Time Series Forecasting using the Evolutionary Fuzzy System

A. Thammano and L. Attavirojanakul (Thailand)


Time Series, Forecasting, Fuzzy System, Evolutionary Algorithm, Hybrid System


This paper proposes a new hybrid fuzzy system, which is the combination of the fuzzy system and the evolutionary algorithm, capable of modeling a complex time series data and forecasting its future values. The evolutionary algorithm is employed in two phases during the training process of the proposed system. In phase 1, the evolutionary algorithm is used to evolve the fuzzy rule base so that a created rule-base is optimal. In phase 2, it is used to fine tune the fuzzy membership functions of the system. The performance of this proposed system is compared with that of the backpropagation neural network and that of the standard fuzzy system. The results obtained are very promising.

Important Links:

Go Back