As shown in table 8, and figure 17, it shows that the ROC area is highest at the Naïve Bayes followed by Bayes Network. The algorithms that produced the best results for the precision criterion are the J48 followed by Naive Bayes and Bayes Network. In conclusion, Naïve Bayes and Bayes Network are the best algorithm to predict the number of injuries, in reference to RMSE, ROC area, accuracy, F-measure, and recall statistical measures. The only exclusion is the precision value, where KNN had the best precision. The KNN showed the best accuracy to predict the number of killed in a collision.