This study looks at a driver’s risk level by incorporating personality traits, demographic characteristics, and driving behaviors. To measure personality traits of the participants, the study used the Chinese version of the Symptom Checklist-90 (SCL-90) scale. To measure the participants' driving, the researchers analyzed the individuals' risky driving behaviors such as speeding, harsh braking, and sharp turns, combining these into a comprehensive risk score. Then, the researchers used ML models, with the Multilayer Perceptron (MP) performing the best, to classify drivers as either being low, medium, or high-risk based on personality traits and demographics. From all the demographics, males tended to be more high-risk drivers compared to females while other variables such as age and driving experience seemed to have little correlation with risk. For mental health characteristics and personality traits, depression, anxiety, and hostility all seemed to increase the likelihood of sharp acceleration, and anxiety combined with paranoid ideation heightened the dangers associated with sharp turning. On the other hand, obsessive-compulsive disorder (OCD) had minimal impact on any risky behaviors. In the discussion, the authors noted how personality traits such as depression, anxiety, hostility, and paranoia were linked to a greater driving risk and that male drivers tended to show more risky behaviors. It was also noted that the MP model was the most effective at identifying risk levels and could be used to create targeted safety training