留言板

尊敬的读者、作者、审稿人, 关于本刊的投稿、审稿、编辑和出版的任何问题, 您可以本页添加留言。我们将尽快给您答复。谢谢您的支持!

姓名
邮箱
手机号码
标题
留言内容
验证码

基于双路聚类的在线学习行为分析研究

彭涛 单志龙

彭涛, 单志龙. 基于双路聚类的在线学习行为分析研究[J]. 华南师范大学学报(自然科学版), 2021, 53(6): 122-128. doi: 10.6054/j.jscnun.2021102
引用本文: 彭涛, 单志龙. 基于双路聚类的在线学习行为分析研究[J]. 华南师范大学学报(自然科学版), 2021, 53(6): 122-128. doi: 10.6054/j.jscnun.2021102
PENG Tao, SHAN Zhilong. An Analysis of Online Learning Behavior Based on Two-way Clustering[J]. Journal of South China normal University (Natural Science Edition), 2021, 53(6): 122-128. doi: 10.6054/j.jscnun.2021102
Citation: PENG Tao, SHAN Zhilong. An Analysis of Online Learning Behavior Based on Two-way Clustering[J]. Journal of South China normal University (Natural Science Edition), 2021, 53(6): 122-128. doi: 10.6054/j.jscnun.2021102

基于双路聚类的在线学习行为分析研究

doi: 10.6054/j.jscnun.2021102
基金项目: 

国家自然科学基金项目 61671213

广州市科技计划项目 201904010195

详细信息
    通讯作者:

    单志龙,Eamil: ZLshan@m.scnu.edu.cn

  • 中图分类号: TP391.9

An Analysis of Online Learning Behavior Based on Two-way Clustering

  • 摘要: 为有效地利用日志文件,更有深度地刻画学习者画像,提出了双路聚类建模方法(Two-way Clustering, TWC),分析挖掘了万余人次学习者在某大学网络教育学院的大量学习行为数据,力图更深刻地展现远程教育学习者的风貌. 考虑到教育数据具有隐含性这一特点,该方法以细粒度数据为核心,通过双角度的聚类计算得到了各学习者在不同模型中的类别,最后基于融合后的模型对学习者进行刻画. 4种经典聚类算法与TWC算法的对比实验结果和TWC算法的聚类结果表明:TWC算法能够增强簇的内聚性,更准确地对学习者进行聚类,从而更深刻、更全面地刻画学习者轮廓.
  • 图  1  TWC框架图

    Figure  1.  The framework of TWC

    图  2  聚类结果可视化

    Figure  2.  The visualization of clustering results

    图  3  不同时刻各类别学习者的注册比例分布

    Figure  3.  The distribution of registration proportion of all categories of learners at different moments

    图  4  不同时刻各类别学习者的结课比例分布

    Figure  4.  The distribution of the course ending proportion of all categories of learners at different moments

    表  1  网络教育学院数据类型

    Table  1.   The type of data of online education institute

    类别 数据类型 具体实例
    1 视频观看 观看视频、暂停视频等
    2 论坛讨论 参与论坛讨论、发帖等
    3 在线作业 提交作业、查看作业等
    下载: 导出CSV

    表  2  5种算法在同一数据集上的性能指标对比

    Table  2.   The comparison of performance indexes between five algorithms on the same data set

    算法 簇内平方和误差 轮廓系数 运行时间/s
    K-means 789.156 0.707 0.882
    K-means++ 612.485 0.835 1.032
    DPC 560.234 0.881 2.254
    RP-DPC 423.451 0.912 1.148
    TWC 390.754 0.954 1.256
    下载: 导出CSV

    表  3  各类别学习者统计特征概览

    Table  3.   The overview of the statistical attributes of learners in different clusters

