计算机科学

计算机科学 ›› 2024, Vol. 51 ›› Issue (10): 67-78.doi: 10.11896/jsjkx.240500002

• 智能教育技术及应用 • 上一篇    下一篇

基于行为演化的学习模式识别及效果预测方法

黄春利1, 刘桂梅1, 姜文君1, 李肯立1, 张吉2, 任德盛3   

  1. 1 湖南大学信息科学与工程学院 长沙 410082
    2 南昆士兰大学 昆士兰 澳大利亚 4350
    3 香港中文大学信息工程学系 香港 999077
  • 收稿日期:2024-05-06 修回日期:2024-07-16 出版日期:2024-10-15 发布日期:2024-10-11
  • 通讯作者: 姜文君(jiangwenjun@hnu.edu.cn)
  • 作者简介:(huangchunli@hnu.edu.cn)
  • 基金资助:
    国家自然科学基金(62172149)

Learning Pattern Recognition and Performance Prediction Method Based on Learners'Behavior Evolution

HUANG Chunli1, LIU Guimei1, JIANG Wenjun1, LI Kenli1, ZHANG Ji2, TAK-SHING Peter Yum3   

  1. 1 College of Computer Science and Electronic Engineering,Hunan University,Changsha 410082,China
    2 University of Southern Queensland,Queensland 4350,Australia
    3 Department of Information Engineering,The Chinese University of Hong Kong,Hong Kong 999077,China
  • Received:2024-05-06 Revised:2024-07-16 Online:2024-10-15 Published:2024-10-11
  • About author:HUANG Chunli,born in 1994,Ph.D candidate,is a member of CCF(No.U0106G).Her main research interests include smart education,behavior analysis and learning optimization.
    JIANG Wenjun,born in 1982,Ph.D,professor,Ph.D supervisor,is a senior member of CCF(No.34099S).Her main research interests include smart education,learning optimization,social network analysis and recommendation system.
  • Supported by:
    National Natural Science Foundation of China(62172149).

PDF (PC)

296

摘要： 在线学习为众多学习者提供了开放灵活的学习机会,却存在着学习者学习积极性不高、学习成绩不理想的问题。已有的在线学习效果预测工作着重从静态角度探究学习行为对成绩的影响,忽略了学习行为随时间的演化规律,缺少对行为背后学习模式和学习动机的深入探讨,而这两者正是影响学习效果的重要因素。为此,提出一种基于学习行为演化的学习模式识别及效果预测方法来建模学习行为与动机对学习效果的影响。首先,依据学习者的付出-收获量化学习效率,按时间构建学习效率动态演化序列;然后,使用高斯混合模型聚类真实学习数据并结合实际学习场景,识别4种典型学习模式;在此基础上,设计学习模式及动机预测模型,结合双向长短期记忆网络,构建学习效果预测模型。利用8门真实课程学习的公开数据,对每一种学习模式学习者的付出、收获演变规律进行细致分析。大量对比实验结果表明所提方法在多个性能指标上提升了6.9%～29.2%。本研究有助于在线学习者、教学者和平台准确理解学习者的学习状态,从而提升在线学习效果。

关键词: 在线学习, 行为演化, 学习模式识别, 学习动机预测, 学习效果预测

Abstract: Online learning provides learners with open and flexible learning opportunities,but suffers low learning engagement and unsatisfactory academic performance.Existing works on academic performance prediction mainly study how behaviors will impact performance from a static perspective,and neglect learners' behavior evolution over time and lack a deep understanding of learning patterns and learners' motivations,which are the key factors in learning performance.Therefore,a method of perfor-mance prediction based on learners' learning pattern and motivation is proposed to model the effects of learners' patterns and motivations on their performances.First,we quantify learning efficiency based on learners' efforts and gains and construct the dynamic evolution sequence of learning efficiency with time.Then,we cluster learners' behavior and identify four typical learning patterns combined with the actual learning scenarios.Based on this,learning pattern recognition and motivation prediction mode-ling are designed.The final performance prediction model is constructed by combining them with the bi-directional long-and short-term memory networks.Furthermore,we conduct a detailed and in-depth data analysis on each type of learning patterŃs efforts and gains in eight online courses.Comparative experiments show that the proposed model performs better on several metrics,with improvements ranging from 6.9% to 29.2%.Our work will help online learners,teachers,and platforms accurately understand learners' learning states and improve online learning performance.

