摘要
[目的/意义]为完善传统的网络舆情预警模型,考虑网络舆情预警指标之间交互性和决策者的风险偏好,合理集结专家的评价意见,提出一种基于直觉模糊Choquet积分的网络舆情预警模型。[过程/方法]采用直觉模糊数刻画舆情预警指标信息的模糊性与不确定性,利用模糊Choquet积分算子对指标属性值进行综合;定义了考虑专家舆情风险偏好的直觉模糊得分函数,以计算舆情风险等级评价值;引入投影法求解专家权重,以合理地对专家们的评价信息进行集结。[结果/结论]通过算例验证了预警模型的合理性和有效性,并对风险偏好因子进行了灵敏度分析,结果显示该舆情预警模型能直观地展示网络舆情风险倾向,为舆情监管部门提供较好的决策支持。
[Purpose/Significance]To improve the traditional network public opinion early warning model,considering the interaction between the network public opinion early warning indicators and the risk preference of decision-makers,the evaluation opinions of experts are reasonably gathered.A network public opinion early warning model based on intuitionistic fuzzy Choquet integral is proposed.[Method/Process]Intuitionistic fuzzy number is used to describe the fuzziness and uncertainty of public opinion early warning index information,and fuzzy Choquet integral operator is used to synthesize the index attribute values;The intuitionistic fuzzy score function considering the risk preference of experts is defined to calculate the evaluation value of public opinion risk grade;the projection method is introduced to solve the expert weight to reasonably aggregate the evaluation information of experts.The rationality and effectiveness of the early warning model are verified by an example,and the sensitivity analysis of risk preference factors is carried out.[Result/Conclusion]The results show that the public opinion early warning model can intuitively display the risk tendency of network public opinion,and provide better decision support for public opinion supervision department.
作者
林玲
陈福集
谢加良
李凤
Lin Ling;Chen Fuji;Xie Jialiang;Li Feng(School of Economics and Management,Fuzhou University,Fuzhou 350108;College of Science,Jimei University,Xiamen 361021;Institute of Big Data Modeling and Intelligent Computing,Jimei University,Xiamen 361021)
出处
《情报杂志》
CSSCI
北大核心
2021年第10期52-58,共7页
Journal of Intelligence
基金
国家自然科学基金“基于知识网格面向网络舆情的政府决策知识供需匹配研究”(编号:71271056)
福建省软科学基金“福建省大数据技术与产业发展分析研究”(编号:B19085)。
关键词
网络舆情
预警模型
直觉模糊
CHOQUET积分
投影法
network public opinion
early warning model
intuitionistic fuzzy
Choquet integral
projection method