Data fusion, ensemble and clustering to improve the classification accuracy for the severity of road traffic accidents in Korea

摘要: Increasing amount of road traffic in 1990s has drawn much attention Korea due to its influence on safety problems. Various types data analyses are done order analyze the relationship between severity accident and driving environmental factors based records. Accurate results such analysis can provide crucial information for prevention policy. In this paper, we use various algorithms improve accuracy individual classifiers two categories accident. Individual used neural network decision tree. Mainly three different approaches applied: classifier fusion Dempster-Shafer algorithm, Bayesian procedure logistic model; ensemble arcing bagging; clustering k-means algorithm. Our empirical study indicate that a classification algorithm works best Korea.