电能计量装置的知识图谱构建与应用

Construction and application of knowledge graph of electric energy metering device

因知识图谱具备强大的实体关系表达推理能力,近年来在各领域得到了广泛应用。文中针对电能计量装置存在标准规范多,建设管理运营成本和门槛高的问题,研究并提出了一种电能计量装置知识图谱的自动构建方法。针对技术规范文档存在的实体嵌套、语义模糊和图谱不全等问题,文中综合采用基于候选实体图匹配的命名实体识别方法,基于多依存句法树融合的关系抽取方法和概念层次图融合的关系推理方法,并在电能计量装置知识图谱构建中进行了验证。实验表明文章构建的知识图谱在实体识别、关系提取和关系推理的准确率以及扩展性方面都取得了较大提升,可大幅度降低相关领域标准规范的应用审查门槛和成本。

Due to the powerful ability of entity relation expression and reasoning,knowledge graph has been widely used in various fields in recent years.Aiming at the problems of many standards and regulations,construction management and operation cost and high threshold for electric energy metering devices,this paper studies and proposes an automatic construction method for the knowledge graph of electric energy metering devices.To solve the problems of nesting entity,semantic fuzziness,and incomplete graph brought by technical specification documents,this paper adopts the named entity recognition method based on candidate entity graph matching,proposes the relationship extraction method based on multidependency syntax tree fusion,and applies the relationship reasoning method based on concept hierarchy graph fusion.The reasoning method is verified in the construction of the knowledge graph of the electric energy metering device.Experiments show that the knowledge map constructed by this method has made a significant improvement in the accuracy and scalability of entity recognition,relationship extraction,and relationship reasoning,which can significantly reduce the application review threshold and cost of standards in related fields.