Research Of Public Health Emergency Decision Method Based On CBR

As the progress of science and technology and the soaring of economy, thecomplexity and uncertainty of human society has been greatly promoted, and thesehave provided fertile soils for the occurrence of emergency incidents. As one kindof the social emergency incidents with a high occurrence, the public healthemergency incident (PHEI) has great impact on peoples’ lives and health as well asthe economic developments and social stability, so the authority has attached agreat importance to the PHEI supervision and rapid response to the PHEI when aPHEI happens.The lack of systematic knowledge required, the incompleteness of informationneeded and the high pressure decision-maker faced are all the common challengesfor PHEI decision. The preplan has always been viewed as the most powerfulweapon for dealing with PHEI, but preplans are usually made by some specialists,the knowledge these specialists command and the extent they understand for thecause and mechanism of PHEI have great impact on quality of the preplan theymade, so the access to the PHEI preplans has became the weak-point of PHEIresponse. In this paper, the case-based reasoning (CBR) methodology will beintroduced for PHEI response. Case is also a tool for problem solving, we can use itas a replacement of preplan, but the main difference between case and preplan isthat preplan is made by human in advance while case are got from history, so thereis no obstacles for getting cases.The structure of this paper is organized based on the process of CBR whichalso known as5-R circulation. Firstly two methods of case representation arepresented for PHEI cases, the object oriented method and XML method, and thenthe principal component analysis is introduce for case feature reduction. Secondly,the author choose the KNN algorithm as case retrieval strategy and designed thesimilarity measure approach for different feature types as well as theweight-assessing approach for case feature. At last, a strategy for PHEI caseadaptation based on case learning combined with user intervention is presented, andat the end of this paper, there are the some details of analysis and design for thisPHEI CBR system.