Enhancing Graph Neural Networks in Large-scale Traffic Incident Analysis with Concurrency Hypothesis

TL;DR

This paper introduces the Concurrency Hypothesis and a novel Concurrency Prior method to improve graph neural network predictions for traffic incidents using large-scale nationwide road network data, validated by extensive experiments.

Contribution

It proposes the Concurrency Hypothesis, introduces new metrics ANCD and ANCC, and develops the Concurrency Prior method to enhance GNN performance in traffic incident prediction.

Findings

Gains of 3% to 13% in F1 score across models

Gains of 1.3% to 9% in AUC metrics

Validation of the Concurrency Hypothesis with real data

Abstract

Despite recent progress in reducing road fatalities, the persistently high rate of traffic-related deaths highlights the necessity for improved safety interventions. Leveraging large-scale graph-based nationwide road network data across 49 states in the USA, our study first posits the Concurrency Hypothesis from intuitive observations, suggesting a significant likelihood of incidents occurring at neighboring nodes within the road network. To quantify this phenomenon, we introduce two novel metrics, Average Neighbor Crash Density (ANCD) and Average Neighbor Crash Continuity (ANCC), and subsequently employ them in statistical tests to validate the hypothesis rigorously. Building upon this foundation, we propose the Concurrency Prior (CP) method, a powerful approach designed to enhance the predictive capabilities of general Graph Neural Network (GNN) models in semi-supervised traffic incident prediction tasks. Our method allows GNNs to incorporate concurrent incident information, as mentioned in the hypothesis, via tokenization with negligible extra parameters. The extensive experiments, utilizing real-world data across states and cities in the USA, demonstrate that integrating CP into 12 state-of-the-art GNN architectures leads to significant improvements, with gains ranging from 3% to 13% in F1 score and 1.3% to 9% in AUC metrics. The code is publicly available at https://github.com/xiwenc1/Incident-GNN-CP.