东南园艺2026，Vol.14Issue(3)：296-306,11.DOI:10.20023/j.cnki.2095-5774.2026.03.007
基于BERTopic模型的我国石斛领域研究主题挖掘与热点识别
Research topic mining and hotspot identification of Dendrobium research in China based on the BERTopic model
摘要
Abstract
[Objective]To reveal the distribution of research hotspots and developmental characteristics in the field of Dendrobium research in China,identify the core research forces,and deeply explore research themes,thereby providing references for the conservation of Dendrobium resources,technological innovation,and industrial development.[Method]A bibliometric approach was adopted,combined with the BERTopic model for thematic analysis.Core research forces were identified based on indicators such as publication output and citation frequency of institutions and authors.The BERTopic model,integrating SBERT semantic embedding,UMAP dimensionality reduction,HDBSCAN clustering,and c-TF-IDF topic representation,was applied for topic identification and clustering analysis to construct a knowledge graph of Dendrobium research.[Result]The Dendrobium research field in China has developed a research system dominated by universities and research institutes,with strong associations with major Dendrobium producing regions.However,regional collaboration and interdisciplinary integration remain insufficient.The BERTopic model identified 20 research topics,which were further grouped into four major clusters:clinical disease prevention and treatment,pharmacological activity and molecular mechanisms,breeding cultivation and molecular biology,and component extraction and application.[Conclusion]The field of Dendrobium research in China has established a relatively stable scientific research structure and thematic framework.However,interdisciplinary and cross-institutional collaborative research remains limited.Future studies should strengthen collaborative innovation among research institutions,universities,and pharmaceutical enterprises in major production areas,focusing on deep processing of Dendrobium products,functional product development,and clinical translation research,thereby enhancing research integration and the application value of scientific achievements.关键词
石斛/主题挖掘/BERTopic/知识图谱Key words
Dendrobium/topic mining/BERTopic/knowledge graph分类
农业科技引用本文复制引用
陈洁,吕凯..基于BERTopic模型的我国石斛领域研究主题挖掘与热点识别[J].东南园艺,2026,14(3):296-306,11.基金项目
安徽省农业科学院院立科研计划项目(2026YL064) （2026YL064）
