基于LDA模型的公众反馈意见采纳研究——共享单车政策修订与数据挖掘的对比分析
|
摘要
【目的/意义】通过数据挖掘、文本分析法整合公众分散的反馈意见,对比研究出台政策,对政府部门吸纳公众意见,制定科学合理的公共政策具有重要意义。【方法/过程】基于大数据技术深度挖掘公众对共享单车政策征求意见稿的反馈内容,采用LDA模型从公众反馈意见中发现并提取低维主题集合,汇总、组织、分析大规模文本信息,以主题领域和隶属概率值两维度,对比公众意见与政策修订的差异。【结果/结论】从制度体系、政府治理、企业经营、产品完善、用户规范五个分类主题及其概率值,评估政策制定中的公众参与度以及政府对公众反馈意见的采纳效果,以期为政府出台政策的科学性提供理论依据和决策建议。
【Purpose/significance】It is important to integrate public dispersed feedback through data mining and text analysis, and take a comparative study on introduce policies, which is significant for government departments to absorb public opinions and formulate scientific and reasonable public policies.【Method/process】The research bases on big data technology to take a deep mining of public feedback content of bicycle sharing draft policy, using the LDA model to find and extract the low dimensional theme, collection, organization, analysis of large scale text information of the public feedback, finding the difference by contrasting public opinion and policy revision through the two dimension in subject areas and the membership probability value.【Result/conclusion】From five classification theme and their probability such as institutional system, government management, enterprise management, the product improvement and users' specification to evaluation policy the participation and the public feedback adoption effect of government, in order to provide theoretical basis and policy recommendations for the government to carry out scientific policy.
【Purpose/significance】It is important to integrate public dispersed feedback through data mining and text analysis, and take a comparative study on introduce policies, which is significant for government departments to absorb public opinions and formulate scientific and reasonable public policies.【Method/process】The research bases on big data technology to take a deep mining of public feedback content of bicycle sharing draft policy, using the LDA model to find and extract the low dimensional theme, collection, organization, analysis of large scale text information of the public feedback, finding the difference by contrasting public opinion and policy revision through the two dimension in subject areas and the membership probability value.【Result/conclusion】From five classification theme and their probability such as institutional system, government management, enterprise management, the product improvement and users' specification to evaluation policy the participation and the public feedback adoption effect of government, in order to provide theoretical basis and policy recommendations for the government to carry out scientific policy.
引文
1中商产业研究院.2017全球与中国市场共享单车深度研究报告[R].2017-03-22.
2 张洪忠.大众传播学的议程设置理论与框架理论关系探讨[J].西南民族学院学报(哲学社会科学版),2001,(10):88-91.
3 COBB R W, ELDER C D. Participation in American Politics:The Dynamics of Agenda-Building[M].Baltimore:John Hopkins University Press,1972.
4 魏淑艳,孙峰.“多源流理论”视阈下网络社会政策议程设置现代化——以出租车改革为例[J].公共管理学报,2016,13(2):1-13,152.
5 邓喆,孟庆国.自媒体的议程设置:公共政策形成的新路径[J].公共管理学报,2016,13(2):14-22,153.
6 ROBERTS M, WANTA W, DZWO T H. Agenda Setting and Issue Salience Online[J].Communication Research,2002,29(4):452-465.
7 汪波.大数据、民意形态变迁与数字协商民主[J].浙江社会科学,2015,(11):41-47,157.
8 胡税根,单立栋,徐靖芮.基于大数据的智慧公共决策特征研究[J].浙江大学学报(人文社会科学版),2015,45(3):5-15.
9 魏航,王建冬,童楠楠.基于大数据的公共政策评估研究:回顾与建议[J].电子政务,2016,(1):11-17.
10 曹光煜,傅昌波.新互联网对公共决策的影响及对策建议[J].中国行政管理,2015,(12):109-112.
11 张楠.公共衍生大数据分析与政府决策过程重构:理论演进与研究展望[J].中国行政管理,2015,(10):19-24.
12 赵妍妍,秦兵,刘挺.文本情感分析[J].软件学报,2010,21(8):1834-1848.
13 邱明涛,马静,张磊,姚兆旭.基于可扩展LDA模型的微博话题特征抽取研究[J].情报科学,2017,35(4):22-26,31.
2 张洪忠.大众传播学的议程设置理论与框架理论关系探讨[J].西南民族学院学报(哲学社会科学版),2001,(10):88-91.
3 COBB R W, ELDER C D. Participation in American Politics:The Dynamics of Agenda-Building[M].Baltimore:John Hopkins University Press,1972.
4 魏淑艳,孙峰.“多源流理论”视阈下网络社会政策议程设置现代化——以出租车改革为例[J].公共管理学报,2016,13(2):1-13,152.
5 邓喆,孟庆国.自媒体的议程设置:公共政策形成的新路径[J].公共管理学报,2016,13(2):14-22,153.
6 ROBERTS M, WANTA W, DZWO T H. Agenda Setting and Issue Salience Online[J].Communication Research,2002,29(4):452-465.
7 汪波.大数据、民意形态变迁与数字协商民主[J].浙江社会科学,2015,(11):41-47,157.
8 胡税根,单立栋,徐靖芮.基于大数据的智慧公共决策特征研究[J].浙江大学学报(人文社会科学版),2015,45(3):5-15.
9 魏航,王建冬,童楠楠.基于大数据的公共政策评估研究:回顾与建议[J].电子政务,2016,(1):11-17.
10 曹光煜,傅昌波.新互联网对公共决策的影响及对策建议[J].中国行政管理,2015,(12):109-112.
11 张楠.公共衍生大数据分析与政府决策过程重构:理论演进与研究展望[J].中国行政管理,2015,(10):19-24.
12 赵妍妍,秦兵,刘挺.文本情感分析[J].软件学报,2010,21(8):1834-1848.
13 邱明涛,马静,张磊,姚兆旭.基于可扩展LDA模型的微博话题特征抽取研究[J].情报科学,2017,35(4):22-26,31.