The idea of industry–university research cooperation has been put forward in the early stage of China’s science and technology system reform. The main purpose is to solve the problem of the “two skins” of science and technology and economy. Many countries in the world regard the strengthening of industry–university research cooperation as important work. At the same time, both at home and abroad, researchers have paid extensive attention to the policy of industry–university research cooperation. In this paper, we first conduct knowledge map analysis of the domestic research situation, so as to obtain some of the directions and themes of the research on the policy of university research cooperation in China.
3.2. Authors of Policy Research on Industry–University Research Cooperation
Through the analysis of 927 documents from the literature by NoteExpress, it is concluded that the total number of authors studying the industry–university research cooperation policy and publishing via Peking University, EI, or CSSCI from 2011 to 2021 is 1606, and the average number of authors of each article is 1.73.
Table 1 clearly shows a list of authors with a high number of papers in the field of industry–university research cooperation policy research (six papers and above). The author with the largest number of papers is Zhu Guilong, with a number of 31, followed by Fan Xia, with 13 articles.
From the relevant data collected and extracted from the top 10 authors,
Figure 4 clearly shows the publishing frequency of each author from 2011 to 2021. It is found that Zhu Guilong is the most active in terms of publishing volume, with an output almost every year, followed by Fan Xia, Cao Xia, and Zhang Yi. Zhu Guilong has the longest research time and Yang Guoliang has the shortest research time.
In order to further understand author cooperation, we ran the CiteSpace software and set the node type as the author to obtain the author’s collinear network knowledge map with the number of nodes,
n = 358, the number of connections,
e = 159, and a density of 0.0025 (see
Figure 4 and
Figure 5). As can be seen from the figure, the scholar with the largest number of nodes is Zhu Guilong, followed by Fan Xia, Zhang Yi, Chen Kaihua, Liu Guowei, Cao Xia, and Chen Guanghua. Due to the mutual exchange and cooperation of various scholars, several sub-author network structures are formed in the atlas, notably a network structure led by Zhu Guilong, Fan Xia, and Liu Guowei. The figure shows that several high-yield authors have low cooperation, which is mainly reflected in the academic exchanges and contact between Zhu Guilong, Zhang Yi, and Chen Kaihua, and less cooperation is seen with other scholars with large amount of papers (such as Liu Guowei, Chen Guanghua, etc.). There are many scholars in separate nodes, such as Wu Jie, Huang Jinsong, and Yuan Yijun. On the surface, it appears that Chinese scholars do not have a strong sense of communication and cooperation in the research of industry–university research cooperation policy, and the academic exchange and cooperation of scholars needs to be strengthened.
3.4. Key Issues of the Policy Research of Industry–University Research Cooperation
Research hotspots reflect the research focus and direction of a research field, which is of great significance for an in-depth understanding and analysis of the research content in this field. Keywords are the core of a document. The high frequency of keywords in a certain field reflects the research hotspot in this field. Keyword cluster analysis is based on keyword co-occurrence analysis, which simplifies the keyword co-occurrence network relationship into a relatively small number of clusters through the method of cluster statistics. This paper analyzes the research hotspots of industry–university research cooperation policy by keyword cluster analysis, and explores the research hotspots of industry–university research cooperation policy.
Using CiteSpace software to extract the keywords of literature data, we set the node type as the keyword, and set the three groups of C, CC, and CCV to have a threshold of 1, 1, and 20; 1, 1, and 20; and 3, 3, and 20, respectively. Other parameter values are set by default. Based on the keyword knowledge network map, the LLR algorithm is selected to obtain the keyword clustering network map as shown in
Figure 7, in which the number of nodes is 963, the number of connections is 1419, and the network density is 0.0031. The figure shows 18 clusters of the phrases “industry university research”, “university”, “mode”, “collaborative innovation”, “cooperative education”, “higher vocational colleges”, “technological innovation”, “innovation network”, “cooperation mode”, “innovation”, “countermeasures”, “benefit distribution”, “knowledge transfer”, “cooperative innovation”, “performance evaluation”, “performance evaluation”, “innovation performance”, “scientific and technological innovation”, and “supporting policies”. This reflects the research hotspots in the field of the policy of domestic university research cooperation.
