Research On Technology Recognition And Evolution Analysis Of CNC Machine Tool Field Based On Knowledge Graph

The manufacturing industry,which supports high-end technology industry and technological innovation,is a solid foundation for a country to achieve sound,stable and rapid economic development.As one of the core technologies in the high-end equipment manufacturing industry,CNC machine tool is regarded as the equipment mother machine for machinery manufacturing.In order to have a clearer understanding of the status quo of the development of CNC machine tools,promote the balanced development of production,education and research,grasping the key technologies and development paths in the field of CNC machine tools and formulating strategic plans for field development have important practical significance.Knowledge map is the scientific research method that has attracted much attention in recent years,it combines scientific metrology,computer graphics and artificial intelligence.From the perspective of statistics to establish a relationship in field of literature,knowledge map helps to dig and understand the hidden knowledge of it and plays an important role in the field of evolutionary path research work.In order to complete the technology identification and evolution analysis in the field of CNC machine tools,this paper takes the literature data in the field of CNC machine tools as the research object and carries out a series of research work around the construction of domain knowledge map.First,a set of phrase mining-based domain technology keyword identification process is constructed,combining two phrase mining models to complete the preliminary identification of technical keywords;In order to improve the shortcomings of traditional keyword recognition methods,a technology keyword extraction method based on patent text centrality and TF-IDF algorithm is proposed,which improves the accuracy of domain technology keyword recognition.Next,a semantic similarity network of keywords is constructed,a similar set of keywords is obtained by solving the largest connected graph and the merging rules of keywords are formulated to realize the attribute alignment of keywords;propose a term extraction method based on the semantic co-occurrence network,use the Textrank algorithm to calculate the term score,and complete the term extraction work in the field of CNC machine tools;build a terminology co-occurrence network and use Louvain algorithm to identify technical topics in the field.Then,the framework and method of knowledge graph construction is designed.According to the patent texts and extracted terms in the field of CNC machine tools,the entity and association relations of patents and terms is established,and the knowledge graph of patent technology in the field is generated.Finally,on the basis of the constructed knowledge graph,the patent text is clustered to guide the division of technology development stages,identify the technical themes of each stage,and calculate the similarity between themes;formulate the association and evolution rules of the theme,complete the technical evolution analysis work in the field of CNC machine tools by means of visualization,and summarize the thinking about the development trend of the field of CNC machine tools.