Training Algorithm for Neuro-Fuzzy Network Based on Singular Spectrum Analysis

TL;DR

This paper introduces a novel training algorithm for neuro-fuzzy networks that combines Singular Spectrum Analysis for noise reduction with neural prediction models, improving long-term prediction accuracy on sunspot data.

Contribution

It presents a new combined approach integrating SSA-based noise filtering with neural network training, enhancing prediction performance over raw data methods.

Findings

Noise reduction via SSA improves data quality.

Enhanced long-term prediction accuracy.

Superior performance on sunspot number prediction.

Abstract

In this article, we propose a combination of an noise-reduction algorithm based on Singular Spectrum Analysis (SSA) and a standard feedforward neural prediction model. Basically, the proposed algorithm consists of two different steps: data preprocessing based on the SSA filtering method and step-by-step training procedure in which we use a simple feedforward multilayer neural network with backpropagation learning. The proposed noise-reduction procedure successfully removes most of the noise. That increases long-term predictability of the processed dataset comparison with the raw dataset. The method was applied to predict the International sunspot number RZ time series. The results show that our combined technique has better performances than those offered by the same network directly applied to raw dataset.