Research on National Scientific Research Cooperation between China and the “One Belt and One Road”—Based on the Frontier Gravity Model

With the continuous development of science and technology and the dee-pening of international academic exchanges, international scientific research cooperation has become an important part of national strategy. Therefore, the research on the influencing factors of scientific research cooperation between China and “One Belt and One Road” countries plays an important guiding role in promoting the cultural, scientific and technological exchanges between them. This article is based on the United States patent and trademark office (USPTO) database in 2008-2017 patent cooperation data. On the basis of studying the current situation of scientific research cooperation between China and countries along the Belt and Road by means of intelligence min-ing, this paper innovate and quote the model of frontier gravity in economics creatively, and conducte an empirical study on the factors affecting scientific research cooperation between China and countries along the Belt and Road. The empirical results show that, except that the countries bordering on each other will inhibit China’s scientific research cooperation with countries along the “One Belt and One Road”, national scientific research strength, national innovation capacity, bilateral relations and economic scale will all promote the development of international scientific research cooperation. However, the geographical distance between the two countries has no significant impact on it. Besides, the limitation of this paper is that the number of patent cooperation between China and “One Belt and One Road” countries is generally small; many countries are not involved in patent cooperation, and data of some countries are missing and incomplete.

Regarding the status quo of international scientific research cooperation: Michel (2010) used the probabilistic affinity to describe the cooperation between Britain, France, Germany, Japan and the United States from 1986 to 1996 [1].
And he found that the scientific research cooperation between the United States and Japan was closer than that between the other countries. Besides, from the perspective of cooperative publications, Tianwei He (2009) used bibliometrics to study the research cooperation between China and G7 countries, and found that the number of international cooperation publications increased exponentially in recent years [2]. Then, Wang and Xu (2013) study the international scientific research cooperation with China from the different countries, institutions and individuals [3]. The results shows that China's scientific research cooperation activities are mainly concentrated in a small number of countries, especially the United States. In recent years, more and more scholars have begun to study the status quo of scientific research within economic organizations. H. Bouabid from the perspective of the national cooperation papers to research the status quo of paper cooperation and the citation rate of cooperative papers in countries along the route [7]- [12]. The results show that with the introduction of the "One Belt and One Road" policy, the number of international cooperation papers between China and countries along the "One Belt and One Road" has been greatly increased and China is at the core of the "One Belt along the "One Belt and One Road" plays a positive role in promoting the international scientific research cooperation between the two countries [23].
By sorting out the above literatures, this paper can be summarized as follows: the existing research on international scientific research cooperation between China and countries along the "One Belt and One Road" route mainly starts from the perspective of international cooperative papers and publications, and few scholars study international scientific research from the perspective of international cooperative patents. In addition, most of the current researches only focus on the current situation of scientific research cooperation based on information science knowledge, and there are few papers using the empirical analysis method of economics to study the factors affecting international scientific research cooperation. It can be seen from the above that international scientific research cooperation is an important way to enhance national innovation, increase national scientific and technological vitality, and strengthen national scientific research exchanges. Studying the influencing factors of China's scientific cooperation with the "One Belt and One Road" countries has an important guiding role in promoting China's cultural and scientific exchanges with the "One Belt and One Road" countries. Therefore, this paper will conduct an empirical analysis of the factors affecting international scientific research cooperation between China and the countries along the "One Belt and One Road" through the study of the status quo of international cooperation patents between China and the "One Belt and One Road" countries.

Research Structure
This paper firstly reviews and sorts out the relevant researches in the field of na-

Distribution of Cooperative Patents
Based on the statistics of the US Patent and Trademark Office (USPTO) database, as of December 2017, China and the countries along the "One Belt and One Road" have cooperated to apply for 1217 international patents (as shown in

Distribution of Cooperative Patent Countries
Based on the statistics of the US Patent and Trademark Office (USPTO) database, Table 1 is given. As can be seen from Table 1 nalized. From this, we can see that the overall degree of patent cooperation between China and countries along the "One Belt and One Road" is not high, and it shows that the distribution of national patent cooperation is uneven, mainly concentrated in few countries.

