[Research purpose]In recent years,society has paid high attention to food safety issues,and the risk of online public opinion caused by food safety incidents has increased,which has a negative impact on social stability.Therefore,by constructing a risk assessment model for the dissemination of public opinion information on food safety networks,this study analyzes the key characteristics that affect risk types,and helps relevant departments timely identify and evaluate public opinion risks,improving risk warning capabilities.[Re-search method]Based on speech act theory,the content of public opinion information is taken as the research object,constructs a feature set of risk influencing factors for public opinion information dissemination from the speaker's speech information,constructs a multi label annotation system for risk types from the speaker's speech information,and selects key features using a sparse multi label learning model to achieve research on the factors influencing the risk of online public opinion information dissemination.[Research conclusion]The experi-mental results indicate that the model can effectively extract key features that affect the types of information dissemination risks,providing decision-making basis for risk monitoring and early warning of food safety network public opinion.
关键词
网络舆情/舆情传播/食品安全/风险评估/多标签学习/特征选择/言语行为理论
Key words
online public opinion/public opinion dissemination/food safety/risk assessment/multi label learning/feature selection/speech act theory
维普
万方数据