Professional Competence Management for University Students Based on Knowledge Graph Technology

Abstract

Effective competence management is beneficial to improve students’ academic performance. Mining the hidden information in teaching data plays an important role in analyzing and modeling professional competence for students. Therefore, we propose a method for constructing a competence management model based on teaching data. By using knowledge graph technology, the model establishes a semantic network which includes seven competence entities and four kinds of relationships between them. To apply and evaluate the competence management method, the model is instantiated by using teaching data from computer science. In the part of named entity recognition, this paper uses the BiLSTM_CNN_CRF model. Compared with the BiLSTM_CRF model, the precision is improved by 1.50%, the recall rate is improved by 1.47% and the F1-score is improved by 1.48%. In the part of entity similarity calculation, in order to help students manage their competence better, the similarity of competence entities and the degree words are calculated respectively. Finally, the Neo4j graphics database is used to store and display the knowledge graph and a competence retrieval service for students is applied in this paper.