Key words: Online learning, Behavior evolution, Learning pattern recognition, Learning motivation prediction, Learning perfor-mance prediction

中图分类号: 

  • TP391
[1]ZHENG Q H,DONG B,QIAN B Y,et al.The state of the art and future tendency of smart education[J].Journal of Computer Research and Development,2019,56(1):209-224.
[2]SU Y,LIU Q W,LIU Q,et al.Exercise-enhanced sequentialmodeling for student performance prediction[C]//Proceedings of the 32th AAAI Conference on Artificial Intelligence.Menlo Park,CA:AAAI,2018:2435-2443.
[3]YUN Y,DAI H,ZHANG Y P,et al.State-of-the-Art survey of personalized learning path recommendation[J].Journal of Software,2021,33(12):4590-4615.
[4]FENG W Z,TANG J,LIU T X.Understanding dropouts inMOOCs[C]//Proceedings of the 33th AAAI Conference on Artificial Intelligence.Menlo Park,CA:AAAI,2019:517-524.
[5]LAN A S,BRINTON C G,YANG T Y,et al.Behavior-based latent variable model for learner engagement[C]//Proceedings of the 10th International Conference on Educational Data Mining.ERIC,2017:64-71.
[6]YAN L S.Higher education psychology[M].Changsha:HunanNormal University Press,2021:153-169.
[7]ZHAO X,ZHU Z W,JAMES C.Rabbit holes and taste distortion:distribution-aware recommendation with evolving interests[C]//Proceedings of the International Conference on World Wide Web.New York:ACM,2021:888-899.
[8]RAMESH A,GOLDWASSER D,HUANG B,et al.Learning latent engagement patterns of students in online courses[C]//Proceedomgs of the 28th AAAI Conference on Artificial Intelligence.Menlo Park,CA:AAAI,2014:1272-1278.
[9]WEN M,ROSÉ C P.Identifying latent study habits by mining learner behavior patterns in massive open online courses[C]//Proceedings of the 23rd ACM International Conference on Conference on Information and Knowledge Management.New York:ACM,2014:1983-1986.
[10]SHI Y L,PENG Z Y,WANG H N.Modeling student learning styles in MOOCs[C]//Proceedings of the 2017 ACM on Confe-rence on Information and Knowledge Management.New York:ACM,2017:979-988.
[11]LI S,WANG S,DU J L,et al.MOOC learners' time-investment patterns and temporal-learning characteristics[J].Journal of Computer Assisted Learning,2022,38(1):152-166.
[12]PENG P,FU W N.A pattern recognition method of personalized adaptive learning in online education[J].Mobile Networks and Applications,2022,27:1186-1198.
[13]ZHANG M,DU X,HUNG J L,et al.Analyzing and interpreting students' self-regulated learning patterns combining time-series feature extraction,segmentation,and clustering[J].Journal of Educational Computing Research,2022,60(5):1130-1165.
[14]YOON M,LEE J,JO I H.Video learning analytics:investigating behavioral patterns and learner clusters in video-based online learning[J].The Internet and Higher Education,2021,50:100806.
[15]MERZDORF H E,DOUGLAS K A.Surveying motivation and learning outcomes of advanced learners in online engineering graduate MOOCs[C]//Proceedings of the 2020 IEEE Frontiers in Education Conference(FIE).Piscataway,NJ:IEEE,2020:1-4.
[16]REYSSIER S,HALLIFAX S,SERNA A,et al.The impact of game elements on learner motivation:influence of initial motivation and player profile[J].IEEE Transaction on Learning Technologies,2022,15(1):42-54.
[17]ORJI F A,VASSILEVA J.Modelling and quantifying learner motivation for adaptive systems:current insight and future perspectives[C]//Proceedings of International Conference on Human-Computer Interaction.Cham:Springer,2021:79-92.
[18]ZAMECNIK A,KOVANOVI V,JOKSIMOVI S,et al.Explo-ring non-traditional learner motivations and characteristics in online learning:A learner profile study[J].Computers and Education:Artificial Intelligence,2022,3:100051.
