学生得分预测:一种基于知识图谱的卷积自编码器
|
摘要
在线个性化学习系统能够根据学生的学习历史,为学生提供个性化的学习资源,辅助学生高效学习.要提供精准的个性化诊断报告和个性化资源推荐,首先要对学生进行学业能力评估,其中一个基础性任务为得分预测.对于得分预测任务,现有的研究和方法存在如下不足:①不能充分利用大数据提升预测精度,②无法解决实际应用场景中常见的冷启动问题,③预测结果不可解释.为此提出并实现了一种基于知识图谱的自编码模型(knowledge-aware auto-encoder model,KAEM)用于学生得分预测.首先介绍了含有教育专家先验知识的一种知识图谱,称之为锚题图谱;然后KAEM采用深度学习自编码技术,将教研对锚题图谱的先验理解作为自编码器的正则化项加入模型中,有效地解决冷启动问题.此外,此类模型的预测结果还可以解释化,为实际个性化学习推荐等应用场景提供教研依据.KAEM已经在国内某在线教育系统上运行,取得了良好的效果;在大规模数据上也实验验证了KAEM的有效性.
To reduce study burden and boost efficiency, online education systems offer personalized learning experience for students. In such systems, ability assessment is a fundamental task as reflected by a basic task, named score prediction. The main drawbacks of existing prediction methods are: ① Inability unable to fully exploit the potential of big data, ② cold start problem,③ lack of reasonable explanations. A novel knowledge-aware auto-encoder model(KAEM) is proposed to address these issues. Specifically, an exercise-knowledge-graph with education experts' prior knowledge is introduced. Then students' performance is modeled using auto-encoders with the combination of information in knowledge graph as regularization item. By encoding and integrating the experts' prior knowledge, KAME can improve both prediction accuracy and model robustness and deal with the cold start problem well. Furthermore, reasonable explanations for recommendations can be generated using this model. KMAE has been applied to a famous online education system. Extensive experiments on large-scale real data clearly demonstrate its effectiveness.
To reduce study burden and boost efficiency, online education systems offer personalized learning experience for students. In such systems, ability assessment is a fundamental task as reflected by a basic task, named score prediction. The main drawbacks of existing prediction methods are: ① Inability unable to fully exploit the potential of big data, ② cold start problem,③ lack of reasonable explanations. A novel knowledge-aware auto-encoder model(KAEM) is proposed to address these issues. Specifically, an exercise-knowledge-graph with education experts' prior knowledge is introduced. Then students' performance is modeled using auto-encoders with the combination of information in knowledge graph as regularization item. By encoding and integrating the experts' prior knowledge, KAME can improve both prediction accuracy and model robustness and deal with the cold start problem well. Furthermore, reasonable explanations for recommendations can be generated using this model. KMAE has been applied to a famous online education system. Extensive experiments on large-scale real data clearly demonstrate its effectiveness.
引文
[1]HONG C M,CHEN C M,CHANG M H,et al.Intelligent web-based tutoring system with personalized learning path guidance[C]//IEEE International Conference on Advanced Learning Technologies.Niigata,Japan:IEEE,2007:787-814.
[2]黄振亚,苏喻,吴润泽,等.一种面向教育评估的智能教育辅助平台[J].中国科学技术大学学报,2015,45(10):846-854.HUANG Zhenya,SU Yu,WU Runze,et al.An intelligent tutoring platform for educational assessment[J].Journal of University of Science and Technology of China,2015,45(10):846-854.
[3]ROMERO C,VENTURA S.Educational data mining:a review of the state of the art[J].IEEE Transactions on Systems,Man,and Cybernetics,Part C:Applications and Reviews,2010,40(6):601-618.
[4]朱天宇,黄振亚,陈恩红,等.基于认知诊断的个性化试题推荐方法[J].计算机学报,2017(1):176-191.
[5]KI SOR Y.The state of educational data mining in2009:A review and future visions[J].Computer Communications,2009,6(2):82-87.
[6]R ASCH G.On general laws and the meaning of measurement in psychology[C]//Proceedings of the4th Berkeley Symposium on Mathematical Statistics.Berkeley,USA:IEEE,1961:321-333.
[7]COR BETT A T,ANDERSON J R.Knowledge tracing:Modeling the acquisition of procedural knowledge[J].User Modeling and User-Adapted Interaction,1994,4(4):253-278.
[8]SA LAKHUTDINOV R,MNIH A.Probabilistic matrix factorization[C]//Advances in Neural Information Processing Systems.Vancourver,Canada:Curran Associates Inc.,2007:1257-1264.
[9]TS CHER A,JAHRER M.Collaborative filtering applied to educational data mining[J].Journal of Machine Learning Research,2010:1-11.
