Road Vision: An Integrated Web-Based System for Road Accident Risk Prediction Using Multi-Algorithm Ensemble Machine Learning
- Version
- Download 0
- File Size 388.61 KB
- File Count 1
- Create Date 21 January 2026
- Last Updated 21 January 2026
Road Vision: An Integrated Web-Based System for Road Accident Risk Prediction Using Multi-Algorithm Ensemble Machine Learning
Mr C. SENTHILKUMARAN., MCA,
(Associate Professor, Master of Computer Applications)
K. SUDHARSHINI., MCA
Christ College of Engineering and Technology
Moolakulam, Oulgaret Municipality, Puducherry – 605010.
Abstract
The escalating global incidence of road accidents necessitates intelligent predictive systems that can assess risk factors and prevent potential collisions. This paper presents RoadVision, a comprehensive web-based Road Accident Risk Prediction System that employs ensemble machine learning models to classify accident risk into three categories: Low, Medium, and High. The system analyzes eight critical driving parameters-weather conditions, road type, light conditions, vehicle type, speed, traffic density, alcohol involvement, and time of day-using two powerful machine learning algorithms: XGBoost and Random Forest. RoadVision achieves remarkable prediction accuracy, with XGBoost reaching 100% and Random Forest achieving 97% on test data. Implemented as a modular Flask web application with interactive visualization dashboards, the system offers users real-time risk assessment, comparative algorithm analysis, and comprehensive reporting capabilities. Experimental results demonstrate that environmental factors (weather, light conditions) and behavioral factors (speed, alcohol involvement) provide the strongest predictive signals for accident risk. The system provides a practical, scalable solution for transportation authorities, insurance companies, and safety researchers, with measurable performance metrics and an intuitive user interface designed for both technical and non-technical users.
Keywords: Road Safety, Accident Prediction, Machine Learning, XGBoost, Random Forest, Ensemble Methods, Web Application, Interactive Visualization, Transportation Analytics
Download