Cooperative Patent Area Distribution
There are 65 countries along the "One Belt and One Road". However, due to the W. Huan complexity of intellectual property rights in the region, the laws and regulations related to intellectual property rights, as well as the level of management and implementation vary. For the convenience of observation, this paperuse the viewpoint of From the viewpoint of Zhang Mingqian and Deng Minmin (2018) and divides the "One Belt and One Road" countries into six regions, namely, Mongolia and Russia, 5 countries in central Asia, 11 countries in southeast Asia, 8 countries in South Asia, 19 countries in central and eastern Europe, and 20 countries in west Asia and the Middle East, according to international practices and other factors such as geography, proximity and culture [15] ( Table 2). From the USPTO database, the study found that the regional distribution of the number of cooperative patents between China and countries along the "One Belt and One Road" is uneven: southeast Asia is the main cooperative region of China, with 706 cooperative patents, more than half of the total number. South Asia, West Asia, and the Middle East each followed 15.201% and 11.832% of the total patents; Central and Eastern Europe, Russia and Mongolia and China have fewer cooperation patents, accounting for only 7.560% and 7.395%, respectively. Besides, it is worth noting that Central Asia and China do not have cooperative patents ( Figure 2).   Table 2. "One Belt and One Road" regional division.

Regional division Nation
North Asia (2 countries

Cooperative Patent Applicant
Before by analyzing the cooperative patent applicants from China and countries along the "One Belt and One Road" route in USPTO database, the applicants who had applied for more than 10 applicants are shown in Table 3. As can be seen from the above list, there are 16 enterprises or organizations have applied for more than 10 cooperative patents, among which 13 are dominated by multinational enterprises, especially the multinational technology companies in American. In terms of international applications, Huawei Technologies Corporation (51), International Business Machines Corporation (44) and General Electric Company (37) were the top three organizations.

Cooperative Patent Application Field
The cooperative patent IPC number will be used as a classification standard to conduct statistics on the cooperation patents between China and the "One Belt and One Road". Therefore, based on the USPTO database, the application field of international patents is analyzed, as shown in Table 4. It can be seen from Table 4 that the patent cooperation between China and the countries along the "One Belt and One Road" mainly focuses on the fields of electrical digital data processing, semiconductor devices, and wireless communication networks. Among them, Singapore, India, Russia and Israel are the major cooperation countries in these fields.

Model Setting
Based on the method of Zhang Cui and Ouyang dongping (2017), this paper introduces the gravity model into the international scientific research cooperation [23]. It combines the current situation of scientific research cooperation, and builds a new gravity model suitable for the study of international scientific research cooperation on the basis of the original gravity model. The mathematical expression equation of the known gravity model is:  In order to facilitate the cooperation of patents between China and the countries along the "One Belt and One Road", an empirical study on the factors affecting international patent cooperation is taken. This paper further refines the most general form of the gravity model and obtains the following mathematical expression: where, , , represents the relationship between object i and object j in time t, Next, the article brings in the explained variables and core explanatory variables, you can get: 1 2 , , 0 , , Collaboration i j t refers to the number of cooperative patents between China i and the countries j along the "One Belt and One Road" in t year,   Geographic distance (gd): Schwens C (2011) shows that the most common statistical method used in geography is to use Google Maps or other means of measurement to calculate the linear distance between the capitals or the distance of the aircraft [27]. However, such a method has its obvious drawbacks and is easily handled as a fixed effect when it is used for gravitational model measurement. But it cannot be recognized by software, so it has been gradually abandoned. This paper uses the measurement method of Jiang Guanhong (2012): the geographical distance is equal to the bilateral distance multiplied by the international oil price [28]. The geographic distance data comes from the CEPII database, and the international oil price comes from the US Energy Information Administration (EIA) database. And the international oil price used here is the weighted international oil price.

