Research on classification algorithm and its application in cased-based reasoning

Cased-based reasoning has been used in decision application with less domain knowledge. This paper presents the novel classification algorithm, which is regarded as the basis of the case-based reasoning application on traffic accident analysis and the early warning system. By using the classified centre vector-based classification algorithm, this method can be used to determine the traffic accident cases' affiliation roughly and as a result, to focus on the relevant retrieval scope. This method has been used in the traffic accident analysis and the corresponding early warning system. The experimental result shows the feasibility of the approach.