Predictive Modeling of Traffic Accidents: A Data-Driven Approach

DOI

10.2991/978-94-6463-866-0_11How to use a DOI?

Keywords

Traffic Accident Prediction; Predictive Models; Data-Driven

Abstract

Traffic accidents are the main cause of deaths, disabilities, and hospital visits in the whole country without stopping. This research is a deep study aiming to choose the best predictive models that can predict traffic accidents. Therefore, cars, whose drivers are essential in managing road safety, should be able to easily understand our findings. That is to say, we are taking a close look to the aspects that drivers are already familiar with, like type of vehicle, age, gender, time of day, and weather, etc., which will enhance our interpretations and therefore predictions and recommendations. Apart from all types of models, two of them Random Forest and Logistic Regression seem to do fine with regard to their predictive performance. Moreover, one should not underestimate the contribution of Optimal Classification Trees due to the fact that they help driver be more clear and practical in their decision making. A different strategy is located in our investigation of the geospatial data with the K-means clustering algorithm to highlight some regions that are less safe than others. The spatial analysis has a fantastic application for the city and traffic engineers as well as the transportation policymakers. The whole point is, by using different approaches, our study is successful not only in discovering the secrets of making predictions as accurate as possible but also in finding the most practical way of using this knowledge in solving real-life problems. Put differently, this research can be considered a good example of the potential of AI-supported models for improving the road safety. Through aiding drivers to effectively watch out for dangerous places and through assisting decision-makers in all improvements, one of the main purposes and benefits of this study is the reduction of the number of traffic-related mishaps, the instilling of the new safer driving art, and the idea of the smart transportation system.

Copyright

© 2025 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  - A. Venkata Kiran
AU  - T. Kiran Sai Pavan
AU  - S. Tanuj Reddy
AU  - S. Suchitra
PY  - 2025
DA  - 2025/10/31
TI  - Predictive Modeling of Traffic Accidents: A Data-Driven Approach
BT  - Proceedings of the International Conference on Intelligent Systems and Digital Transformation (ICISD 2025)
PB  - Atlantis Press
SP  - 107
EP  - 114
SN  - 2589-4919
UR  - https://doi.org/10.2991/978-94-6463-866-0_11
DO  - 10.2991/978-94-6463-866-0_11
ID  - Kiran2025
ER  -