    学习者类别 统计特征 PlayCount PlayTime DiscussCount DiscussAmount SAttitude KCount KEntropy KPassPercent
    Cluster1 Min 5.00 653.00 1.00 0.00 0.01 4.00 0.33 0.00
    Mean 66.58 15 470.47 5.47 178.09 1.62 8.07 0.89 0.25
    Max 287.00 46 656.00 16.00 915.00 89.32 19.00 1.00 0.50
    Cluster2 Min 1.00 1.00 1.00 0.00 0.00 1.00 0.00 0.00
    Mean 10.13 1 886.60 5.70 168.40 1.32 1.92 0.29 0.50
    Max 118.00 30 923.00 16.00 883.00 65.52 13.00 1.00 0.96
    Cluster3 Min 1.00 33.00 5.58 4.00 0.02 1.00 0.00 0.00
    Mean 61.34 5 176.30 1.00 174.50 1.80 4.80 0.76 0.16
    Max 293.00 58 673.00 13.00 917.00 78.04 12.00 1.00 0.40
    Cluster4 Min 100.00 24 984.00 1.00 4.00 0.52 12.00 0.73 0.00
    Mean 291.12 66 978.31 6.00 214.04 17.34 21.33 0.95 0.51
    Max 1 543.00 90 495.24 24.00 1 822.00 331.56 27.00 0.99 1.00
    Cluster5 Min 6.00 73.00 5.00 73.00 0.01 4.00 0.43 0.00
    Mean 68.38 16 061.57 14.81 813.64 1.20 8.61 0.88 0.17
    Max 271.00 40 163.00 85.00 4 458.00 155.46 20.00 0.99 0.46
    下载: 导出CSV
  • [1] 马艳云. 新冠疫情下大学生慕课学习研究——基于疫情防控期间与疫情前慕课学习人数的比较[J]. 中国特殊教育, 2020(5): 91-97. https://www.cnki.com.cn/Article/CJFDTOTAL-ZDTJ202005014.htm

    MA Y Y. Study on MOOC learning of college students under the COVID-based on the comparison of the number of MOOC learning during the prevention and control period and before the COVID-19[J]. China Special Education, 2020(5): 91-97. https://www.cnki.com.cn/Article/CJFDTOTAL-ZDTJ202005014.htm
    [2] RAMOS D B, RAMOS I, GASPAEINI I, et al. A new lear-ning path model for e-learning systems[J]. International Journal of Distance Education Technologies, 2021, 19(2): 34-54. doi: 10.4018/IJDET.20210401.oa2
    [3] 李宇帆, 张会福, 刘上力, 等. 教育数据挖掘研究进展[J]. 计算机工程与应用, 2019, 55(14): 15-23. doi: 10.3778/j.issn.1002-8331.1903-0153

    LI Y F, ZHANG H F, LIU S L, et al. Research progress on educational data minging[J]. Computer Engineering and Applications, 2019, 55(14): 15-23. doi: 10.3778/j.issn.1002-8331.1903-0153
    [4] RUIPEREZ-VALIENTE J A, MUNOZ-MERINO P J, DELGADO K C. Detecting and clustering students by their gamification behavior with badges: a case study in engineering education[J]. International Journal of Engineering Education, 2017, 33(2): 816-830. http://eprints.networks.imdea.org/1760/1/post_print_detecting.pdf
    [5] SHOU Z, LU X, WU Z, et al. On learning path planning algorithm based on collaborative analysis of learning behavior[J]. IEEE Access, 2020, 8: 863-879. http://ieeexplore.ieee.org/document/9127933
    [6] KCA B, MPA B, ADCA B, et al. Predicting student dro-pout in subscription-based online learning environments: the beneficial impact of the logit leaf model[J]. Decision Support Systems, 2020, 135: 22-31. http://www.sciencedirect.com/science/article/pii/S0167923620300804
    [7] 吴青, 罗儒国, 王权于. 基于关联规则的网络学习行为实证研究[J]. 现代教育技术, 2015, 25(7): 88-94. doi: 10.3969/j.issn.1009-8097.2015.07.014

