Impact of Digital Transformation on Corporate Innovation--Empirical evidence from public companies

: In the context of comprehensive penetration and cross-border integration of information technology, digital transformation of enterprises has become an important way to accelerate enterprise innovation and lead economic innovation growth, injecting new vitality into enterprises. Based on the textual information of annual reports of A-share listed companies from 2011-2021, this paper constructs a measure of digital transformation and systematically examines the impact of digital transformation on corporate innovation output and its impact mechanism. It is found that there is a significant positive relationship between digital transformation and innovation output of enterprises, and the finding is robust. The mechanism test indicates that corporate digital transformation promotes corporate innovation by increasing innovation inputs and promoting corporate learning effects. Further study found that the promotion effect of corporate digital transformation on corporate innovation was more pronounced in large enterprises, high-tech enterprises, state-owned enterprises, enterprises whose executives have master's degrees, and enterprises in relatively close geographical proximity to the eastern region. This study not only enriches the empirical research on the consequences of digital transformation and the mechanism of enterprises using new dynamic energy to enhance transformation, but also has important implications for promoting the layout adjustment of national digital innovation-driven strategies.


Introduction and Review of Literature
In recent years, the digital economy has become the most active segment of China's economic development, playing a very important role in promoting consumption and stimulating innovation. From the domestic perspective, China is currently accelerating towards the digital era. According to the information of China Academy of Information and Communication Technology, the scale of China's digital economy has ranked second in the world for many years, the Internet penetration rate has increased to 73%, and the GDP share of digital economy has increased from 21.6% to 39.8%. From an international perspective, according to the 2021 Digital Economy Report released by UNCTAD, China and the United States have the strongest ability to harness data, with the two countries having the highest 5G adoption rates, 50% of the world's hyperscale data centers, 70% of the top AI researchers, 94% of AI startup funding, and 90% of the market capitalization of the world's large digital platforms. The global penetration of digitalization has become an important dependency to drive high-quality economic development, and China's economy has to change its development mindset in a timely manner in line with the times, and to switch from an investment-driven form of growth to an innovation-driven development model. As stated in the "Overall Layout Plan for the Construction of Digital China" released in 2023, digital infrastructure will become the main focus, to solidify digital infrastructure and data resource systems, strengthen digital technology innovation and digital security barriers, and optimize digital development. Enterprise innovation refers to the ability of enterprises to effectively integrate internal and external resources to produce intellectual property or new products adapted to the consumer market, and to achieve innovative returns. Currently, the impact of the digital economy on enterprise innovation has received wide academic attention, and the literature has conducted in-depth studies on enterprise innovation from two main lines: the first line explores the mediating role of enterprise innovation as a mediating variable of the digital economy on high-quality economic development [1]. The second main line explores the relationship between the digital economy and innovation capability [2,3,4] and analyzes the mechanisms at play. With the rapid development of emerging technologies, big data, cloud computing, artificial intelligence and 5G are constantly changing people's habits and consumption needs, in this situation, enterprises must quickly transform and upgrade through R&D innovation, and it is a very wise transformation strategy for enterprises to use digital transformation to grasp the first opportunity of economic growth and achieve business growth through R&D innovation. The core of digital transformation is to use digital technology to collect information, process data and guide decision-making. In terms of theory, there has been a lot of discussion on the connotation of digital transformation among scholars at home and abroad, and the existing studies mainly focus on three aspects: technology scope, transformation area and transformation effect. From the perspective of the technology sphere, digital transformation emphasizes the adaptability of information technology to organizational structure, and enhances the resilience of enterprises to digital impact by raising organizational information costs and inducing downward organizational empowerment [5]. The transformation domain perspective analyzes the adaptive adjustment of the internal management of the enterprise, and the application of digital technology in the operation of the enterprise can adjust the content of the internal functional activities [6], create a digital technology ecosystem, and create more added value for users. The transformation effect perspective argues that at the micro level, the market competition and high innovation incentives brought about by digital transformation will enhance innovation performance [7], and digital transformation leads to the integration of firms into global innovation networks thereby enhancing their innovation performance; at the macro level, digital technology enhances and improves the quality of life of people [8] and promotes industrial development [9]. In order to measure the development status of digital transformation, experts and scholars have constructed various evaluation indicators. Li Xiong and Xuelian Cai used the number of high-tech enterprises as a measure of digital application, which significantly reduced the complexity of indicator calculation, but considered fewer factors, resulting in an imprecise portrayal of the degree of digital transformation; Tao Zhao et al. constructed the digital transformation evaluation system from two dimensions of digital inclusive finance and Internet development level, and selected four indicators of Internet penetration rate, related practitioners, related output and cell phone penetration rate, but gave less consideration to technology-driven and strategic leadership level; Fei Wu et al. used the core structure of digital technology to develop a structured classification, and constructed digital transformation indicators from "underlying Through this perspective, this paper attempts to build a more complete digital transformation index system to measure the degree of digital transformation in a more detailed scale. Through this perspective, this paper tries to build a better digital transformation index system to measure the degree of digital transformation with a more detailed scale. In addition, based on the digital transformation indexes constructed in this paper, exploring the impact of the degree of digital transformation on enterprise innovation will also improve the literature on how the digital economy brings innovation "dividends" and enterprise management changes in the digital economy era, while the literature on whether the digital transformation of enterprises in the context of the digital economy can It is of great importance for China to seize the new opportunities of economic development in the digital era and to promote the innovation-driven strategy of enterprises in depth. In this paper, we will draw on the ideas of Wu Fei and other scholars, use the Python crawler function to organize the information of the annual reports of A-share listed companies, and extract the textual content by means of the Java PDFbox library, use the information about the company's business, operation, and development planning displayed in the disclosed annual reports of listed companies, gather the word frequencies of digital applications as well as technology-driven directions, and use AHP hierarchical analysis We use AHP hierarchical analysis and expert scoring method to assign indicators to six aspects: technology-driven, strategy-driven, organization-empowered, environment-empowered, digital achievement and digital application, to build a relatively complete evaluation system and obtain the digital transformation index. In addition, this paper will apply a more comprehensive perspective to explore the relationship between corporate digital transformation and corporate innovation, not only to investigate the overall impact of corporate digital transformation on corporate innovation, but also to sort out how R&D personnel, R&D investment and corporate learning effects affect corporate innovation as the mechanism of digital transformation. Therefore, compared with the existing literature, the marginal contributions of this paper are summarized as follows: ①Text analysis is used to construct a data pool of characteristic words about digital transformation in annual reports of listed companies, and AHP hierarchical analysis and expert scoring are combined to construct an evaluation index system from six dimensions: technology-driven, strategy-driven, organizational empowerment, environmental empowerment, digital achievement and digital application, and the weighted sum is obtained from the digital The digital transformation index presents the digital transformation development status of A-share listed companies in China more intuitively, and effectively improves the measurement of the digital level of enterprises from the micro level. ② To a certain extent, it expands and improves the research framework and empirical analysis on the mechanism of digital transformation on the promotion effect of enterprise innovation output in China, and further enriches the research on the role of digital transformation on enterprise innovation channels. ③ Clarifying the effects of digital transformation on enterprise innovation in terms of enterprise size, enterprise characteristics, enterprise nature, executive education, and geographic conditions of enterprise location helps summarize successful experiences at the level of digital technology-enabled real economy, and provides references for the subsequent formulation of more detailed digital transformation policies to accelerate the upgrading and transformation of enterprise kinetic energy and promote enterprise digital transformation.

