Feature Group Tabular Transformer: A Novel Approach to Traffic Crash Modeling and Causality Analysis

TL;DR

This paper presents a novel Feature Group Tabular Transformer model for traffic crash prediction and causality analysis, integrating diverse data sources for improved interpretability and performance.

Contribution

The introduction of the FGTT model that organizes data into meaningful feature groups for enhanced prediction and causal understanding in traffic crash modeling.

Findings

FGTT outperforms tree ensemble models in predictive accuracy.

Model interpretation identifies key factors influencing crash types.

Provides new insights into crash causality mechanisms.

Abstract

Reliable and interpretable traffic crash modeling is essential for understanding causality and improving road safety. This study introduces a novel approach to predicting collision types by utilizing a comprehensive dataset fused from multiple sources, including weather data, crash reports, high-resolution traffic information, pavement geometry, and facility characteristics. Central to our approach is the development of a Feature Group Tabular Transformer (FGTT) model, which organizes disparate data into meaningful feature groups, represented as tokens. These group-based tokens serve as rich semantic components, enabling effective identification of collision patterns and interpretation of causal mechanisms. The FGTT model is benchmarked against widely used tree ensemble models, including Random Forest, XGBoost, and CatBoost, demonstrating superior predictive performance. Furthermore, model interpretation reveals key influential factors, providing fresh insights into the underlying causality of distinct crash types.