首页|面向复杂舆论场景的信息内容风险诊断模型构建与仿真研究

面向复杂舆论场景的信息内容风险诊断模型构建与仿真研究

扫码查看
[研究目的]面向复杂舆论场景提取信息内容风险点,构建风险诊断概念模型,为现实工作提供方法参考和工具借鉴.[研究方法]综合SCL、JHA、SWOT及28 位风险专家调查列举的统计数据,挖掘复杂舆论场景中的五大信息内容风险点.借助Delphi Method确定风险诊断指标框架后,运用AHP-DEMATEL加权中心度Mi和初始权重W1i 并作归一化处理而形成诊断方程,结合风险分级标准构建风险诊断I-E-S模型,最后通过案例仿真检验模型性能.[研究结论]风险源和风险处置绩效对诊断指数的影响最大,受众情感反应与行为反应次之,传播力最小.错误内容占比、二次违法违规用户量、重大负面舆情量、线下失范行为量在20 个末项操作化指标中的作用明显.多样本仿真检验发现,I-E-S模型诊断结果与用户风险感知契合度超过82.2%,模型实用性和精准性可初步满足风控工作需求.
Diagnostic Model Construction and Simulation Study of Information Content Risks in Complex Public Opinion Scenarios
[Research purpose]The objective of this study is to extract information content risk points in complex public opinion scenari-os,construct a risk diagnostic model,and provide method references and tool guidance for practical work.[Research method]The statis-tical data obtained by using SCL,JHA,SWOT and a risk expert survey listing method were analyzed to identify five information content risk points in complex public opinion scenarios.After determining the framework of risk diagnostic indicators by the Delphi Method,the AHP-DEMATEL method was applied to weight the centralities Miand initial weights W1iand then standardize them,forming a diagnostic equation.Then,in conjunction with risk classification criteria,the risk diagnostic I-E-S model was constructed.Finally,the model's per-formance was tested through case studies.[Research conclusion]The results indicate that risk source and risk disposal performance have the greatest impact on the diagnostic indicator,followed by audience emotional response and behavioral response.The impact of dissemi-nation power is the smallest.Among the last 20 actionable indicators,the proportion of erroneous content,the number of secondary illegal and non-compliant users,the volume of significant negative public opinion,and the quantity of offline deviant behaviors have significant effects.Through multiple empirical tests,it was found that the risk diagnostic results of the I-E-S model align with users'risk perception with a compatibility rate exceeding 82.2%.The practicability and accuracy of this model can initially meet the needs of risk control.

public opinioncomplex public opinion scenariosinformation content riskrisk diagnosisdiagnostic modeldiagnostic indi-cators

寇杰、李明德、乔亨

展开 >

西安交通大学马克思主义学院 西安 710049

西安交通大学新闻与新媒体学院 西安 710049

陕西延安干部学院 延安 716000

舆论 复杂舆论场景 信息内容风险 风险诊断 诊断模型 诊断指标

国家社会科学基金传播内容认知国家重点实验室开放基金资助项目陕西省社会科学基金年度项目陕西省社会科学基金舆情专项委托项目2023年陕西省软科学研究计划一般项目

20XXW00820K022022M0022021ZXWT022023-CX-RKX-047

2024

情报杂志
陕西省科学技术信息研究所

情报杂志

CSTPCDCSSCICHSSCD北大核心
影响因子:1.502
ISSN:1002-1965
年,卷(期):2024.43(1)
  • 21