Mechanisms and Research Hypotheses
Corporate innovation is highly dependent on the investment of capital and knowledge, and the digital economy influences corporate innovation precisely through the resource effect [10]. Specifically, on the one hand, the development of the digital economy promotes a more suitable financing environment for enterprises, and in this context, companies that undergo digital transformation in a timely manner will have access to more diversified financing channels and receive abundant financial support. On the other hand, the development and application of digital technology has greatly alleviated the long-standing information asymmetry problem among enterprises, and the mobility of knowledge elements is enhanced with the help of information technology platform. According to Chi, I.D., et al.(2020), the hyperlinked nature of digital technology has reshaped the relationship between enterprises and between enterprises and consumers, and the more convenient interconnection is conducive to the dissemination of knowledge elements among innovation subjects, while the inter-enterprise learning effect will also be more obvious, thus bringing positive impact on the innovation output of enterprises, based on which, this paper proposes hypothesis 1: Digital transformation of enterprises has a positive impact on enterprise innovation (H1). Existing studies have found that it is unrealistic to make use of the financial profit growth of enterprises to gain a long and comprehensive competitiveness in the market, and the development of basic research is the key to improve the core competitiveness of enterprises. By increasing the proportion of R&D personnel in the total workforce and the overhead of R&D investment, on the one hand, we can ensure that labor factors have a high degree of specialization and are good at the integration of various links of the value chain, thus improving business processes and mobilizing the enthusiasm of enterprise R&D [11]; On the other hand, R&D personnel with higher market acumen can better grasp market information and transform cutting-edge innovative ideas into innovative results [12]. Based on this, this paper proposes hypothesis 2: Digital transformation of enterprises can increase the share of R&D personnel and the share of R&D expenditure, which in turn promotes the development of enterprise innovation (H2). The learning effect is a fundamental support for the innovative development of enterprises [13], and enterprises should maximize the integration of the learned knowledge with their own production and operation [14]. In addition, based on the findings of Chuzhi and Ou Jinwen, we can conclude that enterprises can promote their innovation output through intra-industry learning of "peer-to-peer" and inter-industry learning of "crossborder integration", thus accelerating their innovation efficiency.Under the background of digitalization, it is more important for enterprises to maintain their core competitiveness in the market continuously by learning and imitating the development patterns of leading enterprises in the industry and integrating knowledge resources to reconstruct their own business forms and business models in the market. Faced with the competitive information society, enterprises must respond to the rapid changes in the environment and optimize and adjust the innovation model to achieve differentiated development. Based on this, this paper proposes hypothesis 3: Enterprise digital transformation promotes innovation development by mobilizing the learning and imitation consciousness of enterprises (H3).

Sample selection and data source
In this paper, the initial research sample of Chinese Ashare listed companies from 2011 to 2021 is used to obtain data from the following two aspects: First, the digital transformation index, which can reflect the degree of digital transformation of listed companies, is obtained by using the annual reports of listed companies from 2011 to 2021, extracting key words from them and using the text analysis method, AHP hierarchical analysis method and expert scoring method. This data is mainly obtained by using the annual reports of listed companies from 2011 to 2021, extracting key words from them, and using the text analysis method, AHP hierarchical analysis method and expert scoring method to obtain them; the second is to obtain the innovation data and data of corporate management and financial level of listed companies, and such data mainly comes from CSMAR database, CNRDS database and Wind database. Then, the samples were screened according to the following principles: ① Only continuous samples with no missing data for at least five years were retained. ②Excluding samples that are treated as * ST, ST, PT during the sample period. ③Excluding samples that do not comply with accounting standards. Finally, in order to reduce the impact of outliers, this paper applies the upper and lower 1% tail-shrinking process to all micro-level continuous variables.

