Causal Factors Quantification of Traffic Accidents Based on Mutual Information

For a variety of reasons, there are more traffic accidents worldwide these days. The causal effect in the context of traffic accidents refers to the impact or influence that specific factors or variables have on the occurrence, severity, or outcome of accidents. Thus, it is up to researchers to determine the causes of accidents. Identifying the causal relationship between the risk factors (such as weather, human circumstances, road condition, etc.) and traffic accidents is the main problem in this scenario. In this study, we present a system for mutual information-based quantification of the causes of traffic accidents. The fatality of the accident was predicted using Support Vector Machine (SVM). The causal effects determined by the mutual information give 90% accuracy in determining whether an accident is fatal or not.