学习分析视角下的个性化预测研究
|
摘要
以学习分析为核心的"数据驱动教学"已成为智能教育时代的重要教学范式。在学习结果预测方面,大多数研究将学习者视为整体进行评估,缺少个性化的分类形式与预测模型,也鲜有较为全面的数据挖掘算法的比较研究。本研究基于中学"互联网+"混合学习场景,基于"大五人格"分类,分析学习行为指标与不同人格特质群体学习结果之间的相关性,利用多元线性回归构建相应的预测模型。通过对28类回归算法与24类分类算法的结果比较,判别具有最佳精度和鲁棒性的预测算法。研究发现,不同人格特质群体的预测变量存在差异,课后测验平均分数出现在所有群体的预测方程中并占有最高的权重;过程性评估成绩、在线学习时长是较为稳健的预测因子;无论在数值预测还是分类预警中,RandomForest算法都具有最佳效能。
Data-driven instruction, chiefly informed by learning analytics, has become an important paradigm in the era of smart education. Most studies about learning outcome prediction take students as a whole,which leads to lack of personalized classification, a personalized prediction model, as well as data mining comparison. This study, drawing from the Internet+ blended learning situations and the Big Five per-sonality traits, analyzed the correlation between learning behaviors and learning outcomes of different personality traits, and constructed a prediction model using multiple linear regression. A prediction algorithm with the most accuracy and robustness was identified through comparing the result of 28 types of regression algorithms and 24 types of classification algorithms. The result shows that there exist prediction variables among different personality traits, with the average score of the after-class test appearing in the prediction equations of all the groups, taking up the most weight; that continuous assessment and online learning time are the two most steady prediction factors; and that random forest algorithm displays the optimal efficiency both in numeric prediction and classification early warning.
Data-driven instruction, chiefly informed by learning analytics, has become an important paradigm in the era of smart education. Most studies about learning outcome prediction take students as a whole,which leads to lack of personalized classification, a personalized prediction model, as well as data mining comparison. This study, drawing from the Internet+ blended learning situations and the Big Five per-sonality traits, analyzed the correlation between learning behaviors and learning outcomes of different personality traits, and constructed a prediction model using multiple linear regression. A prediction algorithm with the most accuracy and robustness was identified through comparing the result of 28 types of regression algorithms and 24 types of classification algorithms. The result shows that there exist prediction variables among different personality traits, with the average score of the after-class test appearing in the prediction equations of all the groups, taking up the most weight; that continuous assessment and online learning time are the two most steady prediction factors; and that random forest algorithm displays the optimal efficiency both in numeric prediction and classification early warning.
引文
陈莉,2005.人格特质、心境对情绪信息加工的影响[D].江西:江西师范大学.
武燕.2012.高低开放性者内隐序列学习进程差异比较研究[D].江苏:苏州大学.
丁梦美,吴敏华,尤佳鑫,等.2017.基于学业成绩预测的教学干预研究[J].中国远程教育(1):50-56.
赵慧琼,姜强,赵蔚,等.2017.基于大数据学习分析的在线学习绩效预警因素及干预对策的实证研究[J].电化教育研究,38(1):62-69.
胡祖辉,施佺.2017.高校学生上网行为分析与数据挖掘研究[J].中国远程教育(02):26-32.
葛子刚,杨丽华,马焕新.2018.割裂还是融合?自适应学习系统设计对于学习风格和认知风格考量的研究现状述评[J].中国远程教育(03):62-69.
李爽,李荣芹,喻忱.2018.基于LMS数据的远程学习者学习投入评测模型[J].开放教育研究,24(01):91-102.
肖巍,倪传斌,李锐.2018.国外基于数据挖掘的学习预警研究:回顾与展望[J].中国远程教育(02):70-78.
张琪,武法提.2018.学习仪表盘个性化设计研究[J].电化教育研究,39(02):39-44,52.
Allport,G.W.(1937).Personality:a psychological interpretation.Ameri?can Journal of Sociology,13(1),48-50.
Nicholls,J.G.(1984).Achievement motivation:conceptions of ability,subjective experience,task choice,and performance.Psychological Review,91(3),328-346.
Jerry M.Burger.(1986)Personality Theory and Research.United States of America:Words Worth Publishing Company.
Costa,P.T.,&McCrae,R.R.(1992).Neo personality inventory-revised(NEO PI-R).Psychological Assessment Resources.
Ravid,G.,Rafaeli,S.(1997).OnLine,Web Based Learning Environ‐ment for an Information Systems course:.Access Logs,Linearity and Performance.In the Proceedings of ISECON 97.
Mischel W.(1999)Instruction to Personality.Sixth edition.Fort Worth:Harcourt Brace College Publishers.
Matthews,G.,Deary,I.J.,Whiteman,M.C.(2003).Personality traits(2nd.ed.).
Kurtz,J.E.,Sherker,J.L.(2003).Relationship quality,trait similarity,and self-other agreement on personality ratings in college roommates.Journal of Personality,71(1),21-48.
Burt,R.S.(2004).Structural holes and good ideas.American Journal of Sociology,110(2),349-399.
Field,A.(2005).Discovering statistics using SPSS(2nd ed.)..Discover?ing statistics using R/.Sage.
Morris,L.V.,Finnegan,C.,Wu,S.S.(2005).Tracking student behavior,persistence,and achievement in online courses.Internet&Higher Ed?ucation,8(3),221-231.
Dollinger,S.J.,Matyja,A.M.,Huber,J.L.(2008).Which factors best account for academic success:those which college students can con‐trol or those they cannot?.Journal of Research in Personality,42(4),72-885.
