| 摘 要: 除了自身的学习动机、学习需求等内因,学习效果也受外部因素的影响,学生的在线学习行为对学习效果有直接影响。选取本校所使用的SPOC在线学习平台超星“学银在线”进行研究,收集参与SPOC平台在线学习的学生行为数据,包括观看教学资源时长、登录系统平台次数、频率及成绩等。利用大数据客户关系管理模型中经典的RFM模型,根据项目研究实际需求提出了RFT模型,并基于此模型及在线学习平台中的大量数据,进行SPOC在线学习行为聚类分析。实验对数据集进行清洗、属性规约处理和标准化处理后,使用K-Means聚类算法对学生进行分组,并通过可视化方式对R、F、T指标进行分析,获取各组学生的在线学习行为差异,深度挖掘学生混合式学习需求。最后针对性地提出不同类型学生提升SPOC在线学习效果的策略与建议。 |
| 关键词: SPOC RFT模型 在线学习 K-Means聚类 可视化 |
|
中图分类号: TP391
文献标识码: A
|
|
| Cluster Analysis of Online Learning Behavior Based on RFT Model |
|
WANG Hongxia
|
(School of Information, Media and Art, Beijing Youth Politics College, Beijing 100102, China)
whx4617@163.com
|
| Abstract: In addition to internal factors such as learning motivatin and neeods, learning outcomes are also affected by external factors, and students’online learning behaviors have a direct impact on their academic performance. This study selected the SPOC online learning platform “Xueyin Online”used by the college as the research object. Behavioral data of participating students were collected, including the duration of viewing teaching resources, login frequency, and academic performance. Drawing on the classical RFM model from big data customer relationship management, this paper proposes an improved RFT model tailored to the research needs, and conducts a cluster analysis of students’SPOC learning behaviors based on this model and platform data. After data preprocessing (including cleaning, attribute reduction, and standardization), the K-Means clustering algorithm was used to group students. Visual analysis of the R, F, and T indicators was then performed to identify behavioral differences among student groups and gain insights into their blended learning needs. Finally, targeted strategies and suggestions are put forward to help different types of students improve their SPOC learning effectiveness. |
| Keywords: SPOC RFT model online learning K-Means clustering visualization |
