A New Approach to Multi-variable Fuzzy Forecasting Using Picture Fuzzy Clustering and Picture Fuzzy Rule Interpolation Method

摘要

In this paper, a new approach to multi-variable fuzzy forecasting using picture fuzzy clustering and picture fuzzy rule interpolation techniques is proposed. Firstly, we partition dataset into clusters using picture fuzzy clustering algorithm. Secondly, we construct picture fuzzy rules based on given clusters. Finally, we determine the predicted outputs based on the picture fuzzy rule interpolation scheme. Our proposed approach is applied to forecast the Taiwan Stock Exchange Capitalization Weighted Stock Index (TAIEX) data. The experimental results indicate that our method predicts better forecasting results than some relevant ones.