Analysis of the Severity of Road Accidents Using Combined Data Mining Techniques

Road traffic accidents represent a critical road safety problem whose severity depends on the complex interaction of multiple factors. The study of these factors and their interrelationship has therefore long been a focus of scientific literature. The objective of this study is to analyze the factors that determine the severity of road accidents, identifying the most important ones and their correlations. An accident dataset incorporating variables related to infrastructure, location, time, and vehicle type was used to predict the Injury Severity Index (ISI), applying Association Rules to identify latent correlations and an Optimized Decision Tree (CART) model for hierarchical risk classification. The results reveal that the Type of Collision is the primary predictor of severity; collisions with objects or pedestrians showed a 100% confidence in resulting in low severity, while maximum severity is associated with heavy traffic and head-on or side-impact collisions. Critical scenarios were also identified during the early morning hours and in rural areas, primarily linked to trucks. The combined use of both tools provides a solid scientific basis for designing interventions on highly vulnerable road segments and during high-risk time periods.