Comparative Research on the Unbalanced Processing of Traffic Accident Data and Multi-model Performance

DOI

10.2991/978-94-6239-648-7_50How to use a DOI?

Keywords

Traffic Accident Prediction; Unbalanced Data; Synthetic Minority Oversampling Technique; Machine Learning; Integrated Learning

Abstract

Traffic Accident prediction is of great significance in the construction of smart cities, however, there is an unbalanced challenge in the data of traffic accidents. In response to this problem, three public data sets, Addis Ababa Sub-city Accident, Us Accident and Barcelona Accident, were selected. In response to data imbalance, the Synthetic Minority Oversampling Technique (SMOTE) method was introduced to sample a small number of samples, comprehensively evaluate the performance of the model through indicators such as Accuracy, Precision, F1-score and AUC. The experimental result show that overall performance of the integrated learning model is more stable, with an accuracy rate of 70% to 84%, and it is obviously better than a single model in terms of balance indicators. The SMOTE method alleviates the deviation caused by data imbalance to a certain extent. The research can provide some references for the selection of subsequent traffic accident prediction models and the methods of unbalanced data processing.

Copyright

© 2026 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Tao Hu
PY  - 2026
DA  - 2026/04/24
TI  - Comparative Research on the Unbalanced Processing of Traffic Accident Data and Multi-model Performance
BT  - Proceedings of the International Workshop on Advances in Deep Learning for Image Analysis and Computer Vision (IWADIC 2025)
PB  - Atlantis Press
SP  - 455
EP  - 462
SN  - 2352-538X
UR  - https://doi.org/10.2991/978-94-6239-648-7_50
DO  - 10.2991/978-94-6239-648-7_50
ID  - Hu2026
ER  -