数据驱动的大群体应急决策公众专家动态协同方法  

Data-driven dynamic collaborative method of public and experts in large-group emergency decision-making

作　　者：徐选华[1] 朱昱承 XU Xuanhua;ZHU Yucheng(School of Business,Central South University,Changsha 410083,China)

机构地区：[1]中南大学商学院,湖南长沙410083

出　　处：《系统工程与电子技术》2023年第12期3875-3886,共12页Systems Engineering and Electronics

基　　金：国家自然科学基金(71971217);中南大学研究生科研创新(自主探索类)项目(2022ZZTS0060);湖南省研究生科研创新项目(CX20220149)资助课题。

摘　　要：针对大群体应急决策中环境复杂多变及公众参与度较低的问题,提出一种数据分析驱动的大群体应急决策公众专家动态协同方法。首先,通过对社交媒体平台中的文本数据进行情感分析,分别从属性层和综合层评价大群体应急决策质量;其次,根据属性层决策质量提出了属性权重动态更新规则;然后,在社会网络环境下基于综合决策质量及专家偏好与群体偏好的距离动态更新专家间的信任关系,并使用改进PageRank算法计算专家权重;最后,通过某飞行器事故的案例应用和对比分析验证了所提方法的有效性和优越性。Aiming at the complex and changeable environment and the low public participation in large-group emergency decision-making,a data-driven dynamic collaborative method of public and experts in large-group emergency decision-making is proposed.Firstly,the sentiment analysis is carried out on text data in social media platforms to evaluate the quality of large-group emergency decision-making from attribute level and comprehensive level respectively.Secondly,according to the decision-making quality of attribute level,the dynamic update rules of attribute weight are proposed.Then,under the social network environment,the trust relationship between different experts is dynamically updated based on the comprehensive quality of decision-making and the distance between experts’preference and groups preference,and the expert weight is calculated by the PageRank algorithm.Finally,the effectiveness and superiority of the proposed method are verified through the case application and comparative analysis of an aircraft accident.

关 键 词：社交媒体数据 情感分析 大群体应急决策 动态协同 飞行器事故 

分 类 号：N945[自然科学总论—系统科学]