|本期目录/Table of Contents|

[1]王金羽詹逸思 冯 起李曼丽.教育质性研究中人机协同文本挖掘技术的运用——以某高校教学评估中文文本数据为例[J].清华大学教育研究,2022,(02):56-63.
 WHAN Jin-yu?ZHAN Yi-si FENG Qi?LI Man-li.Application of Human-Computer Collaboration Text Mining Technology in Educational Qualitative Research——Based on Chinese Text Data of Teaching Evaluation in a University[J].TSINGHUA JOURNAL OF EDUCATION,2022,(02):56-63.
点击复制

教育质性研究中人机协同文本挖掘技术的运用——以某高校教学评估中文文本数据为例
分享到:

清华大学教育研究[ISSN:1001-4519/CN:11-1610/G4]

卷:
期数:
2022年02期
页码:
56-63
栏目:
教育思想与理论
出版日期:
2022-04-20

文章信息/Info

Title:
Application of Human-Computer Collaboration Text Mining Technology in Educational Qualitative Research——Based on Chinese Text Data of Teaching Evaluation in a University
作者:
王金羽1詹逸思2 冯 起3李曼丽1
1.清华大学 教育研究院;2.清华大学 学生学习与发展指导中心;3.清华大学 电机工程与应用电子技术系
Author(s):
WHAN Jin-yu1?ZHAN Yi-si2 FENG Qi3?LI Man-li1
1.Institute?of?Education,?Tsinghua?University;2.Then Center for Student Learning and Development, Tsinghua University;3.Department of Electrical Engineering, Tsinghua University
关键词:
结构主题模型(STM)超大文本挖掘教育质性研究
Keywords:
Structural Topic Model (STM) massive text data mining educational qualitative research
分类号:
G40-034
文献标志码:
A
摘要:
信息时代海量增长的文本资料成为质性研究者开展研究的数据宝藏,但未得到充分研究,其原因在于针对海量中文文本数据的有效分析方法尚待突破。文章率先在质性研究范式中使用了以结构主题模型(STM)为代表的人机协同方法,对某大学在线教学效果评估的课堂观察记录数据展开文本挖掘。以教学评估研究数据分析为例,完整呈现了在教育质性研究中应用STM进行数据挖掘的四个步骤,并分析了其在挖掘海量中文文本资料方面的独特优势。研究表明,跨学科研究方法的尝试有助于解决教育学科甚至人文社科领域内海量中文文本在质性分析上的固有难题。
Abstract:
Although sharp growing massive text data has become a treasure for qualitative researchers, it has not been fully studied because few effective analysis methods for massive Chinese text data have been created. This research initiated the use of a human-computer collaboration method represented by the structural topic model (STM) in the R language in the educational qualitative research paradigm. By mining classroom observation record data of a university’s online teaching effect evaluation, this paper presents the four steps of applying the STM model to data mining in the qualitative research of education and analyzes the strengths and weaknesses of the R language in mining massive Chinese text data. Studies have shown that interdisciplinary research methods can help overcome the inherent challenges in the qualitative analysis of massive Chinese texts in educational research and even beyond the humanities and social sciences.
更新日期/Last Update: 2022-04-20