Research On Teaching Materials Management Model Of Colleges And Related Algorithms Based On Knowledge Graph

Colleges and universities generate a large number of teaching materials during teaching.It is of great significance to collect and manage these teaching materials.However,there are still some problems in Chinese colleges’ existing teaching material management system,which are mainly manifested in three aspects: the failure to effectively organize and integrate the teaching materials,the failure to fully explore the relationship between the teaching materials,and the difficulty for teachers to find the teaching materials that meet their needs.Therefore,this thesis applies the technologies such as knowledge graph and graph database to the management of teaching materials in colleges and universities.Using the excellent performance of knowledge graph and graph database in expressing the relationship between entities,it can better store and process teaching materials in diverse formats,and combines with teaching materials extraction algorithm to explore the inner association between teaching materials to better organize and integrate teaching materials.With the semantic search algorithm of teaching materials,the difficulty in searching teaching materials also can be solved.The main work and results are as follows:The first is to propose a knowledge graph-based ontology model of college teaching materials.Based on the related specification of Chinese teaching resource ontology,the ontology model of college teaching materials is constructed with a seven-step method,and converted into the college teaching materials’ property graph model which is stored into the Neo4 j graph database.It can define the types of entity and the relationships between entities and build up the subject knowledge framework.The second point is to propose a topic extraction algorithm for teaching materials.The text features of teaching material are introduced into the TF-IDF algorithm for improvement.The improved TF-IDF algorithm is used to extract the key knowledge points from the text and experiments are conducted based on the dataset constructed by myself.The experimental results show that the algorithm has improved in performance indexes including recall,precision and F1-Measure,which verifies the accuracy of the algorithm.Through key knowledge points,the association between teaching materials and knowledge points as well as teaching materials is established by associating teaching materials with the constructed college teaching materials’ property graph model.The last is to propose a semantic retrieval algorithm based on the ontology model of college teaching materials.By performing conceptual mapping,semantic extension and semantic reasoning on the keyword set derived from processing search statements,a semantic model that can accurately understand teachers’ search demands is constructed for matching retrieval,which can recommend teaching materials with strong semantic association for teachers and the results are visualized.