[19]MAYA-JARIEGO I,HOLGADO D,GONZÁLEZ-TINOCO E,et al.Typology of motivation and learning intentions of users in MOOCs:the MOOCKNOWLEDGE study[J].Educational Technology Research And Development,2020,68(1):203-224.
[20]XU J.A profile analysis of online assignment motivation:Combining achievement goal and expectancy-value perspectives[J].Computers & Education,2022,177:104367.
[21]QIU J Z,TANG J,LIU T X,et al.Modeling and predictinglearning behavior in MOOCs[C]//Proceedings of the 9th ACM International Conference On Web Search and Data Mining.New York:ACM,2016:93-102.
[22]JIANG Z Y,ZHANG Y,LI X M.Learning behavior analysis andprediction based on mooc data[J].Journal of Computer Research and Development,2015,52(3):614-628.
[23]QIU W,SUPRAJA S,KHONG A W H.Toward better grade prediction via A2GP-An academic achievement inspired predictive model[C]//Proceedings of the 15th International Confe-rence on Educational Data Mining(ERIC).2022:195-205.
[24]EL AOUIFI H,EL HAJJI M,ES-SAADY Y,et al.Predicting learner's performance through video sequences viewing behavior analysis using educational data-mining[J].Education and Information Technologies,2021,26(5):5799-5814.
[25]SUN J W,HU M W,LIU S Y,et al.Multi-dimensional behavior characteristics analysis and academic performance prediction in asynchronous online discussion[J].Journal of Chinese Distance Education,2022(5):56-63.
[26]LIU S,LIU S,LIU Z,et al.Automated detection of emotionaland cognitive engagement in MOOC discussions to predict lear-ning achievement[J].Computers & Education,2022,181:104461.
[27]MA Y L,CUI C R,NIE X S,et al.Pre-course student perfor-mance prediction with multi-instance multi-label learning[J].Science China Information Sciences,2019,62(2):029101.
[28]HUANG Q,CHEN J.Enhancing academic performance prediction with temporal graph networks for massive open online courses[J/OL].https://link.springer.com/article/10.1186/s40537-024-00918-5.
[29]ALSHAMAILA Y,ALSAWALQAH H,ALJARAH I.et al.An automatic prediction of students' performance to support the university education system:a deep learning approach[J].Multimed Tools Applications,2024,83(15):46369-46396.
[30]MOUBAYED A,INJADAT M,ALHINDAWI N,et al.A deep learning approach towards student performance prediction in online courses:challenges based on a global perspective[C]//Proceedings of the 24th International Arab Conference on Information Technology(ACIT).Ajman,United Arab Emirates,2023:1-6.
[31]PENG W H,ZENG D W.Research on content analysis of journal papers on online learning behavior in my country in the past ten years[J].Distance Education in China,2015(1):42-48.
[32]RODRIGUESR L,RAMOS J L C,SILVA J C S,et al.Discoveryengagement patterns MOOCs through cluster analysis[J].IEEE Latin America Transactions,2016,14(9):4129-4135.
[33]LEE I S.Analysis of learners' perceptions and learning styles in a web-based environment mixed with a traditional classroom[J].Advanced Research in Computers and Communications in Education,1999,55(1):468-475.
[34]XIA C Y,ZHANG C W,YANG C H,et al.Zero-shot user intent detection via capsule neural networks[C]//Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing.Stroudsburg,PA:ACL,2018:3090-3099.
[35]LIU G M.Research on pattern recognition and performance pre-diction based on learner behavior [D].Changsha:Hunan University,2021.
[36]WU X Z,ZHOU Z H.A unified view of multi-label performance measures[C]//Proceedings of the International Conference on Machine Learning.Australia:PMLR,2017:3780-3788.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市渝北区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发