Many regions in China are rich in cultural heritage,with cultural and tourism information spanning interdisciplinary knowledge from the humanities,nature,and archaeology.Relying on traditional tour guides and domain experts is insufficient for comprehensively addressing tourists'cultural and tourism queries,and classic keyword-based retrieval and matching knowledge Q&A systems struggle to manage such extensive and complex knowledge effectively.This presents significant challenges for tourists seeking comprehensive cultural tourism knowledge and enhancind cultural experiences.To tackle this issue,this paper proposes an intelligent integration method for cultural and tourism knowledge based on knowledge graphs and large language models(CTK-KG).This method employs large language models to automatically extract entities and relationships within the cultural and tourism domain,constructing a tightly connected cultural and tourism knowledge graph.Based on this graph,the language model is enhanced with targeted retrieval and generation capabilities to provide more comprehensive and accurate answers to cultural and tourism questions.Using the culturally rich Liangzhu region as an application example,an intelligent Q&A system is developed.Experimental results demonstrate that this method effectively integrates multi-domain cultural and tourism knowledge and provides high-quality answers to cultural and tourism questions.
关键词
大语言模型/知识图谱/文旅知识/智能回复/良渚
Key words
large language model/knowledge graph/cultural and tourism knowledge/intelligent response/Liangzhu
维普
万方数据