Taylor-based optimized recursive extended exponential smoothed neural networks forecasting method

The development of a Time series Forecasting System is major concern for Artificial Intelligence researchers. Commonly, existing systems only assess temporal features and analyze the behavior data over time, thus, resulting in uncertain forecasting accuracy. Although many were proposed literature; they have not yet answered attending question. Hence, to overcome this problematic, we propose an innovative method called Taylor-based Optimized Recursive Extended Exponential Smoothed Neural Networks method, abbreviated as TOREESNN. Briefly explained, technique introduces three ideas solve issue: First, building framework univariate time based on theory. Second, designing Elman Classifier model uncertainty prediction order correct forecasted values. And finally hybrading two recurrent one obtain final results. Experimental results demonstrate that has high accuracy both training testing terms Mean Squared Error (MSE) outperforms state-of-the-art Recurrent models Mackey-Glass, Nonlinear Auto-Regressive Moving Average (NARMA), Lorenz, Henon map datasets.