Smart government is an important part of the smart world. The use of big data analysis technology can effectively improve the government’s ability of fine management. Taking China’s bike-sharing industry as the research object, we study the relationship between public-derived big data and industrial policy. First, a feature-enhanced short text clustering method is proposed to perform topic clustering on publicly derived big data. Second, keyword extraction based on word frequency is used to quantify the text of industrial policy. Finally, time is taken as the main line to analyze the co-occurrence of clustering topics and keywords. The results show that (1) the feature enhancement method we proposed can effectively improve the clustering effect. (2) There is a great correlation between the industrial policy and the information mined by Weibo, but there is an obvious lag. Rational use of public-derived big data will effectively help the industrial policy to be released in a better and faster way.
A Study on the Relationship between Public Derivative Big Data and Industrial Policymaking: Taking Bike Sharing as an Example