A knowledge graph-based approach for semantic association and intelligent Q&A on the “Fu” cultural symbol system of Quanzhou

Abstract

Quanzhou's unique "Fu" culture constitutes a diverse, complex, and dynamically evolving symbolic system. To systematically explore, organize, and utilize this precious intangible cultural heritage, this study proposes a semantic association and intelligent question-answering method based on knowledge graphs. First, through the collection and fusion of multi-source heterogeneous data , an ontology model centered on "Fu" cultural symbols was constructed, defining core entities, attributes, and their multi-dimensional semantic relationships. Based on this, a "Quanzhou Fu Culture Knowledge Graph" was implemented using the Neo4j graph database, and an intelligent question-answering system integrating semantic parsing and graph query transformation was designed and implemented. The system uses a BERT+BiLSTM+CRF model for named entity recognition, converting natural language questions into standardized Cypher query statements, retrieving and returning structured answers from the graph database. This study not only provides an innovative technological approach for the digital preservation and knowledge service of Quanzhou's "Fu" culture, but also offers a paradigm for the semantic organization and intelligent application of other complex regional cultures.