[10]PIE CH C,BASSEN J,HUANG J,et al.Deep knowledge tracing[C]//Advances in Neural Information Processing Systems.2015:arXiv:1506.05908.
[11]BERGNER Y,DROSCHLER S,KORTEMEYER G,et al.Model-based collaborative filtering analysis of student response data:Machine-learning item response theory[C]//5th International Conference on Educational Data Mining.Chania,Greece:ERICPress,2012:1-8.
[12]XIE J Y,XU L L,CHEN E H.Image denoising and inpainting with deep neural networks[C]//International Conference on Neural Information Processing System.Lake Tahoe,USA:Curran Associates Inc.,2012:341-349.
[13]WANG H,WANG N,YEUNG D Y.Collaborative deep learning for recommender systems[C]//Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.Sydney,Australia:ACM,2015:1235-1244.
[14]NO VAK J D.Learning,Creating,and Using Knowledge:Concept Maps as Facilitative Tools in Schools and Corporations[M].Taylor&Francis Group,3ed,2010.
[15]ZAHORIAN S A,LAKDAWALA V K,GONZALEZO R,et al.Question model for intelligent questioning systems in engineering education[C]//Frontiers in Education Conference.Reno,USA:IEEE Computer Society,2001:T2B(7-12).
[16]方圆媛.翻转课堂在线支持环境研究---以可汗学院在线平台为例[J].远程教育杂志,2014(6):41-48.
[17]DE VELLIS RF.Classical test theory[J].Medical Care,2006,44(3):50-59.
[18]RUPP A A,TEMPLIN J,HENSON R A.Diagnostic measurement:Theory,methods,and applications[A].Methodology in the Social Sciences[M].New York:Guiford Press,2010:78-79.
[19]GRAVES A,MOHAMED A R,HINTON G.Speech recognition with deep recurrent neural networks[C]//IEEE International Conference on Acoustics,Speech and Signal Processing.Vancouver,Canada:IEEE,2013:6645-6649.
[20]KRIZHEVSKY A,SUTSKEVER I,HINTON G E.ImageNet classification with deep convolutional neural networks[J].Communications of the ACM,2017,60(6):84-90.
[21]M IKOLOV T,SUTSKEVER I,CHEN K,et al.Distributed representations of words and phrases and their compositionality[C]//International Conference on Neural Information Processing Systems.Lake Tahoe,USA:Curran Associates Inc.,2013:3111-3119.
[22]BE NGIO Y,LAMBLIN P,POPOVICI D,et al.Greedy layer-wise training of deep networks[C]//International Conference on Neural Information Processing Systems.British Columbia:MIT Press,2006:153-160.
[23]MASCI J,MEIER U,CIRE爦AN D,et al.Stacked convolutional auto-encoders for hierarchical feature extraction[C]//International Conference on Artificial Neural Networks.Espoo,Finland:Springer-Verlag,2011:52-59.
[24]FUCHS L S,FUCHS D.Effects of expert system consultation within curriculum-based measurement,using a reading maze task[J].Exceptional Children,1992,58(5):436-450.
[25]HUANG Z Y,LIU Q,CHEN E H,et al.Question difficulty prediction for READING problems in standard tests[C]//The 31st AAAI Conference on Artificial Intelligence.San Francisco,USA:AAAIPress,2017:1352-1359.
[26]智学网:http://www.zhixue.com/.
[27]FO GARTY J,BAKER R S,HUDSON S E.Case studies in the use of ROC curve analysis for sensorbased estimates in human computer interaction[C]//Proceedings of Graphics Interface.British Columbia,Canada:DBLP,2005:129-136.
[28]WU R Z,LIU Q,LIU Y P,et al.Cognitive modelling for predicting examinee performance[C]//International Conference on Artificial Intelligence.Buenos Aires,Argentina:AAAI Press,2015:1017-1024.
[29]DIBELLO L V,ROUSSOS L A,STOUT W.31AReview of Cognitively Diagnostic Assessment and a Summary of Psychometric Models[J].Handbook of Statistics,2006,26(6):979-1030.
[2]黄振亚,苏喻,吴润泽,等.一种面向教育评估的智能教育辅助平台[J].中国科学技术大学学报,2015,45(10):846-854.HUANG Zhenya,SU Yu,WU Runze,et al.An intelligent tutoring platform for educational assessment[J].Journal of University of Science and Technology of China,2015,45(10):846-854.
[3]ROMERO C,VENTURA S.Educational data mining:a review of the state of the art[J].IEEE Transactions on Systems,Man,and Cybernetics,Part C:Applications and Reviews,2010,40(6):601-618.
[4]朱天宇,黄振亚,陈恩红,等.基于认知诊断的个性化试题推荐方法[J].计算机学报,2017(1):176-191.
[5]KI SOR Y.The state of educational data mining in2009:A review and future visions[J].Computer Communications,2009,6(2):82-87.
