Multiple Network Public Opinion Event Risk Analysis Model for Probabilistic Language Preference
[Research purpose]To improve the network public opinion risk assessment model,considering the complexity of dynamic cross-evolution of multiple network public opinion events and expert consensus reaching under the environment of trust network group de-cision-making,a probabilistic language preference oriented multiple network public opinion event risk assessment model is constructed.[Research method]A probabilistic language term set is used to describe the fuzziness and uncertainty of public opinion risk indicators,and a PLTSs distance calculation formula considering the degree of hesitation is defined to measure the expert consensus index.A consen-sus model based on the dual feedback mechanism of expert opinion correction and weight reset based on trust-reward and punishment is proposed to form reasonable collective evaluation opinions.The regret theory is integrated into the group decision making method,and the comprehensive risk perceived utility value of public opinion events is obtained.[Research conclusion]The applicability and stability of the model are verified by a numerical example.The results show that the public opinion risk model can intuitively display the risk ranking of multiple network public opinion events,and can provide better decision support for public opinion supervision departments.
multiple network public opinion eventsrisk assessmentpublic opinion governanceprobabilistic linguistic term setsdis-tance measuretrust-reward and punishment dual feedback mechanismregret theory
万方数据
维普
