Knowledge Base Model Research For Emergency Response

The frequent outbreaks of emergencies have caused huge losses and threats to human security and social development.The evolution of emergencies is so complex and unpredictable that the rapid response to emergencies is very difficult.At the same time,the problem of "lack of knowledge" is becoming more and more serious in the big data environment.Therefore,it is urgent to refine and describe the emergency decision of each stage of evolution to historical emergencies,and to build emergency knowledge database which include the related concepts of emergencies evolution,the relationship between the related concepts,and the valuable knowledge.The knowledge base provides the knowledge service for accurately capturing the evolution of unexpected events,and rapidly making emergency decisions.The knowledge base for emergent management is a knowledge cluster to solve the related problems in the field of emergencies.The knowledge base system is mainly composed of case knowledge base,decision knowledge base and rule knowledge base and so on.The classes and grades of emergencies are two important attributes of emergencies.In the process of emergency management,assessment of the classes and grades of emergencies timely can effectively take targeted emergency measures.At the same time,the classification system of emergencies is an important basis for constructing the framework structure of the case database of emergencies.Therefore,the classification and grade model of emergencies that are used to obtain the classification and level knowledge of emergent cases effectively are important parts of the emergency knowledge base model.Based on the theory of knowledge organization and granular computing,this paper focus on building emergency case knowledge base,emergency classification model,emergency grade model,and the emergency rule knowledge base.Specific research work is as follows:(1)Building emergency case knowledge base based on the classification system of emergencies and the rule of emergency evolution.In this paper,based on the idea of multi-level and multi-view of granular computing,the external structure of emergency case database is constructed from two perspectives of vertical and horizontal.From the vertical perspective,the emergency will be organized according to the hierarchical structure of the emergency classification system,and from the horizontal perspective,the emergencies are linked into a whole according to the evolution of emergencies.To design the internal mechanism through the scene and strategy of emergencies,each emergency case is divided into scenes,at the same time,emergency strategies corresponding to the scenes composed of scenarios-strategies.From the internal point of view,the whole emergency case knowledge base is composed of scenario database and strategy database.Emergency case knowledge base constructed in this paper,all kinds of unexpected events will be organized into one,and you can see the incident in full view,which is conductive to evolution extension analysis for emergency,to predict in advance,process in amid,administrate in end.(2)Building an emergency classification model based on the emergent classification feature dictionary.This paper firstly constructs the emergency classification feature dictionary,on the basis of the establishment of emergency text feature matrix representation.The classifier is obtained to realize the automatic classification of the target emergencies based on the machine learning algorithm.The emergent classification model based on the emergent classification dictionary of emergencies,realizes the automatic identification of the cases when the knowledge base of the incident case is constructed.At the same time,the latest information is tracked by identifying the event text.(3)Building an emergency grade model based on the clound model.By analyzing the characteristics of the traditional grade rule,we point out its shortcomings,divide the emergencies into different grades according to the factors such as the degree of the accident,the influence scope and the ability of local control,establish the specific grading index system,and define the level of emergency response according to the principle of grading.In this paper,we introduce the cloud model theory into the emergencies grade,and propose a new grade model of emergencies to solve the uncertainties such as fuzziness and randomness in the grade process.This method is suitable for the early warning,the event processing and the post evaluation,which provides a scientific and effective method for the classification of emergency management.(4)Building an emergency rule knowledge base based on multi-level rough set.The construction of an emergency rule knowledge base includes the acquisition,storage and maintenance of the emergency knowledge.Among them,how to dig out valuable knowledge from a large number of unexpected events,and assist the emergency managers to make effective decisions,is the difficulty of this paper.In this paper,the rough set theory is applied to the acquisition of knowledge,and a new method is proposed to obtain the multi-level knowledge rules.First of all,the analysis of emergency incidents attributes,attribute value types are divided into character,language,numerical value of three types,according to the three types of methods are proposed to construct the emergency attribute hierarchical concept tree.Then,the event data is transformed into the form of multidimensional data cube suitable for data mining.Finally,based on the representation of multidimensional data model,the top-down method is used,then,reduces the data processing,and the knowledge rules are generated by using rough set theory.At the same time,this paper studies the storage and maintenance of knowledge,and designs the data structure and maintenance method.In this paper,the rough set and cloud model are used to study emergency knowledge base model,which has certain theoretical significance and potential practical value.However,these studies are only small parts in the research of emergent intelligence system,mostly stay in the theoretical stage of the discussion,in the future work will be further in-depth study.