Research on Interactive Content Analysis and Appeal Response of Government New Media:Taking the Message Boards of People's Daily Online in Central Megacities as an Example
[Research purpose]Online governance has become one of the main ways of interaction between government and the people in the new era.By fully exploring the characteristics of online political civilian interaction content,we can understand the public's public o-pinion demands and timely identify areas for improvement in the work of local government departments.[Research method]According to the framework theory,natural language processing technology is used to mine the language characteristics,content characteristics and behavior characteristics in the interactive content of leader message boards in the central megacities,and analyze the potential relationship between different characteristics.[Research conclusion]Research has shown that in terms of language characteristics,the proportion of weak message frames among the public is the highest,while the proportion of resistance frames is relatively low.The government prefers to use descriptive discourse and empathetic discourse,while the frequency of using rule-based discourse is relatively low.In terms of con-tent features,there is a correlation between the content features of various fields and the corresponding data of the urban statistical year-book,which can objectively reflect the problems in various fields.Clustering can effectively explore hot topics and assist government re-sponses through entities.In terms of behavioral characteristics,the proportion of public complaints/help seeking messages is related to the per capita Gross regional product.There is a potential correlation between the behavioral characteristics of government responses and the content characteristics of public messages,with the processing type of government actions being most influenced by the content characteris-tics of public messages,lacking attention to language and behavioral characteristics.
new media of government affairsinternet governancemessage boardpublic nessagesgoverment responsegoverment ac-tionsentiment tendency analysistopic clustering
万方数据
