Profiling Festival-Period Traffic Accidents in Thailand: Clustering and Risk Factors

บทคัดย่อ

The analysis of road accident data currently focuses on collecting and examining information from multiple perspectives, including factors such as the date and time of occurrence, details of the injured parties, alcohol consumption, and safety measures such as seatbelt and helmet usage. This study employs clustering techniques to group similar accident events. In addition to clustering, this study integrates logistic regression and decision tree techniques to enhance predictive capabilities. Logistic regression is used to estimate the probability of accident severity based on contributing factors, while decision tree modeling helps identify key decision rules that influence accident outcomes. To analyze the severity of road accidents, logistic regression can be employed to model the probability of severe outcomes based on contributing factors. A case study in Thailand is conducted to explore accident trends, which helps develop effective safety measures and policies. The findings emphasize the need to refine procedures for transferring emergency patients and implement stricter safety protocols to enhance care efficiency and mitigate adverse consequences.