Dependent Variable
The dependent variable in this paper is firm innovation. In recent years, there has been a proliferation of literature using the number of patents to measure the level of innovation output. Combined with the description of invention patents in the statistical yearbook, this paper believes that invention patents are internationally recognized as the core indicator that can reflect the possession of independent intellectual property rights by enterprises or individuals, as well as the core indicator that can reflect the ability of independent innovation, which can not only measure the innovation output of enterprises, but also serve as the performance of the external results of enterprises' basic research investment [15]. Therefore, this paper uses the number of corporate invention patents as a measure of corporate innovation to explore the relationship between digital transformation and corporate innovation output of A-share listed companies in China.

Independent Variable
The independent variable in this paper is the Enterprise Digital Index. As a measure of the degree of digital transformation, the size of the index represents the strength of an enterprise's digital transformation. The suitability of the transformation strategy profoundly affects the success or failure of digital transformation. From a macro perspective and qualitative analysis exploring theoretical descriptions [16,17], digital transformation can be measured using digital economy indicators at the regional or industry level. In recent years, the above-mentioned scholars' qualitative exploration no longer met the academic needs, so scholars Fei Wu et al. redefined the behavioral variable of "digital transformation" from the perspective of quantitative analysis in 2021. The implementation approach chosen by Wu et al. was to first conduct a structured stratification of digital transformation into two dimensions: "practical application of technology" and "use of underlying technology", and then extract the words "digital transformation" or those indirectly describing digital transformation from the annual reports of listed companies and key policy documents. Then, we extracted the words "digital transformation" or words that can indirectly describe digital transformation from the annual reports of listed companies and important policy documents, compiled the statistics, built a digital transformation characteristics map, and finally obtained a digital transformation indicator system. Figure 1 with the help of enterprise data published by Mark Data Network and the White Paper on China's Digital Economy Development (2021) published by China ICT, which shows that the growth trend of digital transformation word frequency disclosed by listed companies is comparable to the growth of digital scale of Chinese enterprises, supporting the use of text analytics to extract the frequency of digital transformation words to measure the correctness of enterprise digitization. However, considering that Fei Wu and other scholars construct the indicator system with a single dimension, it is difficult to reflect the whole picture of digital transformation of enterprises, considering only that in the process of digital transformation, enterprises will first rely on the upgrade of technology system to improve quality and increase efficiency. Combined with the applicability of the research topic and the availability of data, this paper, based on the use of the sub-indicator of technology empowerment, a total of six sub-indicators are weighted and summed with the help of strategic drive, organizational empowerment, environmental empowerment, digital outcome and digital application, to build an evaluation index with relatively complete dimensions of the index system -digital transformation. The index weights are shown in Table 1.

Model Setting
The following econometric model is constructed in this paper.
Inpatent represents the number of patents filed by a company, which is used to quantify the innovation output of a company. Digit represents the digital transformation index, which is used to measure the degree of digital transformation of enterprises, and the construction method has been explained in detail in the previous section. Controls represents the control variables.
Year fixed effects and firm fixed effects are included to control for the effects of macro factors on firm innovation.

Benchmark regression and robustness testing This article begins with benchmark regression. The results
show that there is a significant positive relationship between digital transformation index and enterprise innovation output, and the digital transformation index is significant at the 1% confidence level, confirming hypothesis 1. Meanwhile, based on the consideration of the robustness of the research results, this paper uses the number of enterprise invention patents and the number of patents applied by enterprises in the second year, instead of the digital transformation index for robustness testing. It can be found that the coefficient of the digital transformation index is significantly positive at the 1% level, indicating that the conclusion that digital transformation significantly enhances the innovationdriving effect of firms is robust.

Benchmark regression and robustness
testing In summary, corporate innovation includes the following three aspects [18,19]:①Independent innovation is mainly reflected in the transformation of innovation inputs into innovation outputs by combining new research directions, where innovation inputs include recruiting R&D personnel and increasing research funding. ② Learning innovation is borrowing from the transformation experience of companies with significant digital transformation and proactively receiving external explicit knowledge.③Competitive innovation is that in the digital context, multi-source information connection makes the value creation of similar enterprises more open and transparent, and the comparative competition of similar products will drive the innovation development of enterprises. In order to further investigate the impact of digital transformation on the innovation output of enterprises, this paper uses the percentage of R&D personnel, R&D expenditure and patent citations as estimators to conduct regression tests on the innovation effect of enterprises from each of the three aspects. The mediating effect model is constructed as follows. (3) Inter represents the mediating variable.

