基于鸟类迁徙关联要素数据的知识图谱构建  

Construction of knowledge graph based on bird migration associated factor data

作　　者：李忠伟[1] 李明轩[1] 李永[1] 张文丰 LI Zhong-wei;LI Ming-xuan;LI Yong;ZHANG Wen-feng(College of Oceanography and Space Informatics,China University of Petroleum(East China),Qingdao 266580,China;Qingdao Institute of Software,College of Computer Science and Technology,China University of Petroleum(East China),Qingdao 266580,China)

机构地区：[1]中国石油大学(华东)海洋空间与信息学院,山东青岛266580 [2]中国石油大学(华东)青岛软件学院、计算机科学与技术学院,山东青岛266580

出　　处：《计算机工程与设计》2024年第12期3704-3711,共8页Computer Engineering and Design

基　　金：自主创新科研计划项目(理工科)-战略专项基金项目(22CX01004A)。

摘　　要：为解决当前鸟类迁徙关联要素数据量繁杂,传统方法不足以高效处理这些数据并精准分析关联要素之间的关系等问题,提出一种鸟类迁徙领域知识图谱构建方法。通过构建本体,利用黄河三角洲生态保护和高质量发展研究院提供的鸟类迁徙数据以及互联网大量文本信息来构建鸟类迁徙实体语料库,设计一种基于RoBERTa-BiLSTM-CRF的鸟类迁徙关联要素实体识别方法进行知识的抽取,利用文本相似度技术进行知识融合,将数据存入图数据库Neo4j中。实验结果表明,所提方法简单高效,构建的知识图谱扩充了鸟类迁徙领域的关联要素信息,是知识图谱技术在生态保护领域的应用与探索。To solve the problems that the current amount of data on the related elements of bird migration is complex,and the traditional methods are not enough to efficiently process these data and accurately analyze the relationship between the related elements,a method for constructing the knowledge graph in the field of bird migration was proposed.The ontology model was designed,and the bird migration entity corpus was constructed using the bird migration data provided by the Yellow River Delta Institute of Ecological Conservation and High-Quality Development and a large amount of text information on the Internet.The entity recognition method of bird migration related elements based on RoBERTa-BiLSTM-CRF was designed for knowledge extraction.The text similarity technology was used for knowledge fusion,and the data were stored in the graph database Neo4j.Experimental results show that the proposed method is simple and efficient,and the constructed knowledge graph expands the related factor information in the field of bird migration,which is the application and exploration of knowledge graph technology in the field of ecological protection.

关 键 词：鸟类迁徙 关联要素 实体识别 关系抽取 知识融合 知识图谱 图数据库 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]