- 无标题文档
查看论文信息

中文题名:

 面向“互联网+教育”领域的本体构建研究    

姓名:

 鲍婷婷    

保密级别:

 公开    

论文语种:

 中文    

学科代码:

 0401Z2    

学科专业:

 远程教育    

学生类型:

 硕士    

学位:

 教育学硕士    

学位类型:

 学术学位    

学位年度:

 2021    

校区:

 北京校区培养    

学院:

 教育学部    

第一导师姓名:

 陈丽    

第一导师单位:

 北京师范大学教育学部    

提交日期:

 2021-06-17    

答辩日期:

 2021-06-17    

外文题名:

 RESEARCH ON ONTOLOGY CONSTRUCTION IN THE FIELD OF “INTERNET PLUS EDUCATION”    

中文关键词:

 互联网+教育 ; 领域本体 ; 本体构建 ; 概念抽取 ; 关系抽取    

中文摘要:

随着以互联网技术为代表的新兴技术与教育的深度融合,互联网+教育已成为一个备受关注的、快速发展的创新领域,为推动教育变革、解决教育主要矛盾等提供了新的思路与方法。与此同时,互联网+教育的蓬勃发展导致海量信息资源的迅速产生与积累,一方面造成了互联网+教育领域边界与结构模糊不清,一方面带来了信息爆炸与价值利用间的矛盾现状。因此,对互联网+教育领域中信息的高效组织显得尤为重要。本体作为一种知识组织的技术,可以有效实现领域知识的组织和表示,对互联网+教育领域知识的规范、共享、利用等具有重要作用。因此,本文在对现有领域本体构建方法的分析基础上,针对互联网+教育领域数据的非结构化特征,采用半自动化的方法构建互联网+教育领域本体,以系统梳理互联网+教育领域的体系结构,明晰领域内核心概念及概念间的关系,从而加强对互联网+教育结构特征与核心要素的认识,以更好地促进互联网+教育领域理论与实践的健康发展。主要内容如下:

首先,在文献研究的基础上,分析总结了本体构建研究、本体概念获取、本体关系获取等国内外现状,并对相关核心概念进行了界定,进而从本体构成要素、本体构建原则、本体构建方法、本体构建工具等方面系统介绍了本体构建的相关理论与技术。

其次,介绍了互联网+教育本体构建的思路、方法及数据来源。研究共包括现状梳理、语料收集、本体构建、特征分析四个阶段,并从理论研究、政策文件、实践案例三方面收集研究数据,借助文本挖掘技术、归纳法、本体构建方法、内容分析法等方法实现互联网+教育本体构建。

再次,介绍了互联网+教育领域本体的构建过程,主要包括语料预处理、领域概念获取与主题聚类、概念关系获取、本体形式化表达等内容。语料预处理主要指对文本数据进行数据清洗、数据存储、中文分词、词性标注等预处理操作;领域概念的获取主要采用基于规则的方法和基于统计的方法,最终形成互联网+教育领域概念集,并在此基础上通过K-Means层次聚类得到了互联网+教育领域概念模型的九主题;概念关系的获取主要结合Word2Vec和细化度算法获取概念的分类关系,通过归纳法获取概念的非分类关系,最终形成互联网+教育领域概念关系集;本体形式化表达借助Neo4j图数据库技术,实现对互联网+教育领域本体的形式化表达。

最后,从宏观和微观两个层面系统分析互联网+教育本体结构,形成两级宏观-两级微观的体系结构,互联网+教育领域核心要素结构化可视化具体化,为深入认识互联网+教育领域奠定了理论基础。

外文摘要:

With the deep integration of emerging technologies and education represented by Internet technology, "Internet Plus Education" has become a rapidly developing and innovative field of concern. It provides new ideas and methods for promoting educational reform and solving the main contradiction of education. Meanwhile, the vigorous development of "Internet Plus Education" has resulted in the rapid generation and accumulation of massive information resources, which has resulted in blurring the boundaries and structures of "Internet Plus Education" , and on the one hand, it has brought about the contradiction between information explosion and value utilization. Therefore, it is particularly important to organize information efficiently in the field of "Internet Plus Education". As a knowledge organization technology, ontology can effectively realize the organization and representation of domain knowledge, and it is of great importance to the standardization, sharing and utilization of knowledge in the field of "Internet Plus Education". Therefore, "Internet Plus Education" is applied to build the domain ontology based on the analysis of the existing domain ontology construction methods and the semi-structured method. It is used to systematically comb the system structure , clarify the relationship between core concepts and concepts in the core area, and enhance the structural characteristics and core elements of "Internet Plus Education" so that it will help to promote the healthy development of the theory and practice of "Internet Plus Education". The main contents are as follows:

Firstly, on the basis of literature research, this paper analyzes and summarizes the current situation of ontology construction research, ontology concept acquisition, ontology relationship acquisition at home and abroad, defines the relevant core concepts, and then systematically introduces the relevant theories and technologies of ontology construction from the aspects of ontology elements, ontology construction principles, ontology construction methods, ontology construction tools, etc.

Secondly, it introduces the idea, method and data source of the ontology construction of "Internet Plus Education". The study includes four stages: sorting out the current situation, collecting corpus, constructing ontology and analyzing the characteristics. The research data are collected from three aspects of theoretical research, policy documents and practice cases, and the ontology of "Internet Plus Education" is constructed by means of text mining technology, induction, ontology construction and content analysis.

Thirdly, it introduces the construction process of the domain ontology of "Internet Plus Education", which includes corpus preprocessing, domain concept acquisition and topic clustering, concept relation acquisition, ontology formal expression, etc. Corpus preprocessing mainly refers to data cleaning, data storage, Chinese word segmentation, part-of-speech-tagging and other preprocessing operations.The domain concept acquisition mainly adopts the rule based method and the statistical method, and finally forms the concept of "Internet Plus Education". Based on it, nine themes of the "Internet Plus Education" domain conceptual model are obtained through K-Means level clustering. The acquisition of conceptual relations mainly combines the classification relations of concepts with Word2Vec and refinement algorithm, and obtains the non classification relationship of concepts through induction, and finally forms the concept set of "Internet plus education". Ontology formalization is used to express the "Internet Plus Education" domain ontology formally by means of Neo4j diagram database technology.

Finally, the ontology of "Internet Plus Education" is applied to analyze the structure of "Internet Plus Education" from two levels: macro and micro, and build the two level macro two level micro architecture to make the structural characteristics structured, visualized and concretely of "Internet Plus Education", and lay a theoretical foundation for further understanding "Internet Plus Education".

参考文献总数:

 116    

馆藏号:

 硕0401Z2/21008    

开放日期:

 2022-06-17    

无标题文档

   建议浏览器: 谷歌 360请用极速模式,双核浏览器请用极速模式