The Use of Rule-Based Knowledge Discovery Techniques to Profile Black Spots

In Belgium, traffic safety is currently one of the highest topics on the list of priorities of the government. The identification of black spots and black zones and profiling them in terms of accident related data and location characteristics must provide new insights into the complexity and causes of road accidents which, in turn, provide valuable input for government actions. Data mining is the extraction of information from large amounts of data. The use of data mining algorithms is therefore particularly useful in the context of large datasets on road accidents. In this paper, association rules are used to identify accident circumstances that frequently occur together. The strength of this descriptive approach lies within the definition of different accident types and the identification of relevant variables that make a strong contribution towards a better understanding of accident circumstances. An analysis of the produced set of rules, describing underlying patterns in the data, indicates that five aspects of traffic accidents can be discerned: collision with a pedestrian, collision in parallel, sideways collision, week/weekend accidents and weather conditions. For each of these accident types, different variables play an important role in the occurrence of the accidents.