• Control Variables
Innovative capability (in): Existing research shows that innovation is an important factor affecting the development of national science and technology. The stronger the innovation capability, the greater the number of international cooperation patents in emerging technology fields and the higher the quality. The article here refers to the Innovation Indicators in the Global Competitiveness Index Report as a measure of the innovation capabilities of China and the countries along the "One Belt and One Road".
Bilateral relationship (re): when measuring the relationship between two countries, this paper classifies and assigns the value of bilateral partnership. This article, by consulting the official website of the Ministry of Foreign Affairs of the People's Republic of China, finds the bilateral relationship between China and the countries along the "One Belt and One Road" and classifies the bilateral relations according to the closeness of the country. It is divided into: general partnerships, strategic partnerships, and comprehensive strategic partnerships with assignments of 0, 1, and 2, respectively.
Economic scale (gdp): DuanQingfeng (2018) shows that capital has a direct impact on patent cooperation that the larger the economy of a country, the closer the patent cooperation [29]. In this paper, when measuring the size of a country's economy, this paper uses the indicator of gross national product, which is derived from the gross national product of the countries along the "One Belt and One Road" based on the 2010 constant US dollar disclosed by the The World Bank. To prevent data from being too large, this article records the unit of data as $100 billion.

• Virtual Variables
Neighboring countries (tig): In measuring the geographical location of China along with the "One Belt and One Road" countries, this paper has added a measure of whether there is border bordering. This article regards whether China and the countries along the "One Belt and One Road" have borders as a virtual variable. If two countries border on each other, it's one; otherwise, it's zero.

W. Huan American Journal of Industrial and Business Management
Language (lan): this paper also introduces common language as dummy variable. If there is a common language, it is denoted as 1; otherwise, it is denoted as 0.
The specific explanatory variables and data related information are shown in Table 5: Up to now, the number of countries along the "One Belt and One Road" has reached 65. This paper selects its data from 2008-2017 for a total of ten years for analysis. However, due to wars and inconsistent statistical methods, there is a lack of relevant variable data. Finally, this article deducts the countries along the "One Belt and One Road" that cannot obtain data information, and finally selects 45 countries along the "One Belt and One Road" as samples. Descriptive statistical analysis of each variable is shown in Table 6 and the specific sample countries and regions can be seen in the footnotes 1 .

Empirical Analysis
In this paper, using the 14th edition stata software, the panel data model is used to establish the gravity model. Then through the Huasman test to determine the model should use the time trend fixed effect model, and the specific test results are shown in Table 7. Since the number of collaborative papers in the interpretative variable is a non-negative integer measure and there is a zero value, a negative binomial fixed-effects regression model is used. Besides, considering that there may be endogeneity between the interpreted and explanatory variables,

Full Sample Regression
This sample regression test first regresses the core variables, and then continuously adds new control variables to the original ones, and then obtains the final empirical results. Table 8 is the full sample regression results: The variable Lnpatent1, the data passed the 1% significance level, and the coefficient is positive, indicating that China's scientific research strength has a significant positive correlation with the number of patents between China and the countries along the "One Belt and One Road". Moreover, the Lnpatent1 coefficient is larger than 0.7, indicating that the number of Chinese scientific research patents changes by one unit, and the number of international cooperation patents changes by more than 0.7 units. From that, it can be seen the number of 1 Sample countries: Mongolia, Russia, Kazakhstan, Singapore, Malaysia, Indonesia, Thailand, Cambodia, Vietnam, Brunei, the Philippines, India, Pakistan, Bangladesh, Sri Lanka, Nepal, Bhutan, Poland, Lithuania, Estonia, Latvia, Croatia, the Czech republic, Hungary, Slovenia, Bulgaria, Albania, Macedonia, Ukraine, Iran, Turkey, Jordan, Israel, Saudi Arabia, Oman, the United Arab emirates, Qatar, Kuwait, bahrain, Greece, Cyprus, Egypt, Georgia, Azerbaijan, Armenia.     It has the characteristics of being more convenient and easy to operate. The emergence of new communication methods such as network, mail and video can greatly reduce the distance constraints.
As for other control variables and virtual variables, we can see that innovation ability, economic scale have a significant impact on China and the "One Belt and One Road" countries, and they all play a positive role in promoting national patent cooperation. Besides, They passed the significance test on the basis of 5% W. Huan and 1%, respectively. One the other hand, the bilateral relations, language and bordering between countries have an impact on international patent cooperation, but the impact is small that they only pass the significant at 10% degree.
The border between the two countries shows that it has a hindrance to national patent cooperation at 10% significant. The possible reason is that the countries