Based on the knowledge map of keyword clustering, the log likelihood rate (one of the clustering tag word extraction algorithms) is obtained in “cluster explorer”, and the keyword co-occurrence network clustering table is obtained (see
Table 3).
By analyzing the keywords in each cluster, it is found that the research contents of each cluster intersect with each other. Therefore, it can be seen that the domestic research on the policy of industry–university research cooperation can be summarized into three thematic areas: “industry–university research cooperation”, “innovation”, and “universities”. Three important related clusters of “industry university research”, “universities”, and “innovation” are taken as an example (
Figure 8). In the literature, the industry university research cluster first appeared in Tian Huajie’s [
1]
Effect and Problem Thinking of Industry–University Research Cooperation Mode in 2011. Since 2011, the number of results in this cluster has increased, and the number of studies containing the keyword “industry university research” has increased rapidly. Moreover, over time, the attention paid to this cluster has been maintained at a high level. The keyword emergence diagram (
Figure 9) also shows that the keyword “industry university research” achieved important results from 2019 to 2021, consistent with the recovery trend of the number of documents issued, as mentioned above. For example, Chen Yufen [
2] used the social network analysis method to quantitatively analyze the characteristics of an industry–university research cooperative innovation network and its impact on innovation performance from two aspects: different stages and different industries. Guided by the vision of science and technology development in 2035, Ren Xingxin [
3] discussed a feasible path to realize the effective combination of industry, university, and research by 2035, by dissecting Bi Yusui, Professor of Shandong University of Technology, and his team’s research and development of “chlorine free fluoropolyurethane new chemical foaming agent”. In the literature, a cluster of the keyword “innovation” first appeared in Jiang Zhengguo’s [
4]
Research on the Innovation of Teaching Operation Mechanism under the Background of Industry University Research Cooperative Education in Newly Built Local Undergraduate Colleges in 2011. In the same year, the clustering results began to increase and the research intensity also gradually increased. Results in the literature on the clustering of the keyword “universities” first appeared in Huang Yanfei’s
An Analysis of the Ways to Form the Advantages of Independent Innovation of University Official Industry Research University Cooperation—Taking the Universities in the Yangtze River Delta as an Example in 2011 [
5]. The clustering results began to increase from 2011, but on the whole, the intensity of the three clusters decreased with the passage of time. At the same time, it can be seen from the emergence chart that the emerging keyword from 2015 to 2021 is “innovation performance”; from 2018 to 2021, the emergent keyword is “evolutionary game”; and from 2019 to 2021, the term “industry university research” emerged, and the emergence rate of these three terms has continued to this day, which shows that these three keywords represent the main development trend of domestic university research cooperation policy research.
With the continuous development of the economy alongside science and technology, innovation performance, evolutionary game, and industry university research will continue to become hotspots of future research.
In the era of the knowledge economy, open innovation has gradually become an important way for enterprises to obtain the key resources needed for technological innovation. Under the open innovation paradigm, the organizational boundary is no longer closed, and the innovation resources inside and outside the organization can complement each other. R&D activities are no longer limited to within the organization. Enterprises can reduce the uncertainty of technological innovation in the market by carrying out cross-organizational cooperation. From the perspective of forms of knowledge, the knowledge of enterprises, university research institutions, and scholars is heterogeneous and complementary. Therefore, enterprises will spontaneously carry out R&D cooperation with universities and research institutions according to their own interests and demands. However, with the increase in the degree of open innovation, enterprises will also face many problems, such as excessive searching, high transaction costs, difficult cooperative management, insufficient knowledge absorption capacity, and leakage of technical knowledge. As a result, open innovation may have a negative impact on enterprise innovation performance. Therefore, in the process of technological innovation, the relationship between the openness of cross-organizational cooperation and innovation performance is complex [
6]. In the future, the research on innovation performance will become increasingly popular. In the process of inter-organizational cooperation, the impact of technological factors on industry–university research cooperation needs to be fully considered. Therefore, the game payment matrix of industry–university research cooperation innovation based on technological maturity and technological innovation is constructed. The equilibrium solution of industry–university research cooperation innovation is obtained by the evolutionary game method. Through numerical and case analysis, the impact of different changes in technological maturity and technological innovation on the willingness of industry–university research cooperation innovation is explored. Judging the technical reasons affecting the willingness of industry–university research cooperation innovation, evolutionary game theory plays an important role in the transformation of industry–university research achievements and the improvement of enterprise innovation performance [
7]. Therefore, it will also exist in the future as a research hotspot.
