摘要
针对特大突发事件应急决策中大群体专家存在偏好信息不完全的问题,提出了一种新的不完全偏好信息大群体应急决策方法.首先,利用TF-IDF(term frequency-inverse document frequency)算法对特大突发事件相关的微博大数据文本流进行关键词提取,获取事件属性及其权重;其次,根据专家给出的偏好信息计算专家的犹豫度,进而获得专家的权重;再次,根据不完全偏好信息矩阵进行属性关联测度和方案接近度测度,提出了基于属性关联和方案接近度的新的补值模型,获得完全偏好信息矩阵;然后,结合专家权重和属性权重进行信息集结和方案择优;最后,通过江西洪涝灾害事件验证所提方法的可行性和有效性.
In this study,we propose a novelemergency decision-making method for large groups with incomplete preferential in formationto solve the problem associated with the availability of incomplete information fora large group of experts during emergency decision making.First,we use the term frequency-inverse document frequency(TF-IDF)algorithm for extracting keywords from microblog texts related to anemergency to obtain attributes related to the event and theirweights.Second,we evaluate the hesitation of the experts according to the preferential information provided by the experts to obtain the expert weights.Third,we propose a new complement model to obtain complete preferential information matricesby measuring the correlation between the attributes and the proximity between alternatives based on the incomplete preferential information matrices.Fourth,we aggregate information and select alternatives by combining the attributes’weights and experts’weights.Finally,the feasibility and effectiveness of the proposed method are verified based on theflood disaster events in Jiangxi.
作者
徐选华
余艳粉
XU Xuanhua;YU Yanfen(School of Business,Central South University,Changsha 410083,China)
出处
《信息与控制》
CSCD
北大核心
2019年第6期678-686,693,共10页
Information and Control
基金
国家自然科学基金资助项目(71671189,71971217)
国家自然科学基金重点资助项目(71790615).
关键词
微博大数据
犹豫不完全偏好
补值方法
属性关联
大群体应急决策
micro-blog big data
hesitant incomplete preference
complement method
attribute association
largegroup emergency decision making
作者简介
徐选华(1962-),男,教授,博士生导师,研究领域为大数据决策理论与方法,信息系统与决策支持系统,应急管理与决策,风险分析与管理,复杂工程决策方法等;通信作者:余艳粉(1993-),女,硕士生,研究领域为大数据决策理论与方法,应急管理与决策,风险分析与管理,2733983727@qq.com
