基于学习者评论数据挖掘的MOOC课程质量影响因素研究  

Research on the Influencing Factors of MOOC Course Quality Based on Learner Review Data Mining

作　　者：刘清堂[1] 尹兴翰 吴林静[1] 曹天生 陈亮[1] Liu Qingtang;Yin Xinghan;Wu Linjing;Cao Tiansheng;Chen Liang(Faculty of Artificial Intelligence In Education,Central China Normal University,Wuhan Hubei 430079)

机构地区：[1]华中师范大学人工智能教育学部,湖北武汉430079

出　　处：《远程教育杂志》2023年第1期80-90,共11页Journal of Distance Education

基　　金：国家自然科学基金项目“数据驱动的在线学习协作会话过程监测与干预机制研究”(项目编号:72174070);华中师范大学信息化与基础教育均衡发展省部共建协同创新中心重点项目“中小学教师信息化教学胜任力指标构建及智能测评”(项目编号:xtzd2021-008);中央高校基本科研业务费资助(项目编号:CCNU22JC011)的研究成果。

摘　　要：理解学习者MOOC课程学习的体验和需求,是促进MOOC高质量可持续发展的重要途径。基于学习者评论数据挖掘,可以揭示学习者情感体验与课程质量因素之间的关系,并为分析不同学科课程的差异提供有效支持,弥补现有研究的不足。为此,开展了基于学习者视角的MOOC课程质量影响因素研究,首先以文本挖掘技术为基础,客观地从MOOC课程学习者评论数据中提取出课程管理、课程设计、学习平台、学习任务、学习材料、课程教师、课程内容及学习体验八个课程质量影响因素。其次,构建了基于学习者满意度和关注度的KANO分类模型,以评估课程质量影响因素的重要性程度,并对两类课程进行差异性分析和对影响因素之间进行相关性分析。结果表明:学习者对人文社科类和自然课程类课程的评价在课程设计、学习材料、课程教师、课程内容及学习体验上存在显著差异;而学习任务在两类课程中的评价均为最低;并且通过可视化揭示了不同类别课程的KANO模型分类结果和影响因素之间的关系。最后,针对不同学科课程的特点,提出课程优化建议,对MOOC课程质量的改进与提升具有重要指导意义。Promoting the high-quality and long-term development of MOOC requires an understanding of learners’ needs and experiences with online learning. Data mining based on learner reviews can reveal the relationship between learners’ affective experiences and course quality factors, and provide effective support for analyzing differences in courses across disciplines, bridging the gaps in existing research. In this paper, we objectively collected eight-course quality influencing factors from the perspective of learners, including course management, course design, learning platform, learning tasks, learning materials, course teachers, course content,and learning experience. Secondly, in order to assess the degree of importance of the factors influencing course quality, we constructed a KANO classification model based on learner satisfaction and attention, and conducted a difference analysis and correlation analysis between the influencing factors for both types of courses. The results show that: learners’ evaluations of humanities and social science and nature courses differ significantly in terms of course design, learning materials, course teachers, course content and learning experience;while learning tasks are rated the lowest in both categories;and we reveal the relationship between the KANO model classification results and influencing factors for different categories of courses through visualization. Finally, we propose course optimization suggestions for the characteristics of different subject courses, which are important for the improvement and enhancement of MOOC course quality.

关 键 词：MOOC 课程质量影响因素 KANO模型 在线评论分析 质量提升 

分 类 号：G420[文化科学—课程与教学论]