[6]R ASCH G.On general laws and the meaning of measurement in psychology[C]//Proceedings of the4th Berkeley Symposium on Mathematical Statistics.Berkeley,USA:IEEE,1961:321-333.
[7]COR BETT A T,ANDERSON J R.Knowledge tracing:Modeling the acquisition of procedural knowledge[J].User Modeling and User-Adapted Interaction,1994,4(4):253-278.
[8]SA LAKHUTDINOV R,MNIH A.Probabilistic matrix factorization[C]//Advances in Neural Information Processing Systems.Vancourver,Canada:Curran Associates Inc.,2007:1257-1264.
[9]TS CHER A,JAHRER M.Collaborative filtering applied to educational data mining[J].Journal of Machine Learning Research,2010:1-11.
[10]PIE CH C,BASSEN J,HUANG J,et al.Deep knowledge tracing[C]//Advances in Neural Information Processing Systems.2015:arXiv:1506.05908.
[11]BERGNER Y,DROSCHLER S,KORTEMEYER G,et al.Model-based collaborative filtering analysis of student response data:Machine-learning item response theory[C]//5th International Conference on Educational Data Mining.Chania,Greece:ERICPress,2012:1-8.
[12]XIE J Y,XU L L,CHEN E H.Image denoising and inpainting with deep neural networks[C]//International Conference on Neural Information Processing System.Lake Tahoe,USA:Curran Associates Inc.,2012:341-349.
[13]WANG H,WANG N,YEUNG D Y.Collaborative deep learning for recommender systems[C]//Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.Sydney,Australia:ACM,2015:1235-1244.
[14]NO VAK J D.Learning,Creating,and Using Knowledge:Concept Maps as Facilitative Tools in Schools and Corporations[M].Taylor&Francis Group,3ed,2010.
[15]ZAHORIAN S A,LAKDAWALA V K,GONZALEZO R,et al.Question model for intelligent questioning systems in engineering education[C]//Frontiers in Education Conference.Reno,USA:IEEE Computer Society,2001:T2B(7-12).
[16]方圆媛.翻转课堂在线支持环境研究---以可汗学院在线平台为例[J].远程教育杂志,2014(6):41-48.
[17]DE VELLIS RF.Classical test theory[J].Medical Care,2006,44(3):50-59.
[18]RUPP A A,TEMPLIN J,HENSON R A.Diagnostic measurement:Theory,methods,and applications[A].Methodology in the Social Sciences[M].New York:Guiford Press,2010:78-79.
[19]GRAVES A,MOHAMED A R,HINTON G.Speech recognition with deep recurrent neural networks[C]//IEEE International Conference on Acoustics,Speech and Signal Processing.Vancouver,Canada:IEEE,2013:6645-6649.
[20]KRIZHEVSKY A,SUTSKEVER I,HINTON G E.ImageNet classification with deep convolutional neural networks[J].Communications of the ACM,2017,60(6):84-90.
[21]M IKOLOV T,SUTSKEVER I,CHEN K,et al.Distributed representations of words and phrases and their compositionality[C]//International Conference on Neural Information Processing Systems.Lake Tahoe,USA:Curran Associates Inc.,2013:3111-3119.
[22]BE NGIO Y,LAMBLIN P,POPOVICI D,et al.Greedy layer-wise training of deep networks[C]//International Conference on Neural Information Processing Systems.British Columbia:MIT Press,2006:153-160.
[23]MASCI J,MEIER U,CIRE爦AN D,et al.Stacked convolutional auto-encoders for hierarchical feature extraction[C]//International Conference on Artificial Neural Networks.Espoo,Finland:Springer-Verlag,2011:52-59.
[24]FUCHS L S,FUCHS D.Effects of expert system consultation within curriculum-based measurement,using a reading maze task[J].Exceptional Children,1992,58(5):436-450.
[25]HUANG Z Y,LIU Q,CHEN E H,et al.Question difficulty prediction for READING problems in standard tests[C]//The 31st AAAI Conference on Artificial Intelligence.San Francisco,USA:AAAIPress,2017:1352-1359.
[26]智学网:http://www.zhixue.com/.
[27]FO GARTY J,BAKER R S,HUDSON S E.Case studies in the use of ROC curve analysis for sensorbased estimates in human computer interaction[C]//Proceedings of Graphics Interface.British Columbia,Canada:DBLP,2005:129-136.
[28]WU R Z,LIU Q,LIU Y P,et al.Cognitive modelling for predicting examinee performance[C]//International Conference on Artificial Intelligence.Buenos Aires,Argentina:AAAI Press,2015:1017-1024.
[29]DIBELLO L V,ROUSSOS L A,STOUT W.31AReview of Cognitively Diagnostic Assessment and a Summary of Psychometric Models[J].Handbook of Statistics,2006,26(6):979-1030.