Research On Emergency Case Representation And Similarity Calculation Based On Knowledge Element

With the increasing number of emergencies under the complex natural and social background,providing support and reference for emergency management and decision-making process based on existing cases and case bases has gradually become an important way to deal with unconventional emergencies.Many emergency decisions are done by case-based reasoning.The success of case-based reasoning in emergency management relies on a mature emergency case base,which requires a reliable case representation form and a effective case similarity algorithm to ensure the implementation.Aiming to the field of emergency management,this paper explores the representation form the similarity calculation method of emergency cases,which are two important aspects in the construction and application of emergency case base.Furthermore,we discusses the support function of case similarity results for case base maintenance,providing theoretical and practical meanings to the construction and maintenance of emergency case base and to the application of case-based reasoning.In this paper,we analyze the content of emergency cases and use the common knowledge element model as the unified knowledge representation form to extract and depicted actual object knowledge involved in emergency cases.On the basis of the structured representation and storage of emergency cases,we propose a new calculation framework of emergency case similarity based on the content of emergency cases and knowledge element model.By hierarchical division of case content,the amount of calculation is reduced and the calculation efficiency will be improved.In the meanwhile,the calculation resultes of emergency case similarity provide support for following case base maintenance and useful case enrichness.The similarity results are used to determine whether a case is redundant or whether it provides new knowledge,which is helpful for keeping the case base reliable and contributing to the success of case-based reasoning.