In addition, we used the CiteSpace software to extract the keywords from the literature and create the keyword time zone map, in order to reflect the research content of this topic changing with time, and then observed the research trend in a certain time period. We set the top n of the selection criteria to 10 and the thresholds of C, CC, and CCV to 3, 3, and 20; 3, 3, and 20; and 3, 3, and 20, respectively. We obtained a keyword time zone map of industry–university research cooperation policy research (
Figure 10).
Based on the above research, it can be seen from the keyword time zone map that the focus and amount of attention differ by period. Observing the keyword distribution in the past two years, it can be found that the keywords “cooperation breadth”, “cooperation depth”, “information disclosure”, “collusion”, “rent-seeking”, “opportunism”, and “innovation quality” appear in the time zone map in 2020; the keywords in 2021 include “engineering education”, “institutional change”, and “policy change”. Therefore, it can be concluded that the future research trend of industry–university research cooperation policy research can be roughly divided into three parts.
The first part is “the comprehensive deepening of industry–university research cooperation”. Liu Feiran, Hu Lijun, Fan Xiaoqun [
5], and others analyzed and tested the impact of industry–university research cooperation on enterprise innovation quality based on the data of nearly one million patent applications of Chinese listed companies from 2005 to 2008, and concluded that the breadth of enterprise participation in industry–university research cooperation has an inverted “U”-shaped impact on innovation quality, and the depth of cooperation has a continuous positive impact on innovation quality. Gao Xia, Cao Jieqiong, Bao Lingling [
8], and others used the jointly authorized patent data of China’s ICT industry from 1999 to 2015, and extracted a large sample of panel data from 14,596 enterprises. Using the negative binomial regression model, they concluded, in terms of the two dimensions of cooperation breadth and depth, that universities play an important role in improving the innovation performance of China’s ICT enterprises, followed by scientific research institutions. This proves the importance and necessity of research in the direction of cooperation breadth and depth when studying industry–university research cooperation policy.
The second part is “industry–university research cooperation and innovation performance”. Qiu Yangdong [
9] discussed the innovation incentive effect of the combination of industry, university, and research from the three dimensions of state ownership, market orientation, and intellectual property protection, and found that the cooperation of industry, universities, and research has a significant effect on the quantity and quality of enterprise innovation. It was proposed that, in order to improve the innovation performance and ensure innovation quality, we should actively promote market-oriented reform and the construction of an intellectual property protection system, so as to effectively improve the innovation performance of industry, university, and research cooperation. It can be seen that innovation performance will become the direct purpose of industry–university research cooperation policy, which will maintain an important position in the field of industry–university research cooperation policy research.
The third part is “the change of industry–university research cooperation policy”. Peng Lin and Brent Jesiek [
10] studied the changes in cooperative education policy in the United States and obtained the logic of the changes to the cooperative education policy structure and the historical changes to cooperative education policy in the United States. In terms of background and policy variables, the complexity of society is becoming increasingly prominent, and the top-level design, policy structure, informal institutional factors, and market forces have put forward new demands for cooperative education policy. In addition, industry–university research cooperation policy is at the key node of recovery, so under the theme of industry–university research cooperation policy, the trend of policy change and institutional change is beyond a doubt, and it will also become a research hotspot of industry–university research cooperation policy in the future.