Research On Technology Opportunities Analysis Of NC Machine Tool Based On Technical Knowledge Graph And Deep Learning

With the continuous development of artificial intelligence,cloud computing,Internet of Things and other technologies,the intelligent and networked technology of NC machine tool has obtained a new opportunity.In this context,it is of great significance to identify the technology opportunities in the field of NC machine tool quickly and accurately.In this way,countries and enterprises to seize the first opportunity and gain competitive advantages in the emerging technology field.At present,the research of technology opportunities analysis are constantly enriched,but there are still four problems as follows: Firstly,in the role of domain experts,there is a lack of effective embedding of expert knowledge.Secondly,the dimension of data information is relatively single.Thirdly,current research usually uses a single rule to filter the knowledge elements that reflect technology opportunities,resulting in a relatively limited search space for technology opportunities.Fourth,most of the representation methods of technology opportunities represented by keywords and patent classification numbers are poor in interpretability and specificity.Aiming at the above shortcomings,this paper proposes a compound technology opportunities analysis method.The first step is to construct the technical knowledge graph of NC machine tool and carry out technology development trend analysis.These two outputs accurately and comprehensively reflect the current development status of NC machine tool technology,and provide basic knowledge support for subsequent analysis of technology opportunities.In this way,the first problem was solved.For the second problem,this paper uses the semantic and co-occurrence information of CPC to construct the CPC co-occurrence network.Then the graph neural network and link prediction method are combined to mine the technology opportunities in the network.In order to further explore the specific connotation of technology opportunities,this paper builds a deep neural network model to learn the implicit fusion association between patents.The trained model can identify patent pairs that may provide technology convergence opportunities.Then extract the SAO structure of patents to obtain a description of specific technology convergence opportunities.Finally,the last two problems are solved through this process.Through methods and practice,this paper summarizes the construction process of the technical knowledge graph and completes the formulation of the technical knowledge graph of NC machine tools.On this basis,the identification of technology opportunities of NC machine tool technology is completed in the field of information technology and artificial intelligence technology.The results of the identification can provide forward-looking decision support for the formulation of future technology development strategy of CNC machine tool.