Ma, R., Zheng, M., & Xu, J. (2023). Visualized Evaluation and Intelligent Recommendation of International Chinese MOOCs Based on Learning Data Mining. Journal of Technology and Chinese Language Teaching, 14(2), 1-24.
[马瑞祾, 郑明鉴, & 徐娟. (2023). 基于学习数据挖掘的国际中文慕课评价可视化及智能推荐. 科技与中文教学 (Journal of Technology and Chinese Language Teaching), 14(2), 1-24.]

Full paper

Abstract/摘要：

In the context of digital transformation in education, artificial intelligence is steering international Chinese education toward a direction that emphasizes both scaled education and personalized training. However, the evaluation of scaled international Chinese language education, such as Massive Open Online Courses (MOOCs), from a learners’ perspective remains unclear. This paper, following data-driven approaches, collected and analyzed online comments from L2 learners participating in 51 international Chinese MOOCs on Coursera and Chinese University MOOC, resulting in a total of 10,050 valid comments. Employing a series of advanced statistical and textual analyses, including semantic network analysis, text clustering, topic modeling, and sentiment analysis, the article proposes a quantitative evaluation framework for international Chinese MOOCs from a bottom-up approach. This study also designed and developed the 'International Chinese MOOCs Smart Learning Companion System,' achieving the visual presentation of multi-dimensional evaluation results of MOOCs, as well as providing intelligent and personalized recommendations. Based on the results of data mining, the paper puts forward four suggestions for the construction of high-quality international Chinese MOOCs.

在教育数字化转型的背景下，人工智能技术正推动国际中文教育朝着规模化教育与个性化培养并重的方向发展。然而，如何从学习者的角度来评价规模化的国际中文教育，比如大规模的在线开放课程（MOOCS）仍不清楚。首先，文章在数据驱动的理念下，对“Coursera”和“中国大学MOOC”平台上的51门国际中文慕课在线评论文本进行采集和分析，共得到10050条有效的二语学习者评论文本。其次，文章通过语义网络分析、文本聚类、LDA主题模型、情感分析等一系列地技术操作，“自下而上”地构建了国际中文慕课课程评价量规，设计并研发了“国际中文慕课智慧学伴系统”，实现慕课多维度评价结果的可视化呈现，以及课程的智能化、个性化推荐。最后，文章立足数据挖掘结果，探讨了该研究带来的理论和实践贡献，并为建设高质量地国际中文慕课高质量提出了四条建议。