. That can be used to overcome the conOicting rule by detennining each rule degree. The weakness of fuzzy model based on table lookup scheme is that lhc fuzzy rules may not be COm;Jletc so the fuzzy rules can not cover...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article PeerReviewed |
| Language: | English |
| Published: |
JOURNAL OF QUANTITATIVE M ETHODS
2009
|
| Subjects: | |
| Online Access: | https://repository.ugm.ac.id/32963/1/4.pdf https://repository.ugm.ac.id/32963/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Universitas Gadjah Mada |
| Language: | English |
| Summary: | Table lookup scheme is a simple method to construct fuzzy rules of fuzzy model. That can be used to overcome the conOicting rule by detennining each rule degree. The weakness of fuzzy model based on table lookup scheme is that lhc fuzzy rules may not be COm;Jletc so the fuzzy rules can not cover all val ues in the domain. In this paper a new method to generate fuzzy rules from training data will be proposed. In thsi method, all complete fu7.zy n1lcs are identi ficd by firing strength of each p()ssible fw.zy rule. Then, the resulted fuzzy rules arc used to design fuzzy model. Appli cations of the proposed method to predict the Indonesian inflation rc1t.e and interest rate of 3ank Indonesia Certificate (BJC) will be disc ussed. The predictions of the Indonesian Inflation rate and interest rate of BI C using the proposed method h ave a higher accuracy than those using the table lookup scheme. |
|---|