A Semantic Mining-based Method To Analyze And Evaluate Technology Innovation Pathway

With rapid technological advances and globalization,technical innovation is becoming increasingly important.All nations are conducting strategies to strengthen their technical innovation and trying to position themselves favorably in emerging industries,such as new energy,information technology,and biology.Accordingly,how to identify technology opportunity and forecast technology innovation pathways has become an important theme,which can be helpful for technology innovation decision-making and has attracted increasing attentions from enterprisesand researchers worldwide.This thesis constructs a systematic model to forecast innovation pathway(FIP)based on SAO semantic method.The model follows the order of “theoretical framework and methodological basis”?“problem & solution trajectory identification”?“technology morphological forecasting”?“commercial potential evaluation”.The method being used in the model includes text mining,technology roadmapping,morphological analysis,fuzzy consistent matrix and subject–action–object(SAO)semantic mining method.The research is supported by the General Program of National Natural Science Foundation of China(Grant No.71373019)and Basic Scientific Research Fund of Beijing Institute of Technology(Grant No.20132142008).The main innovation achievements are summarized as follows:(1)Constructing a“problem &solution”trajectory identification modelThe thesis proposes a new method to identify technology development trajectory based on subject–action–object(SAO)semantic analysis.Different from keyword-based quantitative methods,the SAO-based method allows to match technology solutionswith key goals,and to identify the relations among different technology solutions accurately.To analyze technology solutions for each goal,we construct comprehensive “goal & solution” evolutionary diagrams.In it,nodes are key technical phrases in the form of adjective + subject/ adjective + object,which can express a complete and clear meaning so that it can show key technical findings more accurately and completely than keyword-based method.Besides,based on the detail information in SAO structures,we introduce technology comparison diagrams to compare the performance of technology solutions,which help us to better identify important and promising technology solutions of the goals.Finally,to illustrate the feasibility of the proposed method,the dye-sensitized solar cells(DSSCs)technology is used as the case study.(2)Constructing a technology morphological forecasting model based on SAO semantic analysis and morphological analysisThe thesis constructs a technology morphology forecasting model based on SAO semantic analysis and morphological analysis.SAO semantic analysis can clearly represent how keywords are related and can identify the purpose of a given technology.Accordingly,this thesis combines SAO semantic analysis method and keywords frequency analysis to construct morphological structure,which can make up for the limitations of keyword-based approaches and build morphological structure more objectively.To evaluate and forecast future morphological configurations,the thesis combines patent analysis and bibliometrics method to build a systematic and comprehensive technology evaluation index system.Using regression analysis,a morphology forecasting model is constructed based on historical data.Finally,to validate the proposed method,an empirical study is conducted using DSSCs technology as an example.(3)Constructing a new technology commercial potential evaluation model based on fuzzy consistent matrixThe thesis constructs a new technology commercial potential evaluation model for technical solution or new technology combination,which integrates methods of fuzzy consistent matrix,entropy weight method and expert estimation.The model divides the analytical process into three steps: screening technical solutions,constructing evaluation index system and conducting technology evaluation based on fuzzy consistent matrix.As for application research,we also choose DSSCs technology as an example to conduct empirical anlaysis.The feasibility and validity of the model are illustrated.Besides,the analysis results will be helpful for enterprise technology innovation and industry development decision in DSSCs filed.