中国安全科学学报 ›› 2018, Vol. 28 ›› Issue (8): 186-191.doi: 10.16265/j.cnki.issn1003-3033.2018.08.031

• 公共安全 • 上一篇    

不完备信息条件下的应急案例检索方法

郑晶1,2, 副教授, 王应明2 教授   

  1. 1 福建江夏学院 电子信息科学学院,福州 福建 350108;
    2 福州大学 决策科学研究所,福州 福建 350116
  • 收稿日期:2018-05-11 修回日期:2018-06-30 出版日期:2018-08-28 发布日期:2020-11-25
  • 作者简介:郑晶(1980—),女,福建福州人,博士,副教授,主要从事应急决策、信息管理等方面的研究。E-mail:zhengjing80@qq.com。
  • 基金资助:
    国家自然科学基金资助 (61773123);福建省自然科学基金资助(2017J01513)。

Method for retrieving emergency cases under incomplete information condition

ZHENG Jing1,2 , WANG Yingming2   

  1. 1 College of Electronics and Information Science,Fujian Jiangxia University,Fuzhou Fujian 350108,China;
    2 Decision Sciences Insititue,Fuzhou University,Fuzhou Fujian 350116,China
  • Received:2018-05-11 Revised:2018-06-30 Online:2018-08-28 Published:2020-11-25

摘要: 突发事件中存在不完备信息,对其表示不当将负面影响应急响应的效果。为提高案例检索的有效性,引入置信度分布表示突发事件的信息,提出一种不完备信息条件下的应急案例检索方法。该方法先应用置信度分布表示不完备信息,再构建最优化模型得到分布等级上的得分来确定不完备信息的属性值;计算案例相似度和方案实施效果的效用值;在此基础上,集结案例相似度和效用值得到案例的综合效用值,并依据综合效用值检索出最有效的历史案例;最后,对该方法进行实例应用。结果表明:用该方法能够检索存在不完备信息的案例,而且检索结果较准确。

关键词: 案例推理(CBR), 信息不完备, 数据包络分析, 置信度分布区间, 应急案例库

Abstract: The emergencies contain incomplete information,if it is not expressed appropriately,the effect of emergency response will be affected adversely.This paper introduces the belief degree distribution to represent the information on emergencies,and presents a method for retrieving emergency cases under the incomplete information condition.Firstly,the proposed method applies the belief degree distribution to represent incomplete information,and builds an optimization model to obtain the score of the distribution grade for determining the attribute value of incomplete information.Then,the case similarities and the utility values of the alternative implementation effect are calculated.Furthermore,the comprehensive utilities of the historical cases are obtained by aggregating case similarities and utilities,and the most effective historical case is retrieved according to the comprehensive utilities.Finally,the method was used for analyzing an example.The results show that the method can be used to deal with the emergency cases containing incomplete information,and the results of the retrieval are more accurate.

Key words: case-based reasoning(CBR), incomplete information, data envelopment analysis, interval belief degree distribution, emergency case base

中图分类号: