An Enhancement of a Learning Procedure in Neuro-Fuzzy Model
Alexander Vlasenko, Nataliia Vlasenko, Olena А. Vynokurova, Yevgeniy V. Bodyanskiy · 2018
In this paper we propose an extended version of the learning algorithm for the hybrid five-layer neuro-fuzzy model. Backpropagation algorithm has been applied to the first layer membership functions centers tuning. The experimental results have shown that introduced enhancement significantly improves accuracy in time-series prediction benchmark tests without significant performance cost.