The Analysis On The Policy Of Industry-University-Research Collaborative Innovation Based On Text Mining

With the development of the global economy,the transformation of scientific and technological achievements into productivity has attracted more and more attention from various countries.In this context,research on collaborative innovation of industry,university and research has become a hot spot.For my country’s local governments at all levels and the main bodies of industry-university-research institutes,to better promote the process of industryuniversity-research collaboration innovation requires a deep understanding of policies.The implementation of the relevant policies of industry-university-research collaborative innovation can give full play to the respective advantages of schools,scientific research units,and enterprises,and promote the high-quality development of the three.Research on productionuniversity-research policies can provide reference for the formulation and adjustment of supporting policies for local governments at all levels,improve the efficiency of policy implementation,and is of great significance for advancing my country’s supply-side structural reform.Based on the above analysis,this article mainly does the following work:First of all,the acquisition of relevant policy documents for production,education and research.The research object of this article is the industry-university-research-related policy text after my country promulgated and implemented the "National Medium and Long-term Science and Technology Development Plan Outline(2006-2020)" in 2006.The data sources are mainly from the Peking University Legal Information Network and government portals.After collecting the data,do exploratory data analysis and preprocessing of the text to prepare for analysis.Second,excavate the policy text.(1)Count the effective word frequency of the policy text;(2)Generate a high-frequency word co-occurrence matrix based on the result of the word frequency statistics;(3)Use SPSS software to cluster the policy text;(4)Establish the LDA topic model,Extract the policy text subject.The results show that the policy text mainly contains three themes,namely: enterprise,university scientific research,and talent.Finally,from the perspective of quantitative evaluation of policy texts,research on industry-university-research policies is carried out.The specific method is: combined with the results of LDA theme mining,select the most representative 12 policies from the corpus,from publishing organization,policy object,policy nature,incentive measures,policy function,effectiveness level,field of action,policy means,policy evaluation The document quotes ten dimensions for quantitative research,establishes a PMC index model,calculates the PMC index score based on policy content and text mining methods,draws a PMC surface diagram,and finally gives conclusions and suggestions.The results show that the connotation of the industry-university-research policies issued by my country is gradually being improved and gradually standardized at the regulatory level.Among them,in terms of policy objects and incentive measures,the text analysis results reflect that the current policies are effective in promoting governments or institutions at all levels to increase the impact on talents.It is of great significance to attach importance to intensity and promote the introduction of local innovation support policies.However,there are still shortcomings in the application of policy tools,such as public services and financial support.When local governments formulate policies,they must adapt to time and local conditions to optimize the effectiveness of policies.In addition,the establishment of a scientific research performance evaluation system should be actively promoted.The policy encourages cooperation between university scientific research talents and enterprises,strengthens the motivation of innovation,and mobilizes the enthusiasm of various innovation entities.