Research On The Construction Of Academic Inheritance Network Based On TCM Personage Knowledge Map

1 Research PurposeAfter in-depth analysis and careful collation of various dictionary reference books,using bibliography,history,complex network theory analysis and other methods,this paper uses Python and Pandas toolkit,Gephi and other tools to study the relationship between TCM figures as the main object,and discusses the construction of TCM figures knowledge graph(TCM figures network).The academic inheritance of TCM is studied by exploring the relationship between TCM personages,in order to provide reference for the study of academic inheritance of a family,a discipline and a school.2 Research materials and methodsThe ontological data of people and documents in this study come from a wide range of sources,including official history,general records,related books,medical books,dictionaries,local records,Fu Records,related foreign books and databases,etc.Dictionaries include Biographies of ancient Chinese Doctors,Characters Dictionary of Traditional Chinese Medicine,and Literature Dictionary of Traditional Chinese Medicine.On the basis of studying and analyzing the ontological data one by one,the author also compares the Metadata Specification and Description Rules of Ancient Books of National Library and Friend-of-a-Friend(abbreviated as FOAF),Functional Requirements for Normative Data,and Standards for Normative Records of International Archives(Group,Individual,and Family),etc.Define metadata of people and documents in combination with the characteristics of traditional Chinese medicine.Referring to Cao Feifei’s Research on the Spatio-temporal Data Modeling of the Relationship between Figures in the Six Dynasties,the biographical database of Chinese characters in the past dynasties,the database of Jinshi in the past dynasties,the knowledge map of Chinese medical scholars and other relevant research on the relationship between figures and literature,the author defines the characters and documents as a unified carrier from the perspective of academic inheritance in the form of "human and book integration".Manual indexing based on defined metadata and relationships.Then Python and Pandas toolkit were used to process the indexed data,and the processed data was imported into Gephi to generate the TCM human relationship knowledge map.Finally,combined with complex network theory and technology,the relationship between TCM personage is analyzed.3 Research ResultsIn Chapter 1,there are 14 elements defined by TCM character knowledge graph metadata and 10 elements defined by literature metadata.In Chapter 2,the definition of TCM character knowledge graph relationship,the character relationship is divided into 7 first-level relationships,which are subdivided into 39 second-level relationships,17 third-level relationships and 4 fourth-level relationships.In Chapter 3,on the one hand,defines the data sources as the main data sources of figures,the main data sources of literature,and other supplementary data sources.On the other hand,it is determined that the steps of TCM human knowledge map construction are model construction,data indexing,knowledge fusion,knowledge map generation,inference verification.In Chapter 4,it is clear that Gephi tool software is used to realize visualization,and network features such as degree distribution,average clustering coefficient and average path length are used to represent the topology structure of TCM personal-relationship network,and four network indicators such as PageRank,degree value,intermediate number centrality and feature vector centrality are used to comprehensively evaluate the importance of nodes.Community division was carried out by louvain algorithm.In Chapter 5,603 people related to febrile disease are finally obtained through screening.Then,a human indexing table was established to manually extract the attributes and relationships of information from different sources in the entity and ontology data.The final triplet structure(node-edge-node)was obtained,that is,4183 nodes(entities)and 7828 edges(relationships),and the data was imported into Gephi to visually present the knowledge graph of people related to febrile disease(relationships of people related to febrile disease).On the other hand,it is found that①the network of people related to febrile disease has the characteristics of small world and scale-free network.②The evaluation of the importance of network nodes requires comprehensive consideration of multiple indicators.③It is found that among the top three associations with nodes accounting for the ratio of main nodes,the figures related to the typhoid school and the River School have an important influence on the formation of the Febrile disease school.④ A total of 140,000 relational data were deduced,and the number of simultaneous relationships accounted for 92%of the total,which showed the importance of simultaneous relationships for the construction of a more comprehensive and complete knowledge map of people related to febrill disease.The 3413 erroneous data were verified and the specification and textual research of attribute data were the key to further research on the knowledge map of people related to febrile disease.4 Research ConclusionsConclusion 1.To study the academic inheritance of TCM is actually to study the academic inheritance among TCM figures.Conclusion Two,the character metadata and literature metadata,in fact,is the "symbol" of the character and literature,in line with the general law,but also different.Conclusion 3,related figures and related literature elements are the key elements of integrating figures or documents into the whole TCM academic inheritance.Conclusion 4.The relationship in TCM academic inheritance can be defined as a whole according to the principle of "integrating human and book".Conclusion Five,it is appropriate to construct knowledge map of TCM personages in order to include all TCM personages and TCM documents as much as possible,so as to construct a comprehensive and complete knowledge map of TCM personages.Conclusion 6.In the early stage of the construction of TCM personage knowledge map,the data should be processed manually,and the efficiency and accuracy of data processing should be considered.Conclusion 7:Construct the TCM personage knowledge map,visualize the specific situation of inheritance by graphical means,and describe the general situation of inheritance in a digital way.Conclusion 8.To study the academic inheritance of a family,school or discipline,it is necessary to collect as many relevant figures and literatures as possible,so as to build a more complete human knowledge map.