基于本体库的知识抽取及图谱构建技术  

作　　者：李津[1] 

机构地区：[1]中南林业科技大学涉外学院信息与工程学院,长沙410211

出　　处：《科技创新与应用》2024年第11期37-40,共4页Technology Innovation and Application

摘　　要：该文首先介绍领域内本体库的组成架构,以及基础数据分析和WordNet节选的主要功能。随后,提出一种基于本体库的实体数据抽取技术,在不同的实体之间建立语义关系,为知识抽取做好铺垫。在实体信息抽取过程中,首先要判断网页是否在领域内,在确定网页属于领域后按照特定的标签划分网页内容,进而抽取出有价值的实体数据。将抽取到的实体数据存储到Neo4j数据库中,定期更新知识图谱内的数据。当需要调用数据时,可以从知识图谱中检索需要的数据,从而实现数据资源的整合利用,发挥数据的价值。This paper first introduces the architecture of ontology library in the domain,as well as the main functions of basic data analysis and WordNet excerpt.Then,an entity data extraction technology based on ontology library is proposed to establish semantic relationships between different entities and pave the way for knowledge extraction.In the process of entity information extraction,we should first judge whether the web page is in the domain,divide the web page content according to the specific label after determining whether the web page belongs to the domain,and then extract the valuable entity data.The extracted entity data is stored in Neo4j database,and the data in the knowledge graph is updated regularly.When the data needs to be called,the needed data can be retrieved from the knowledge graph,so as to realize the integrated utilization of data resources and give full play to the value of the data.

关 键 词：本体库 实体数据抽取 Neo4j数据库 数据检索 知识图谱 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]