基于Neo4j的农业专家知识图谱的构建  

The construction of agricultural expert knowledge map based on Neo4j

作　　者：周源 阳兵 李博[1] 李东晖[1] ZHOU Yuan;YANG Bing;LI Bo;LI Donghui(School of Information and Intelligent Science and Technology,Hunan Agricultural University,Changsha 410128,Hunan,China)

机构地区：[1]湖南农业大学信息与智能科学技术学院,湖南长沙410128

出　　处：《农业装备与车辆工程》2025年第2期158-164,共7页Agricultural Equipment & Vehicle Engineering

摘　　要：针对农业领域专家资源分散,缺乏系统化管理,限制了知识的共享与创新的问题,构建了基于Neo4j的农业专家知识图谱,以实现专家信息的结构化表示和多维度关联分析。通过Scrapy爬虫技术、正则表达式和BERT-BiLSTM-CRF深度学习模型,从多源数据中提取专家背景与学术成果;设计知识本体并定义关系规则,构建专家与文献、机构等实体间的关联;利用Neo4j图数据库实现存储与可视化。最终图谱涵盖3类18877个实体和13类94053条关系,支持基于研究方向与合作网络的多维度查询分析,有效揭示专家知识间的内在关联。本研究为农业专家信息的智能化管理提供了新工具,并为未来图谱的自动更新与跨领域扩展奠定了基础。Expert resources in the field of agriculture are scattered and lack of systematic management,which limits the sharing and innovation of knowledge.The knowledge graph of agricultural experts based on Neo4j was constructed to realize the structured representation and multi-dimensional correlation analysis of expert information.By combining scrapy crawler technology,regular expression and BERT-BiLSTM-CRF deep learning model,expert background and academic achievements are extracted from multi-source data.The knowledge ontology was designed and the relationship rules were defined to construct the association between experts and entities such as literature and institutions.Neo4j graph database was used to realize storage and visualization.The final map covers 18877 entities in 3 categories and 94053 relationships in 13 categories.It supports multi-dimensional query analysis based on research direction and cooperation network,and effectively reveals the internal relationship between expert knowledge.This study provides a new tool for the intelligent management of agricultural expert information,and lays a foundation for the automatic update and cross-domain expansion of the future map.

关 键 词：知识图谱 农业学者 Neo4j 知识抽取 

分 类 号：S126[农业科学—农业基础科学] TP391.1[自动化与计算机技术—计算机应用技术]