基于文本挖掘的 ISO 标准术语自动识别与标准术语知识图谱构建研究

摘要：

ISO标准术语蕴含特定的领域知识，是ISO标准文本数据的重要组成。在标准数字化转型下，ISO术语自动识别技术面临迫切的发展需求。本研究通过深入分析ISO标准术语的编写要求，总结了ISO标准术语核心要素的文本特性，基于此采用基于规则的文本挖掘方法构建了ISO标准术语自动识别模型及结构化和可视化加工路径，在ISO 26262标准上完成验证与应用，生成ISO 26262的标准术语知识图谱。本研究的技术路径能够为ISO标准实体抽取和相关标准数字化平台的构建提供一定的参考。

Abstract:

ISO standard terminology contains specific domain knowledge and is an important component of ISO standard text data. In the context of the digital transformation of standards, ISO terminology automatic recognition technology is facing urgent development needs. This study conducted an in-depth analysis of the requirements for writing ISO standard terminology and summarized the text characteristics of the core elements of ISO standard terminology. Based on this, a rule-based text mining method was used to construct an automatic recognition model for ISO standard terminology and a structured and visualization processing path. The model was validated and applied on the ISO 26262 series of standards. The study can provide some reference for the extraction of ISO standard entities and the construction of related standard digital platforms.

基金项目：本文受上海市质量和标准化研究院院立项目“国际标准核心要素标注方法研究”（项目编号：YRY202406）资助。

作者简介：方思怡，硕士研究生，工程师，研究方向为标准数字化、标准数据挖掘、标准知识图谱。