Mechanism test: the effect of the number of R
Based on the regression results in the previous section, this paper infers that the mechanism of digital transformation to promote the innovation output of enterprises is to increase the proportion of R&D personnel in total employees, thus mobilizing the motivation of enterprise R&D innovation. Firms with a higher percentage of R&D staff will have a head start in the implementation of digital transformation strategies, which will have a more significant effect on the innovation output of such firms. For this purpose, this paper obtains data on the percentage of R&D personnel from the CSMAR database to examine the mechanism effect. The test results show that the innovation capacity of enterprises significantly increases at the level of 1%, confirming that the digital transformation of enterprises is a way to increase the motivation of enterprises to innovate by increasing the number of innovators.

Mechanism test: the impact of R&D funding
As the first productive force of enterprise development, technology is the most active factor in the production process, and securing sufficient R&D funding is the only way to maintain the momentum of enterprises' continuous innovation activities. Therefore, this paper expects that if enterprises want to better apply digitalization to complete their innovation drive, they must pay attention to the role of science and technology R&D to support digital transformation, and hypothesizes that the mechanism of digital transformation to promote the improvement of enterprises' innovation output is to increase the proportion of R&D expenditure in total expenditure, which ensures the sustainability of enterprises' R&D innovation. To address this inference, this paper conducts a mechanism test and obtains the results of the mechanism test exploration as shown in Table 2. It is found that the estimated coefficient of the mediating variable R&D expenditure is significantly positive, and the digital transformation index is still significant at the 1% confidence level, indicating that the digital transformation of enterprises is to meet the sustainable R&D experiments of R&D personnel by increasing R&D expenditure, and the adjustment of R&D expenditure effectively stimulates the innovation vitality of enterprises, thus increasing the research funding becomes one of the mechanisms for digital transformation of enterprises to promote their innovation output. -0.107 0.413 -0.103 Note: *** indicates significant at the 1% level, ** indicates significant at the 5% level, * indicates significant at the 10% level, and t-statistics are indicated in parentheses; if not otherwise specified, the following tables are annotated identically.

Mechanism test: the impact of enterprise learning effect
The digital era has created a relatively open internal environment for enterprises to receive new information, which has greatly widened the access to information resources. Enterprises promote their innovation output through intra-industry learning of "peer-to-peer" and inter-industry learning of "cross-border integration", thus accelerating their innovation efficiency. Therefore, this paper tries to explore the mechanism between digital transformation and enterprise innovation from the perspective of inter-industry learning effect. The test results are shown in Table 3, which shows that with the implementation of enterprise digital transformation strategy, enterprise innovation capability is significantly improved, and the estimated coefficient of mediating variable patent citation is significantly positive. digital transformation enhances its own product development capability through learning and borrowing, thus realizing enterprise innovation.

Heterogeneity test
The relationship between enterprise digital transformation and enterprise innovation output may be affected by various objective conditions. This paper explores whether enterprise digital transformation drives enterprise innovation output by five aspects: whether it is a large enterprise, whether it is a high-tech enterprise, whether it is state-controlled, the level of education of enterprise directors, and the geographical distribution of enterprises.

Heterogeneity test -enterprise size
The sample is divided into large enterprises and small and medium enterprises based on the number of employees and business revenue of the enterprises in the year, and heterogeneity is tested by referring to the "Statistical Classification of Large, Small, Medium and Micro Enterprises (2017)" issued by the National Bureau of Statistics. The results illustrate that digital transformation of enterprises can increase the innovation-driving effect of both large and small and medium-sized enterprises, but has a relatively small effect in enhancing the innovationdriving effect of small and medium-sized enterprises.