    WU Q, LUO R G, WANG Q Y. An empirical study on network learning behavior based on association rules[J]. Modern Educational Technology, 2015, 25(7): 88-94. doi: 10.3969/j.issn.1009-8097.2015.07.014
    [8] WANG Y, JIANG Y L. Research on learning behavior based on ARCS learning motivation model[J]. Advances in Science and Technology, 2021, 105: 356-363. http://www.scientific.net/AST.105.356
    [9] 张莉, 卢星凝, 陆从林, 等. 支持向量机在高考成绩预测分析中的应用[J]. 中国科学技术大学学报, 2017, 47(1): 1-9. https://www.cnki.com.cn/Article/CJFDTOTAL-ZKJD201701001.htm

    ZHANG L, LU X N, LU C L, et al. National matriculation test prediction based on support vector machines[J]. Journal of University of Science and Technology of China, 2017, 47(1): 1-9. https://www.cnki.com.cn/Article/CJFDTOTAL-ZKJD201701001.htm
    [10] WIDYAHASTUTI F, TJHIN V U. Performance prediction in online discussion forum: state-of-the-art and comparative analysis[J]. Procedia Computer Science, 2018, 135: 302-314. doi: 10.1016/j.procs.2018.08.178
    [11] 刘淇, 汪飞, 王新. 教育资源的深度表征与智能应用[J]. 人工智能, 2019(3): 44-54. https://www.cnki.com.cn/Article/CJFDTOTAL-DKJS201903007.htm

    LIU Q, WANG F, WANG X. Deep representation and intelligent application of educational resources[J]. Artificial intelligence, 2019(3): 44-54. https://www.cnki.com.cn/Article/CJFDTOTAL-DKJS201903007.htm
    [12] 陈妍, 张新慧, 朱海萍, 等. 面向网络教育学院的学习行为分析[J]. 计算机应用, 2016, 36(S1): 224-227. https://www.cnki.com.cn/Article/CJFDTOTAL-JSJY2016S1058.htm

    CHENG Y, ZHANG X H, ZHU H P, et al. Analysis of learning behavior for online education colleges[J]. Journal of Computer Applications, 2016, 36(S1): 224-227. https://www.cnki.com.cn/Article/CJFDTOTAL-JSJY2016S1058.htm
    [13] 兰荣亨, 朱格, 杨文, 等. 基于聚类的网络直播群体行为建模分析[J]. 计算机系统应用, 2019, 28(1): 69-74. https://www.cnki.com.cn/Article/CJFDTOTAL-XTYY201901010.htm

    LAN R H, ZHU G, YANG W, et al. Modeling and analysis of community behavior on live streaming platform using clustering approach[J]. Computer Systems & Applications, 2019, 28(1): 69-74. https://www.cnki.com.cn/Article/CJFDTOTAL-XTYY201901010.htm
    [14] 蒋卓轩, 张岩, 李晓明. 基于MOOC数据的学习行为分析与预测[J]. 计算机研究与发展, 2015, 52(3): 614-628. https://www.cnki.com.cn/Article/CJFDTOTAL-JFYZ201503009.htm

    JIANG Z X, ZHANG Y, LI X M. Learning behavior analysis and prediction based on MOOC data[J]. Journal of Computer Research and Development, 2015, 52(3): 614-628. https://www.cnki.com.cn/Article/CJFDTOTAL-JFYZ201503009.htm
    [15] 林鹏飞, 何秀青, 陈甜甜, 等. 深度学习视阈下MOOC学习者流失预测及干预研究[J]. 计算机工程与应用, 2019, 55(22): 258-264. doi: 10.3778/j.issn.1002-8331.1812-0312

