An Application of CBR in Emergency Management ——Based on Text Mining in Chinese
Emergency management involves a large amount of knowledge from different areas. The reuse of the information and knowledge in historical cases could support the emergency management to a great extent; hence the CBR (case-based reasoning) has a nice scenario with the application in emergency management. This paper proposed a procedure based on test mining to accomplish a general application of CBR. Firstly, the test-based files of disaster cases are transformed into VSM (vector space model) through a series of preprocessing, such as word segmentation, stop-word removal, extraction of keyword, etc. Secondly, the emergency cases, in the form of VSM with labels and values, are compared in similarity as a classification task. At last, based on the fact that the cases with the highest similarity should have similar solutions, the emergency management dealing with a new disaster can consult the experience of historical cases retrieved through previous steps. The testing has been done using several cases to illustrate the feasibility and efficiency of this procedure.
Huang Chao Huang Quanyi Shen Shifei
Department of Engineering Physics Institute of Public Safety Research Tsinghua University Beijing, China
国际会议
北京
英文
686-688
2011-09-23（万方平台首次上网日期，不代表论文的发表时间）