Heterogeneity test -firm characteristics
The regression results of the grouping of enterprises show that digital transformation of enterprises significantly enhances the innovation-driven effect of enterprises for both high-tech and non-high-tech enterprises at 1% confidence level.

Heterogeneity test -nature of enterprises
State-owned enterprises grasp the lifeblood of the stateowned economy and have distinctive and strong institutional advantages. The unified decision-making of state-controlled enterprises can build a digital system exclusively for the coordination and deployment within state-owned enterprises and give full play to the function of the digital coordination system network to achieve intelligent development. If we can actively promote the innovative application of the new generation of information technology and take the initiative to lead the new generation of technological changes to complete the transformation of the state-owned economy, it will lay a good start for the digital transformation work and greatly mobilize the transformation and development of the overall economy of China. Therefore, this paper distinguishes by state-controlled enterprises and non-state-controlled enterprises and conducts group tests. The results show that digital transformation significantly enhances corporate innovation at the 1% confidence level in SOEs, while it is relatively less significant in non-SOEs. There are two possible reasons for this result: First, compared to the policy constraints of private enterprises, SOEs can receive more preferential policies at the government level, and the internal assessment and evaluation system and incentive mechanism of SOEs are more conducive to the promotion of corporate innovation. Second, compared to the small economic scale of private enterprises, SOEs have a larger asset volume and are suitable to carry digital information systems and realize large-scale investment in smart manufacturing.

Heterogeneity test -executive education
The development of enterprise innovation is a complex and multifaceted technical and economic activity. A good digital innovation system is inseparable from the forwardlooking control of digital frontier technology by business managers, and the talent, boldness, pioneering and innovative spirit and social responsibility of entrepreneurs themselves are more important in the context of leading the development of enterprise innovation in the digital era. Therefore, we use education as an important indicator to estimate entrepreneurship, and divide the education level of corporate directors by bachelor's, master's, and doctoral degrees to obtain the grouping test results. The results show that the digital transformation of a director's company has a significant effect on promoting corporate innovation when the director has a master's degree, while the digital transformation of a company increases the innovation-driving effect of the company at a 5% confidence level when the director has a doctoral and bachelor's degree, but the company led by a director with a doctoral degree performs better in the area of using digital transformation to promote corporate innovation.

Heterogeneity test -geographical conditions
The institutional and competitive environments faced by enterprises are highly correlated with the regions where they are located, so the effect of digital transformation of enterprises must be influenced by geographic conditions. The regression test of digital transformation on innovation is conducted by dividing the regional subsamples into Northeast, East, Central and West. The results show that the east region have the most significant effect in enhancing the innovation-driving effect of enterprises, enterprises in the west region significantly enhance the innovation-driving effect of enterprises at 5% confidence level, and the northeast region have 10% level. In the central region, the effect of digital transformation is not significant. The possible reasons for this situation are that the eastern region has the Yangtze River Delta economic zone and mostly coastal cities, and the market openness and circulation in this region is much higher than other regions, which makes it easier for resource sharing and efficient business cooperation, and the learning effect among enterprises is stronger, so the practice ability is better than other regions, and they can integrate into the tide of digital transformation and promote enterprise innovation development more quickly. Compared with the eastern region, the western and northeastern regions also show significance in the enhancement of enterprise innovation by digital transformation, but the significance is relatively weak because: the agglomeration effect in the production side is not high in the western and northeastern regions, and the manpower, material resources and fixed assets of enterprises cannot be sensitively regulated according to the economic environment.

