Research on Railway Rescue Scheme Matching Method Based on Confidence Mixed Similarity

In recent years, the railway emergency accompanied by the rapid development of China’s railway can not be ignored. The important enlightenment to us from the accident death toll and huge property loss is that we need to make a reasonable dispatch of rescue resources in time to provide decision support for the formulation of rescue plan. Case-Based Reasoning (CBR), is to select one or more most similar cases from the case base according to the current new events combined with certain calculation methods, and modify its solution according to the actual situation to get the solution of the new problem. Therefore, this paper proposes a confidence mixed similarity case matching method based on case-based reasoning system. First of all, the rough set theory is used to determine the weight of each attribute of the railway emergency. Secondly, the rescue cases of similar structures are obtained by using the calculation of structural similarity. Third, in order to better mine the similarity between the attributes of accident cases, the attributes are classified to calculate the attribute similarity. Fourth, the historical cases can be screened by combining structural similarity and attribute similarity. Finally, the decision tree gain algorithm is used to establish a decision tree to determine the confidence index of the rescue plan. At the end of this paper, we extract some cases from the case base and calculate the similarity and confidence between them and the target cases to obtain the rescue plan, which verifies that the results obtained by the analysis are logical and reliable.