교통사고, 유형을 읽다: 클러스터링을 통한 분석과 예방 전략

Domestic traffic accidents are a serious social problem causing human and material damage, with the annual number of accidents not significantly decreasing and the number of traffic accident deaths among OECD countries remaining high. Accordingly, analyzing the fundamental causes of traffic accidents and devising practical preventive measures are emerging as important tasks. This study aims to cluster traffic accident types by applying the K-modes clustering (Cluster Analysis) technique using domestic traffic accident data, and to analyze the characteristics of each accident type to propose effective preventive measures. In particular, it seeks to enhance the effectiveness of traffic accident prevention policies through more comprehensive data analysis, overcoming the limitations of previous studies that were limited to specific regions, age groups, and vehicle types. This study conducted a more comprehensive analysis of traffic accident types by utilizing traffic accident data from all regions of Korea, and complemented the limitations of previous studies by suggesting effective preventive measures in administrative, institutional, and educational aspects. In practice, it can be used as basic data for the police and traffic safety agencies to establish more effective crackdowns and preventive policies, and academically, it can promote a multidisciplinary approach to traffic safety research.