Carter,M.(2009).Visible learning:a synthesis of over 800 meta‐analy‐ses relating to achievement.Educational Psychology,29(7),867-869.
Richardson,M.,Abraham,C.,Bond,R.(2012).Psychological correlates of university studentsacademic performance:a systematic review and meta-analysis.Psychological Bulletin,138(2),353-387.
Romero,C.,López,M.I.,Luna,J.M.,Ventura,S.(2013).Predicting stu‐dentsfinal performance from participation in on-line discussion fo‐rums.Computers&Education,68,458-472.
Wong,L.(2013).Student Engagement with Online Resources and Its Im‐pact on Learning Outcomes.(Vol.2013,pp.129-146).
Yu,T.,Jo,I.H.(2014).Educational technology approach toward learn‐ing analytics:relationship between student online behavior and learn‐ing performance in higher education.International Conference on Learning Analytics and Knowledge(pp.269-270).
Zacharis,N.Z.(2015).A multivariate approach to predicting student outcomes in web-enabled blended learning courses.Internet&High?er Education,27,44-53.
Pardo,A.,Han,F.,Ellis,R.A.(2017).Combining university student self-regulated learning indicators and engagement with online learn‐ing events to predict academic performance.IEEE Transactions on Learning Technologies,10(1),82-92.
武燕.2012.高低开放性者内隐序列学习进程差异比较研究[D].江苏:苏州大学.
丁梦美,吴敏华,尤佳鑫,等.2017.基于学业成绩预测的教学干预研究[J].中国远程教育(1):50-56.
赵慧琼,姜强,赵蔚,等.2017.基于大数据学习分析的在线学习绩效预警因素及干预对策的实证研究[J].电化教育研究,38(1):62-69.
胡祖辉,施佺.2017.高校学生上网行为分析与数据挖掘研究[J].中国远程教育(02):26-32.
葛子刚,杨丽华,马焕新.2018.割裂还是融合?自适应学习系统设计对于学习风格和认知风格考量的研究现状述评[J].中国远程教育(03):62-69.
李爽,李荣芹,喻忱.2018.基于LMS数据的远程学习者学习投入评测模型[J].开放教育研究,24(01):91-102.
肖巍,倪传斌,李锐.2018.国外基于数据挖掘的学习预警研究:回顾与展望[J].中国远程教育(02):70-78.
张琪,武法提.2018.学习仪表盘个性化设计研究[J].电化教育研究,39(02):39-44,52.
Allport,G.W.(1937).Personality:a psychological interpretation.Ameri?can Journal of Sociology,13(1),48-50.
Nicholls,J.G.(1984).Achievement motivation:conceptions of ability,subjective experience,task choice,and performance.Psychological Review,91(3),328-346.
Jerry M.Burger.(1986)Personality Theory and Research.United States of America:Words Worth Publishing Company.
Costa,P.T.,&McCrae,R.R.(1992).Neo personality inventory-revised(NEO PI-R).Psychological Assessment Resources.
Ravid,G.,Rafaeli,S.(1997).OnLine,Web Based Learning Environ‐ment for an Information Systems course:.Access Logs,Linearity and Performance.In the Proceedings of ISECON 97.
Mischel W.(1999)Instruction to Personality.Sixth edition.Fort Worth:Harcourt Brace College Publishers.
Matthews,G.,Deary,I.J.,Whiteman,M.C.(2003).Personality traits(2nd.ed.).
Kurtz,J.E.,Sherker,J.L.(2003).Relationship quality,trait similarity,and self-other agreement on personality ratings in college roommates.Journal of Personality,71(1),21-48.
Burt,R.S.(2004).Structural holes and good ideas.American Journal of Sociology,110(2),349-399.
Field,A.(2005).Discovering statistics using SPSS(2nd ed.)..Discover?ing statistics using R/.Sage.
Morris,L.V.,Finnegan,C.,Wu,S.S.(2005).Tracking student behavior,persistence,and achievement in online courses.Internet&Higher Ed?ucation,8(3),221-231.
Dollinger,S.J.,Matyja,A.M.,Huber,J.L.(2008).Which factors best account for academic success:those which college students can con‐trol or those they cannot?.Journal of Research in Personality,42(4),72-885.
Carter,M.(2009).Visible learning:a synthesis of over 800 meta‐analy‐ses relating to achievement.Educational Psychology,29(7),867-869.
Richardson,M.,Abraham,C.,Bond,R.(2012).Psychological correlates of university studentsacademic performance:a systematic review and meta-analysis.Psychological Bulletin,138(2),353-387.
Romero,C.,López,M.I.,Luna,J.M.,Ventura,S.(2013).Predicting stu‐dentsfinal performance from participation in on-line discussion fo‐rums.Computers&Education,68,458-472.
Wong,L.(2013).Student Engagement with Online Resources and Its Im‐pact on Learning Outcomes.(Vol.2013,pp.129-146).
Yu,T.,Jo,I.H.(2014).Educational technology approach toward learn‐ing analytics:relationship between student online behavior and learn‐ing performance in higher education.International Conference on Learning Analytics and Knowledge(pp.269-270).
Zacharis,N.Z.(2015).A multivariate approach to predicting student outcomes in web-enabled blended learning courses.Internet&High?er Education,27,44-53.
Pardo,A.,Han,F.,Ellis,R.A.(2017).Combining university student self-regulated learning indicators and engagement with online learn‐ing events to predict academic performance.IEEE Transactions on Learning Technologies,10(1),82-92.