| In recent years,with the rapid development of technology,the phenomenon of technology convergence in fields such as information,energy,and biology has also emerged one after another.Since the proposal of ‘NBIC converging technology’,technology convergence has attracted much attention.Biotechnology has a typical feature of technology convergence,and the phenomenon of technology convergence,both within and between biotechnology and other technology,is fairly common,bringing new transformations and breakthroughs to it.Therefore,it is important to carry out a systematic analysis of the development situation of technology convergence of biological technology.This study focuses on the construction of patent-based technology convergence analysis model,and analyzes the development situation of technology convergence by constructing multi-level patent co-occurrence networks.By using ‘link prediction +random forest’ method,the potential technology convergence relationship is predicted from multiple levels.The field of synthetic biology was selected for empirical analysis,which verified the feasibility,validity and practicability of the patent-based biological technology convergence analysis model proposed in this study.The main research work and innovative achievements of this paper are as follows:(1)The relevant theories and methods of technology convergence analysis and research progress are systematically studied.Through the domestic and foreign literature research on the technology convergence of biological technology based on patent data,this paper systematically analyzes the research progress of patent citation analysis method,patent co-occurrence analysis method,data mining method and index method,and deeply analyzes the existing shortcomings,which lays a solid foundation for the construction of technology convergence analysis model and technology convergence prediction.(2)A technology convergence analysis model based on patent is constructed.Three levels of international patent classification code are adopted in this study: IPC3,IPC4 and IPC8 respectively construct patent co-occurrence network,and combine social network analysis,link prediction,random forest and other methods,and use Python,Gephi and other tools to build a patent-based technology convergence analysis model.The model mainly consists of the following five levels: Patent data layer: Obtain the required patent data from the selected patent database;Patent co-occurrence matrix layer: Three levels of IPC are extracted from the obtained patent data,and the IPC3 co-occurrence matrix,IPC4 co-occurrence matrix and IPC8 co-occurrence matrix are constructed by writing code in Python;Patent co-occurrence network layer: On the basis of the co-occurrence matrix of the above three levels,Gephi is used to transform the pair-to-pair convergence relationship between technical fields revealed by IPC co-occurrence matrix into patent co-occurrence network,and IPC3 co-occurrence network,IPC4 co-occurrence network and IPC8 co-occurrence network are obtained respectively;Technology convergence situation analysis layer: Based on the patent co-occurrence network of three IPC levels constructed above,analysis of technology convergence situation is carried out from the two perspectives of identification of hotspots of technology convergence and network index analysis,and the interaction relationship between different IPC levels is observed.Network index analysis mainly uses network density,network clustering coefficient,Betweenness Centrality and other indicators;Technology convergence relationship prediction layer: the method of "link prediction + random forest" is adopted to predict the potential technology convergence relationship from three levels,providing information support for the future development of technology convergence.(3)Taking synthetic biology as an example,an empirical study of technology convergence analysis based on patent data was carried out.The technology field of synthetic biology was selected as the research object of empirical analysis,and the technology convergence situation of synthetic biology during 2012-2014,2015-2017 and2018-2020 was analyzed by using the constructed multi-level IPC technology convergence analysis model,and the potential technology convergence relationship of synthetic biology in the future was predicted.The validity of the prediction method of potential technology convergence relationship was tested by the indicators such as AUC and Precision,and the feasibility and practicability of the model were verified.The above analysis and prediction results have a strong reference value for the innovation and development of synthetic biology in the future.The patent-based technology convergence analysis model constructed in this study can effectively analyze and predict the situation of technology convergence in the field of biotechnology.This model not only further enriches the theoretical methods of technology convergence research,but also provides a new method for a more comprehensive understanding of the convergence situation in the technology field. |
