Time-Aware Fuzzy Neural Network Based on Frequency-Enhanced Modulation Mechanism

Fuzzy neural network (FNN) is regarded as a prominent approach in application of time-series modeling. With the capability of fuzzy reasoning, FNN can capture temporal patterns from the time-series samples. However, the existing FNNs may suffer from the temporal pattern distortion because possibly multiscale features cannot be explored sufficiently. To address this problem, a time-aware fuzzy neural network, based on the frequency-enhanced modulation mechanism (FEM-TAFNN), is developed for time-series prediction in this article. First, a Fourier-based decoder is established to extract the multiscale features. This decoder employs the frequency-domain model to orthogonally separate the time-scale features with different frequencies into independent temporal patterns based on the Fourier basis, which prevents the overlap of temporal patterns using time-domain analysis. Second, a frequency-enhanced modulation mechanism is designed to shape fuzzy rules of FNN based on the contribution of different temporal patterns in the frequency spectrum. It enables FEM-TAFNN to modulate out the realistic multiscale temporal patterns. Finally, the proposed FEM-TAFNN is tested on four multiscale time-series datasets. The empirical results confirm its superior prediction performance than other methods.

I. Introduction

Fuzzy neural network (FNN) is a prominent modeling approach in recent years, which integrates the advanced fuzzy reasoning capability and the powerful learning ability [1], [2], [3]. Numerous research articles have proved that FNN can establish the accurate model of time series under reliable samples [4], [5], [6]. However, there are often multiple temporal patterns in the time series that present multiple time scales and are entangled together [7], [8]. FNNs do not explicitly tackle the multiscale relationship between temporal patterns, which causes the temporal pattern distortion during modeling [9].