Multitask Hypergraph Convolutional Networks: A Heterogeneous Traffic Prediction Framework

Traffic prediction methods on a single-source data have achieved excellent results in recent years, especially the Graph Convolutional Networks (GCN) based models with spatio-temporal dependency. In reality, various modes of urban transportation operate simultaneously. They influence and complement each other in common space-time occasions, constituting the transportation system dynamically. Thus, traffic data from multiple sources is ostensibly heterogeneous, but internally correlated. The typical single data driven models are, however, not universally applicable for heterogeneous traffic data. To address this issue, we propose a Multi-task Hypergraph Convolutional Neural Network (MT-HGCN) for the multi-source traffic prediction problem. The framework consists of a main task and a related task. Both tasks are based on Hypergraph Convolutional Neural Networks (HGCN) and are devoted to two prediction problems. Furthermore, the tasks are bridged by a feature compress unit, which models the correlation and shares the latent feature to improve the performance of the main task. The node-level forecasting has been evaluated on historical datasets of Beijing to verify the effectiveness of the proposed method. Compared with the state-of-the-arts, the superior performance of the proposed method can be obtained.

I. Introduction

Traffic prediction is the cornerstone of intelligent transportation systems. Accurate prediction methods are critical for practical traffic applications. For instance, road traffic speed and state forecast is of great significance for dynamic management of urban traffic, improvement of quality and efficiency of the logistics industry, and route planning of public travel. Due to the irregular structure of the traffic network and the variable temporal characteristics of traffic flow, the traffic forecasting problem presents challenges that traditional time series forecasting problems do not possess.