Research On The Patent Value Of Artificial Intelligence Based On Patent Analysis

With the rapid rise and development of Industry 4.0,it has witnessed the booming and prosperity of artificial intelligence.In recent years,Chinese government attaches great importance to the technological progress and industrial development of artificial intelligence.AI industry has become the focus of national economic development,and China's artificial intelligence has attached worldwide attention.Moreover,patents as an important form of technological innovation,have become a symbol of the development of artificial intelligence technology.With the introduction of high-value patents,patent value has become an important way to measure the technological innovation capability of enterprises.Therefore,it is necessary to evaluate the patent value of artificial intelligence in China so that we can better develop artificial intelligence technology,create high-quality patents,and cultivate high-value patents.Firstly,this paper defines the concept of artificial intelligence and patent value.Then this paper introduces some methods to measure the patent value in the exiting literature,such as real option theory,theory capture theory and comprehensive evaluation.After analyzing the current situation of artificial intelligence technology,this paper analyzes the state of technology competition in the field of artificial intelligence in China,and finds the main problems of patent development in the artificial intelligence field.Secondly,this paper constructs a comprehensive evaluation system based on the three dimensions of patent value,excavates effective information from patent literature information.In addition,this paper makes use of the entropy-TOPSIS method to measure and calculate the patent value index to evaluate the patent value from the vertical and horizontal dimensions.In the vertical dimension,this paper analyses the development of patent value in the field of artificial intelligence during 2004 to 2014.And in horizontal dimension,this paper analyses the distribution of patent value across the regions which contain 29 provinces and cities in China by using the patent sample data of artificial intelligence applied in 2014.Then,by using the invention patents in the field of artificial intelligence applied in China with the priority years in 2000-2010,retrieving from the inco Pat patent database.This paper takes citation lag as the proxy of patent value and further use cox regressions to analyze the influencing factors of patent value.Evidently the empirical result shows that the patent family size,number of IPC classifications,backward citations and forward citations have significant positive effects on patent value,the number of claims has significant negative effects on patent value.Finally,this paper puts forward some policy inspirations to promote the development of high-quality patents and the suggestions for the progress of artificial intelligence technology according to the empirical analysis results.