The Study Of Public Emergency Event Case Retrieval Method Based On Semantic Similarity

Recent years, with the occurring of masses of emergencies, emergency management has become a new field of academic research. By means of information science, many scholars began to study the issues related to emergency management by case-based reasoning(CBR) method. Academic institutions and government departments have begun to build cases database of emergency events. Based on existing cases and knowledge to provide decision support for crisis management process is necessary means to improve crisis management capabilities. The similarity computation based on semantic similarity case has a high value to raise the case clustering and retrieval accuracy. However, in CBR of public crisis event, there is little research on semantic similarity of case-based reasoning and retrieval model.In this paper, in-depth study of the relevant literature analysis of international cases Reasoning, Describing the basic cycle, hierarchy model, as well as method of the case-based reasoning. At the same time, analysis the existing method of text processing and text feature extraction on semantic similarity calculation. And finally, through the analysis of the structure of public crises case base case, Learn from existing text processing and the extraction and semantic similarity calculation method, Complete the public crises Case semantic similarity calculation model. The verification of results is carried out through the use of case base in two different categories of the four cases, the results show that the similarity calculation model has validity, Provides a reference for public crises of case-based reasoning.