Time Period Regression
In order to better analyse the changes in the factors affecting scientific research cooperation between China and countries along the "One Belt and One Road" before and after the implementation of the "One Belt and One Road" Strategy, we are now carrying out time segmentation and time-specific regression tests. Data for 2008-2017 are divided into two phases: the pre-"One Belt and One Road" strategy (2008-2012) and the post-"One Belt and One Road" strategy (2013-2017). And because virtual variables have less influence on patent cooperation, only control variables are introduced in the later regression. Table 9 is the result of time-limited regression: The pre-"One Belt and One Road" Strategy (2008-2012): China and countries along the "One Belt and One Road" continue to have an important influence on international patent cooperation at 1% significant. However, the innovation power of one country, bilateral relations between the two countries, and the size of the country's economy have no significant impact on international patent cooperation, because they don't pass the significant test. It is speculated that the reason is that most of the "One Belt and One Road" countries are developing countries, and their scientific research strength is weak and their economic base is poor so that their enthusiasm for cooperative scientific research in the early days was not high. The post-"One Belt and One Road" Strategy (2013-2017): With the continued investment in scientific research and innovation between China and countries along the "One Belt and One Road", innovation has become another important factor that has affected scientific research cooperation between countries after the strength of national scientific research at 1% significant test. However, geographical distance, bilateral relations between the two countries, and the country's economic size variables do not pass the significant test and had nothing to do with international patent cooperation. Since the "One Belt and One Road" Strategy, the impact of China's scientific research on international patent cooperation has diminished. However, the strength of Chinese scientific research is still the core factor in promoting patent cooperation between China and countries along the "One Belt and One Road".
And with the emphasis on scientific research innovation, innovation has become one of the driving factors that affect international scientific research cooperation.

Robustness Test
From a theoretical perspective, there may be missing variables in the model constructed in this paper, which makes the accuracy of the empirical results biased.
Therefore, it is necessary to test the robustness of the model. The robustness test can generally be implemented by replacing the estimation method or the substitution variable. This paper will use the least squares method to test the robustness of the article, as shown in Table 10. From the test results, it can be found that the original conclusion has not changed substantially, which also indicates that the empirical analysis of the model is robust and reliable.

Conclusions and Policy Recommendations
This paper uses the data of national patent cooperation between China and the "One Belt and One Road" in the USPTO database from 2008 to 2017 to analyze the status quo of scientific cooperation between China and countries along the route. And it makes an empirical study on factors affecting patent cooperation between China and countries along the "One Belt and One Road" by means of W. Huan Note: ***, **, *, respectively, are significant at the level of 1%, 5%, and 10%, and the standard deviation of robustness in parentheses.
gravitational model. The study found that the total number of patents between China and the countries along the "One Belt and One Road" has been increasing year by year, but its cooperation is not high and its distribution is uneven. The current situation shows that the top five countries in the cooperation have con- For example, on the one hand, the government should attach importance to innovation and practice education, liberate students' thinking. On the other hand, government should improve students' competitiveness and cultivate world-class talents with innovative spirit and creative ability. Second, China should strengthen its political, economic, and cultural ties with countries along the "One Belt and One Road" to lay a good foundation for promoting scientific research cooperation between China and the countries along the "One Belt and One Road". For example, China should further expand its foreign investment and import and export trade to countries along the "One Belt and One Road" by strengthening China's economic cooperation with countries along the "One Belt and One Road". In addition, China can promote scientific research cooperation through economic cooperation, and promote Chinese multinational enterprises to carry out patent cooperation with countries along the "One Belt and One Road".
Third, the uneven distribution of scientific research cooperation areas between China and the "One Belt and One Road" countries requires a new situation of scientific research cooperation between China and the "One Belt and One Road" marginalized countries. Therefore, China needs to further deepen its patent cooperation with its superior partners and consolidate the achievements of existing scientific and technological cooperation. China can set up research institutes and other international research conferences by holding research institutes and other research institutes in the "One Belt and One Road" countries to promote international academic resources, science and technology, professional talents mobility and complementarity. It will have a great help for the "One Belt and One Road" countries to participate in the process of international patent cooperation.

Conflicts of Interest
The author declares no conflicts of interest regarding the publication of this paper.