Construction of Programming Knowledge Graph Based on Student Knowledge Needs

Programming is one of the basic skills necessary for information technology talents. At present, from primary and secondary education to university education, great emphasis is placed on cultivating students' programming abilities. Programming learning involves a large amount of knowledge, diverse language types, and is full of difficulties and challenges. How to help students improve their programming abilities and interest in programming during the learning process is crucial. There is a wealth of programming-related knowledge and code, but it is scattered in every corner of the Internet and books, making it difficult to obtain relevant knowledge. It is a good choice to solve the problems existing in programming education by building a programming knowledge graph. There have been some studies related to programming knowledge graphs, but they did not fully consider the actual needs of students. This article builds a programming knowledge graph based on the analysis of students' actual needs. First, student needs were obtained from multiple data sources and an ontology model was constructed. Then we collected a large amount of semi-structured and unstructured data through crawlers and other methods. For unstructured data, We use the BiLSTM+CRF model to efficiently identify programming entities and relationships. We then store all programming entities and relationships in the Neo4j database. Finally, we used the Cypher query language in Neo4j to evaluate the programming knowledge graph, and the results showed that the programming knowledge graph was effective.