# PT-TDGCN: Pre-Trained Trend-Aware Dynamic Graph Convolutional Network for Traffic Flow Prediction

## TL;DR

This paper introduces PT-TDGCN, a new method for predicting traffic flow that improves accuracy by learning dynamic spatial and temporal patterns.

## Contribution

The novel PT-TDGCN framework combines pre-training with dynamic graph learning and trend-aware attention for traffic prediction.

## Key findings

- PT-TDGCN outperformed 14 baseline models on four real-world datasets.
- The method achieves superior predictive accuracy and robustness.
- Dynamic graph learning and trend-aware attention enhance modeling of spatiotemporal patterns.

## Abstract

Accurate traffic flow prediction is vital for intelligent transportation systems, yet strong spatiotemporal coupling and multi-scale dynamics make modelling difficult. Existing methods often rely on static adjacency and short input windows, limiting adaptation to time-varying spatial relations and long-term patterns. To address these issues, we propose the Pre-trained Trend-aware Dynamic Graph Convolutional Network (PT-TDGCN), a two-stage framework. In the pre-training stage, a Transformer-based masked autoencoder learns segment-level temporal representations from historical sequences. In the prediction stage, three designs are integrated: (1) dynamic graph learning parameterized by tensor decomposition; (2) convolutional trend-aware attention that adds 1D convolutions to capture local trends while preserving global context; and (3) spatial graph convolution combined with lightweight fusion projection for aligning pre-trained, spatial, and temporal representations. Extensive experiments on four real-world datasets demonstrated that PT-TDGCN consistently outperformed 14 baseline models, achieving superior predictive accuracy and robustness.