    LIN P F, HE X Q, CHEN T T, et al. Research on MOOC learner loss prediction and intervention from the perspective of deep learning[J]. Computer Engineering and Applications, 2019, 55(22): 258-264. doi: 10.3778/j.issn.1002-8331.1812-0312
    [16] 任占广, 尚福华. 基于行为分析的在线课程成绩预测模型[J]. 计算机技术与发展, 2019, 29(11): 139-143. doi: 10.3969/j.issn.1673-629X.2019.11.028

    REN Z G, SHANG F H. An online course grade prediction model based on behavior analysis[J]. Computer Technology and Development, 2019, 29(11): 139-143. doi: 10.3969/j.issn.1673-629X.2019.11.028
    [17] 吴行斌, 郭强, 张林兵, 等. 基于网络社团划分方法的多维数据聚类研究[J]. 计算机应用研究, 2020, 37(2): 421-423. https://www.cnki.com.cn/Article/CJFDTOTAL-JSYJ202002021.htm

    WU X B, GUO Q, ZHANG L B, et al. Multi-dimensional data clustering based on network community partitioning method[J]. Application Research of Computers, 2020, 37(2): 421-423. https://www.cnki.com.cn/Article/CJFDTOTAL-JSYJ202002021.htm
    [18] 张林兵, 郭强, 吴行斌, 等. 基于多维行为分析的用户聚类方法研究[J]. 电子科技大学学报, 2020, 49(2): 315-320. https://www.cnki.com.cn/Article/CJFDTOTAL-DKDX202002022.htm

    ZHANG L B, GUO Q, WU X B, et al. Research on user clustering method based on multi-dimensional behavior analysis[J]. Journal of University of Electronic Science and Technology of China, 2020, 49(2): 315-320. https://www.cnki.com.cn/Article/CJFDTOTAL-DKDX202002022.htm
    [19] 池丽萍, 辛自强. 大学生学习动机的测量及其与自我效能感的关系[J]. 心理发展与教育, 2006, 22(2): 64-70. doi: 10.3969/j.issn.1001-4918.2006.02.012

    CHI L P, XIN Z Q. The measurement of college students' learning motivation and its relationship with self-efficacy[J]. Psychological Development and Education, 2006, 22(2): 64-70. doi: 10.3969/j.issn.1001-4918.2006.02.012
    [20] 傅德胜, 周辰. 基于密度的改进K均值算法及实现[J]. 计算机应用, 2011, 31(2): 432-434. https://www.cnki.com.cn/Article/CJFDTOTAL-JSJY201102042.htm

    FU D S, ZHOU C. Improved K-means algorithm based on density and its implementation[J]. Computer Applications, 2011, 31(2): 432-434. https://www.cnki.com.cn/Article/CJFDTOTAL-JSJY201102042.htm
    [21] JAISWAL R, KUMAR A, SEN S. A simple D2-sampling based PTAS for k-means and other clustering problems[J]. Algorithmica, 2014, 70(1): 22-46. doi: 10.1007/s00453-013-9833-9
    [22] RODRIGUEZ A, LAIO A. Clustering by fast search and find of density peaks[J]. Science, 2014, 344: 1492-1496. doi: 10.1126/science.1242072
    [23] 纪霞, 姚晟, 赵鹏. 相对邻域与剪枝策略优化的密度峰值聚类算法[J]. 自动化学报, 2020, 46(3): 168-181. https://www.cnki.com.cn/Article/CJFDTOTAL-MOTO202003015.htm

    JI X, YAO S, ZHAO P. Density peak clustering algorithm for relative neighbors and pruning strategy optimization[J]. Acta Automatica Sinica, 2020, 46(3): 168-181. https://www.cnki.com.cn/Article/CJFDTOTAL-MOTO202003015.htm
  • 加载中
图(4) / 表(3)
计量
  • 文章访问数:  153
  • HTML全文浏览量:  82
  • PDF下载量:  33
  • 被引次数: 0
出版历程
  • 收稿日期:  2021-05-24
  • 网络出版日期:  2022-01-10
  • 刊出日期:  2021-12-25

目录

    /

    返回文章
    返回