Research On The Construction Method Of Educational Knowledge Graph

With the development and progress of science and technology,the way of obtaining network information has also undergone tremendous changes.Semantic networks have demonstrated a semantic relationship and knowledge in a graphical manner,and solved the problem of incomplete access to traditional channel resource data.As a structured semantic knowledge base,knowledge graph use symbols to describe the physical concepts of the real world.It shows the knowledge system through the network structure of triplet,entity and related attribute structure interconnections.Knowledge Graph have been studied in many foreign countries,but research on domestic knowledge Graph has just started.Most of them are researches on the construction of general knowledge graphs.The construction of knowledge graphs in related fields is mostly concentrated on medicine and social management,etc.The domain knowledge Graph shows the knowledge network structure in related fields.Compared with general knowledge Graph,the knowledge coverage is deeper,the knowledge system is more structured,and the domain features are displayed more completely.Therefore,the domain construction method is different from general knowledge.This paper proposes a method of constructing knowledge Graph in the field of education based on this method.For the proposed method of constructing the knowledge graph in the field of education,this paper first analyzes the characteristics of the knowledge system in the field of education that is different from the general knowledge graph,and provides the precondition for the construction of the next knowledge graph.Then,this paper gives the overall formalized representation of the knowledge graph of this paper and the construction of the knowledge graph.It also gives a formal representation and construction method for the two layers of the logically divided pattern layer and data layer:(1)At the model level,the ontology database is used to manage the upper concept of the knowledge graph,and the structure and hierarchical relationship of the education ontology are proposed.(2)The data layer,using the data extraction method as the ontology concept entity,defines the knowledge of the education domain.Point entity relationships and unique attributes and attribute values.Secondly,the data layer data extraction is completed.For the extraction of entity relationships,a bootstrapping-based extraction method is improved.The semantic constraint relation model is added to reduce the offset problem of extracting relational instances.The form of the package completes the determination of the attributes of the education feature.Finally,for entity alignment problems in multi-source data entity extraction,an alignment method based on attributes and classes is proposed to reduce knowledge redundancy in the knowledge base.Finally,in order to verify the feasibility and validity of the construction method proposed in this paper,the entity relationship extraction method and the entity alignment method in the paper are compared with the current related technologies and methods.Under the condition of ensuring the same experimental environment,data set and evaluation index,the method of entity relationship extraction and entity alignment proposed in this paper is verified,which proves the feasibility and validity of the construction method of education knowledge graph proposed in this paper.