Research On Methods For Updating And Maintaining Educational Knowledge Graph

Knowledge Graph is a combination of theoretical methods of applied mathematics,graphics,information visualization technology,and information science,and metrology citation analysis methods.It also visualizes the core structure,development history,frontier areas and overall Knowledge architecture achieves the modern theory of multidisciplinary integration.It displays complex knowledge areas through data mining,information processing,knowledge measurement,and graphic rendering,revealing the dynamic development laws in the field of knowledge and providing practical and valuable references for subject research.So far,its practical application has been gradually expanded in developed countries and has achieved good results,but it is still in the initial stage of research in China.As an important branch of knowledge graphs,educational knowledge graphs have obvious differences from other areas of knowledge graphs.Its many professional terms,the complex Chinese and English concepts,and the close connection between knowledge points make it necessary to redefine the ontology framework,optimize the extraction method of entities and attributes,and establish a new update-completion algorithm in the construction of educational knowledge maps.Because the construction process of the knowledge map is not an overnight process,based on the extensibility of the knowledge,the knowledge map needs to be constantly updated and maintained.The updating and maintenance of the Knowledge Graph needs to be gradually established during the construction of the knowledge map.Considering the relationship between entities and entities to be updated,the updating of knowledge maps is generally divided into two levels: the data layer and the data model layer.In the data layer,add and modify the attributes of entities and entities in the map.At the data schema layer,entity deletion,entity prediction completion,and entity relationship completion modification operations are performed.However,the data layer and the data schema layer are not completely isolated.The data schema layer also updates the entity.Therefore,this paper first establishes an update framework for the education knowledge map and determines the updated entity category.According to the general characteristics of educational knowledge maps,"educators","educated persons" and "education measures" are identified as the three main entities of the educational knowledge map.Focus on the relationship between the educated and the education measures,supplement and renew the relationship between knowledge points,determine the weight of knowledge points,and the weight of knowledge is up to the knowledge map of knowledge.Compared with the traditional storage model,show the advantages of the knowledge in the form of “map” and “spectrum”.Secondly,this paper improves the classical algorithm of transcendental knowledge map inference algorithm Trans E.Due to the simplicity and high efficiency of Trans E,it has been respected by the field of knowledge mapping at home and abroad.However,for 1-N,N-1 and N-N types of entity relationship description is not accurate enough,and it is easy to cause ambiguity between entities.Therefore,the improved algorithm Trans EM is used as the main algorithm of the educational knowledge map data model layer inference updating.Finally,the method proposed in this paper is verified by experiments.Word Net's subsets WN11 and WN18 were used in the experiment.And Freebase's subsets FB13,FB15 K datasets,these two datasets are dedicated datasets for verifying the performance of the knowledge map.The structure defines the relationship between entities and entities.From the experimental results,it can be concluded that the improved reasoning update algorithm is superior to the original algorithm in complementing missing entities and tuple classification judgments.This proves the feasibility and superiority of the update completion algorithm for educational knowledge map proposed in this paper.