Dynamic Frequency Domain Graph Convolutional Network for Traffic Forecasting

TL;DR

This paper introduces DFDGCN, a novel traffic forecasting model that uses Fourier transform and embeddings to better capture dynamic spatial dependencies amid noise and time-shifts, outperforming existing methods.

Contribution

The paper proposes a dynamic frequency domain graph convolution network that mitigates time-shift effects and noise in traffic data by combining Fourier transform, sensor and time embeddings, and multiple graph types.

Findings

Outperforms baseline models on four real-world datasets.

Effectively captures spatial dependencies despite noise and time-shifts.

Demonstrates robustness and accuracy in traffic forecasting.

Abstract

Complex spatial dependencies in transportation networks make traffic prediction extremely challenging. Much existing work is devoted to learning dynamic graph structures among sensors, and the strategy of mining spatial dependencies from traffic data, known as data-driven, tends to be an intuitive and effective approach. However, Time-Shift of traffic patterns and noise induced by random factors hinder data-driven spatial dependence modeling. In this paper, we propose a novel dynamic frequency domain graph convolution network (DFDGCN) to capture spatial dependencies. Specifically, we mitigate the effects of time-shift by Fourier transform, and introduce the identity embedding of sensors and time embedding when capturing data for graph learning since traffic data with noise is not entirely reliable. The graph is combined with static predefined and self-adaptive graphs during graph convolution to predict future traffic data through classical causal convolutions. Extensive experiments on four real-world datasets demonstrate that our model is effective and outperforms the baselines.