A novel fuzzy time series forecasting method based on the improved artificial fish swarm optimization algorithm

Recently, many forecasting methods have been proposed for the analysis of fuzzy time series. The main factors that affect the results of the forecasting of these models are partition universe of discourse and determination of fuzzy relations. In this paper, a novel fuzzy time series forecasting method which uses a hybrid artificial fish swarm optimization algorithm for the determination of interval lengths is proposed. Firstly, we introduce the chemotactic behavior of Bacterial foraging optimization into foraging behavior. Secondly, the Levy flight is used as the mutation operator for a mutation strategy. Finally, the new proposed method is applied to a fuzzy time series forecasting and the experimental results show that the proposed model obtain better forecasting results than those of other existing models. It proves the feasibility and validity of above-mentioned approaches.