| As an important industry of the real economy,the information industry occupies an important position in economic development.In recent years,Yunnan Province has continuously increased its support and attention to the development of the information industry,and technological innovation continues to emerge,the number of patents brought about by this has steadily increased year by year.The patent literature implies the most advanced technical points in this field at this stage,and its use value is extremely high.At present,the patent analysis methods mainly include statistical analysis methods and technical analysis methods.This article combines the subject knowledge and researches the patent data from a brand-new point of view in order to show the current patent development situation and provide suggestions for the future improvement of the patent level of the industry.The main research contents of the paper are as follows:1.This paper collected literature and other data to complete the theoretical knowledge.Retrieved Yunnan's information industry patent data from January 1,2007 to December 31,2017,adopted a patent information statistical analysis method to examine the annual trend of patent applications from a macro perspective,patent types and legal status,main rights holders,statistics of major IPC distribution,patent authorization status and application day distribution.Compared with the number of patent applications in the western region,we analyzed the overall development trend of Yunnan's information industry patents.2.Based on co-word clustering analysis method to tap the hotspots of patent applications in recent years.The high-frequency keywords obtained after the word segmentation of the patent name are selected to build a co-occurrence matrix,to solve the problem of actual errors caused by large individual elements,the co-occurrence matrix is converted into a correlation coefficient matrix and the clustering is dug based on this.Eventually the key words of five kinds of high relevancy were determined.Also selected the IPC classification properties to obtain a high-frequency IPC classification of small class to verify and supplement the hot patent technical field.Through research,we learned the effective technology hot spot in the Yunnan information industry for patent applications,combined with the current status of industrial development in the country to obtain future development trends and research directions,and to provide references for future industrial patent applications.3.A method based on C4.5 decision tree to analyze the authorizing factors of patent data in Yunnan information industry is proposed.Selected the type of application,the number of inventors,the number of subordinate claims,the application date,and the number of IPC classification as the research attributes and made data preprocessing.Imported data on the Weka platform,then using the C4.5 decision tree,K-nearest neighbor,Naive Bayesian classification method to analyzed,at last the C4.5 decision tree algorithm with the accuracy rate of 87.08%was selected to complete the construction of the classification model.The decision tree of the structure was used to summarize the specific influence rules of the above five attribute on the authorization.Provide reliable advice when making a patent application for the patent applicant.Through the above research,the combination of the hot trend of patent applications in the province's information industry and the impact of other information on project authorization will provide targeted recommendations for the future development of Yunnan's information industry patents,thus enhancing the information industry patents in Yunnan.At the same time,it will also play a complementary role in promoting the good and rapid progress of the province's information industry. |
PDF Full Text Request