标准知识数字化表达通用模型与自动抽取技术研究  

Research on Universal Model of Digital Representation of Standards Knowledge and Automatic Extraction Technology

作　　者：马小雯 孙红军 刘彦林 甘克勤[2] MA Xiao-wen;SUN Hong-jun;LIU Yan-lin;GAN Ke-qin(Zhejiang Lab;China National Institute of Standardization)

机构地区：[1]之江实验室 [2]中国标准化研究院

出　　处：《标准科学》2024年第1期83-87,共5页Standard Science

基　　金：之江实验室开放课题“面向图表识别的标准数字化知识提取标准研究”(项目编号:K2022NH0AB02);中国标准化研究院院长基金项目“数字标准馆标准体系构建及关键标准研制与应用”(252023Y-10411)资助。

摘　　要：标准数字化是国内外标准化发展的重要领域和方向。研究以食品与农产品领域标准为研究对象开展标准知识数字化表达的通用模型与自动抽取技术研究,明确了国内外标准化数字化研究现状与问题,提出了标准知识数字化表达的通用模型,开展了标准知识数字化自动提取技术研究,实现对表达模型的知识要素的自动标注和抽取,并据此形成具有语义关联的标准知识。最后,以2000项食品与农产品领域标准为例进行标准知识数字化表达模型与自动提取技术的实证研究。Standards digitization is an important field and direction of development at home and abroad.This research takes the standards in the field of food and agricultural products as the research object to carry out the research on the universal model of digital expression of standards knowledge and the automatic extraction technology of standards knowledge,defines the current situation and problems of standardization digitization research at home and abroad,puts forward the universal model of digital expression of standards knowledge,and carries out the research on digital automatic extraction technology of standards knowledge.The knowledge elements of the representation model are automatically labeled and extracted,and the standards knowledge with semantic association is formed accordingly.Finally,the digital representation model and automatic extraction technology of standards knowledge are studied by taking 2,000 food and agricultural product standards as examples.

关 键 词：标准知识 数字化 通用模型 自动抽取 语义关联 

分 类 号：TP391.1[自动化与计算机技术—计算机应用技术] F203[自动化与计算机技术—计算机科学与技术]