Endogeneity
There are endogenous challenges caused by inverse causation in the empirical research of this paper, to mitigate the estimation bias caused by reverse causality effects on the empirical findings, telephone penetration and fiber optic cable length are used as instrumental variables for endogeneity testing [20,21]. This paper select telephone penetration and fiber optic cable length indicators as instrumental variables in the situation of telecommunication communication service level in each province and city as well as the situation of telecommunication infrastructure in each province and city [22]. The regression results of the 2SLS test can found that the coefficient of the digital transformation index is significantly positive, confirming the validity of the instrumental variable selection, indicating that digital transformation has a significant enhancement effect on the innovation drive of enterprises.

Conclusions and Recommendations
The use of text analysis to explore the impact of corporate digital transformation on corporate innovation output is an important research topic in the context of the current digital economy. This paper constructs a digital transformation index by combining the initiatives of digital transformation of A-share listed companies in six aspects: strategic drive, organizational empowerment, environmental empowerment, digital results, and digital application, empirically examines the impact of digital transformation on corporate innovation output, the impact mechanism, and examines the impact of digital transformation on corporate innovation output from a total of five external The heterogeneity is explored from five external conditions: whether it is a large enterprise, whether it is a high-tech enterprise, whether it is statecontrolled, the level of education of enterprise directors, and the geographical distribution of enterprises. First, there is a strong correlation between enterprise digital transformation and enterprise innovation, and enterprise digital transformation has a significant positive promotion effect on enterprise innovation output. Second, the digital transformation of enterprises mobilizes the enthusiasm of enterprise R&D by optimizing the proportion of R&D personnel in total employees and the overhead of R&D investment, which in turn promotes the development of enterprise innovation. Third, the digital transformation of enterprises promotes innovation output through intra-industry learning of "peer-to-peer" and inter-industry learning of "cross-border integration", thus accelerating the efficiency of enterprise innovation. Fourth, the effect of digital transformation on enterprise innovation is more significant in state-owned enterprises, high-tech enterprises, and eastern regions. The policy implications of this paper are as follows. Firstly, digital transformation of enterprises should follow the principle of differentiation, and enterprises should give full play to their own conditions and advantages to create digital transformation solutions with characteristics. ① SMEs have less liquidity, so they should focus on encouraging integrated innovation and re-innovation led by digestion and absorption when conducting digital transformation. As for large enterprises, they should pay attention to improving the awareness of independent innovation and realizing original innovation mainly by independent research and development. ② In the perspective of the nature of property rights, so they should focus on making up for the long-standing low efficiency of them as well as improving the technology ratio of statecontrolled enterprises when carrying out digital transformation. Considering private enterprises are inferior to state-controlled enterprises in terms of volume and scale, the flexibility of their innovation mechanisms and the activeness of their R&D teams give them superior institutional advantages, they should grasp the institutional advantages, maintain the high efficiency of technological innovation, and implement patent strategic plans suitable for the company. ③ The government should pay attention to the role of cluster effect in regional innovation, and enterprises should strengthen the awareness of resource interoperability and learning from peer enterprises. Secondly, open the chain of data integration and realize refined operation. ①For enterprises, facing the complex and changing economic environment, they should accelerate digital transformation to broaden the access to data information. ② For the government, it should accelerate the construction of new infrastructure represented by 5G, gradually reduce the imbalance in the development of digital technology between regions, by continuously improving the advanced nature of China's infrastructure ensure the completeness of the digital transformation system. Thirdly, improve the talent training mechanism and establish an integrated innovation system of industry, research and academia. Enterprises should integrate the resource advantages of innovative talents in universities, form relatively stable alliances between industries, universities and research institutes, establish a certain amount of independent innovation bases, realize the gathering of scientific and technological resources, and maximize the use of external technological resources to promote the production and R&D system. Lastly, deepen the division of labor among domestic regions and industries, and build a new development pattern of economic efficiency. In this context, the government should combine the resources of each region and the technical level of each industry, actively improve the management mechanism of the market, optimize the administrative approval process, target the filling of regional infrastructure, create an open platform for digital information exchange, supervise enterprises to promote digital transformation division of labor, and drive realize enterprise innovation.