University Knowledge Graph Construction Based on Academic Social Network

Yanzhen Yang, Jingsong Leng, Ronghua Lin, Jianguo Li, Feiyi Tang

Abstract: Knowledge graph is an important knowledge representation method in the era of big data. It has become one of the key technologies of artificial intelligence and has been applied in different fields. However, there are relatively few studies on university knowledge graphs combined with academic social networks. Therefore, in this paper, we combine the academic social network SCHOLAT to complete the construction of the university knowledge graph. We first construct the ontology of the knowledge graph, then extract and fuse knowledge from data that come from different sources, and add the output knowledge to the knowledge graph. The university knowledge graph has 191,089 entities and 1,638,275 relationship pairs after the construction is completed, and we store it in the Neo4j database to provide knowledge reserve for subsequent applications. In addition to the construction, we also conduct an application analysis to study its application in university knowledge graph-based Q &A system.