Predicting Traffic Accident Severity with Deep Neural Networks

2025-09-03

AI Summary

This study addresses key challenges in traffic accident severity prediction—namely, high-dimensional input features, severe multicollinearity, and class imbalance. To tackle these issues, we propose a hybrid deep learning architecture integrating an autoencoder with a deep feedforward neural network. First, feature multicollinearity analysis is performed to identify and retain discriminative variables. Next, an unsupervised autoencoder performs dimensionality reduction while preserving task-relevant latent representations. Finally, a fully connected classifier is trained for multi-level severity classification. The proposed method significantly enhances model generalizability and robustness on imbalanced traffic datasets. Under cross-validation, it achieves a classification accuracy of 92%, outperforming conventional approaches. Moreover, the architecture offers improved interpretability through feature selection and latent-space regularization. This work provides a high-accuracy, explainable deep learning framework for intelligent traffic risk assessment.

Technology Category

• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• Computer Vision: Multi-modal Vision
• Natural Language Processing: Safety and Robustness

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Traffic accidents can be studied to mitigate the risk of further events. Recent advances in machine learning have provided an alternative way to study data associated with traffic accidents. New models achieve good generalization and high predictive power over imbalanced data. In this research, we study neural network-based models on data related to traffic accidents. We begin analyzing relative feature colinearity and unsupervised dimensionality reduction through autoencoders, followed by a dense network. The features are related to traffic accident data and the target is to classify accident severity. Our experiments show cross-validated results of up to 92% accuracy when classifying accident severity using the proposed deep neural network.

Problem

• Predicting traffic accident severity using deep neural networks
• Analyzing feature colinearity and dimensionality reduction techniques
• Classifying accident severity with high accuracy on imbalanced data

Innovation

• Deep neural networks for accident severity classification
• Autoencoders for unsupervised dimensionality reduction
• Dense network achieving 92% cross-validated accuracy