元模板在高等教育文本抽取管理语义的应用  

The Application of Meta-Template in Higher Education Text Extraction Management Semantics

作　　者：周斌 王笔扬 刘林[1] 张敏[2] 杨现民[3] ZHOU Bin;WANG Bi-yang;LIU Lin;ZHANG Min;YANG Xian-min(School of Education Science,Jiangsu Normal University,Xuzhou 221116,China;School of Software Engineering,East China Normal University,Shanghai 200062,China;School of Intelligent Education,Jiangsu Normal University,Xuzhou 221116,China)

机构地区：[1]江苏师范大学教育科学学院,江苏徐州221116 [2]华东师范大学软件工程学院,上海200062 [3]江苏师范大学智慧教育学院,江苏徐州221116

出　　处：《数学的实践与认识》2024年第7期158-168,共11页Mathematics in Practice and Theory

基　　金：江苏省社会科学基金(23XZB009);江苏省2024年研究生科研与实践创新计划项目(KYCX24_3003)。

摘　　要：知识图谱是经典的语义抽象(信息抽取)技术,随着数学方法和网络技术等的发展,语义分析成为高等教育文本研究新的路向之一。根据中文语言特点,构建一种开放域关系抽取模板,将经典三元组中的关系、修饰语均抽取为知识实体;实体间的关系设定为预先定义的有限域.进而在领域专业知识实体基础上,使用词向量技术建立文本总体语义倾向的量化体系.最后,根据高等教育领域目标文本构建基于元模板的高校综合改革知识图谱,设计高校综合改革学术权力与行政权力倾向的评分体系,完成对教育部直属高校综合改革方案中学术权力与行政权力的倾向性评价.The knowledge graph is a classic semantic abstract(information extraction)technology.With the development of mathematical methods and network technology,semantic analysis has become a new direction for higher education text research.According to the Chinese language characteristics,an open domain relation extraction template can be constructed.Extract the relationships in the classic triples(Subject-Predicate-Object)and modifiers as knowledge entities;The relationship between entities is a pre-defined finite field.Simultaneously,based on expertise knowledge entities,word vector technology is used to establish a quantification system for the text's overall semantic tendency.Finally,we construct a meta-template-based knowledge graph of comprehensive reforms of colleges and universities according to target text and design a scoring system for its academic power and administrative power.Furthermore,we give out our comments on academic and administration power's tendency concluded from the comprehensive reform plans reported by some of the universities directly under the Ministry of Education.

关 键 词：元模板 知识图谱 高等教育文本 语义推理 

分 类 号：TP391.1[自动化与计算机技术—计算机应用技术]