面向公共政策的网络媒体内容文本分析应用框架与实证--以新能源汽车政策为例  

Application Framework and Empirical Study of Web Media Content Analysis on Public Policy--New Energy Vehicle Policy

作　　者：赵菲菲 王宇琪 孔庆超[2] 周庆山[1] ZHAO Fei-fei;WANG Yu-qi;KONG Qing-chao;ZHOU Qing-shan(Department of Information Management,Peking University,Beijing 100871,China;State Key Laboratory of Management and Control for Complex Systems,Institute of Automation,Chinese Academy of Sciences,Beijing 100190,China)

机构地区：[1]北京大学信息管理系,北京100871 [2]中国科学院自动化研究所复杂系统管理与控制国家重点实验室,北京100190

出　　处：《情报科学》2020年第4期140-146,共7页Information Science

摘　　要：【目的/意义】随着互联网的迅速发展,网络媒体成为反映社会舆论的主要载体。如何有效地从网络媒体获取公共政策相关的社情民意以引导公共政策的传播议程设置是政府职能部门所关注的重要问题之一。【方法/过程】本文基于网络媒体数据,应用数据挖掘、机器学习等数据分析技术,提出了一个面向公共政策的网络媒体内容文本分析框架。利用文本语义分析方法,从主题识别、情感分析等角度对网络主流媒体的公共政策传播议程设置与社交媒体网民舆论进行挖掘和对比,并以新能源汽车政策为例对该分析框架的有效性进行了验证。【结果/结论】通过实证发现当下网络媒体报道的议题与社交媒体上公众对有关公共政策的关注焦点之间存在较大偏差,就新能源汽车政策为例,公众对于其政策的关注偏向于衡量自身获利的多少,而网络媒体报道更多以描述政策传递信息为主。建议政府职能机构针对公共政策使用网络媒体进行传播时,可根据公众关注焦点话题进行议程的设置和调整,以增强公众对此政策的认可度。【Purpose/significance】With the rapid development of the Internet, the web media has become the primary channel for reflecting social public opinion. It has become an important issue for the government and related departments to effectively boost the policy related public opinion collection, and more importantly, to guide mainstream media agenda settings in policy information dissemination.【Method/process】Based on the data analysis technology like data mining, machine learning and et al, facing public policy,this paper develops a framework for analyzing web media content.Using textual semantic anaylsis method, the paper compares the differences between the public’concerns and the media agenda setting on an empirical study of the new energy vehicles policy from the perspectives of topic modeling, opinion mining and sentiment analysis.【Result/conclusion】The results show the bias exists between the web media agenda reports and the public’s concerns regarding the public policy. The public focus on the benefits they can achieve from the public policy, while web media reports describe the nature of policy-driven information. It is suggested for the government agencies to adjust the context of web media agenda reports according to the public’s concerns topics, in order to enhance the public’s acceptance toward a policy.

关 键 词：公共政策 网络媒体 舆论分析 媒体议程设置 

分 类 号：G254[文化科学—图书馆学]