[1]陈 成,董永权,贾 瑞,等.FKA-DKT:融合知识与能力的深度知识追踪模型[J].南京师大学报(自然科学版),2024,(02):129-139.[doi:10.3969/j.issn.1001-4616.2024.02.015]
　Chen Cheng,Dong Yongquan,Jia Rui,et al.FKA-DKT:Deep Knowledge Tracing Model Based on the Fusion of Knowledge and Ability[J].Journal of Nanjing Normal University(Natural Science Edition),2024,(02):129-139.[doi:10.3969/j.issn.1001-4616.2024.02.015]
点击复制

FKA-DKT:融合知识与能力的深度知识追踪模型()
分享到：

《南京师大学报（自然科学版）》[ISSN:1001-4616/CN:32-1239/N]

卷:
期数:
2024年02期
页码:
129-139
栏目:
计算机科学与技术
出版日期:
2024-06-15

文章信息/Info

Title:
FKA-DKT:Deep Knowledge Tracing Model Based on the Fusion of Knowledge and Ability
文章编号:
1001-4616(2024)02-0129-11
作者:
陈 成1; 董永权1; 2; 3; 贾 瑞1; 刘 源1
(1.江苏师范大学计算机科学与技术学院,江苏 徐州 221116)
(2.江苏省教育信息化工程技术研究中心,江苏 徐州 221116)
(3.徐州市云计算工程技术研究中心,江苏 徐州 221116)
Author(s):
Chen Cheng1; Dong Yongquan1; 2; 3; Jia Rui1; Liu Yuan1
(1.College of Computer Science and Technology,Jiangsu Normal University,Xuzhou 221116,China)
(2.Jiangsu Engineering Technology Research Center of ICT in Education,Xuzhou 221116,China)
(3.Xuzhou Cloud Computing Engineering Technology Research Cente
关键词:
知识追踪; 深度知识追踪; 个人能力建模
Keywords:
knowledge tracing; deep knowledge tracing; personal ability modeling
分类号:
TP391
DOI:
10.3969/j.issn.1001-4616.2024.02.015
文献标志码:
A
摘要:
知识追踪(KT)是智能教育中的一个重要研究问题,其通过分析学生的历史交互来预测其未来的答题情况. 现有的主流KT模型仅根据学生的知识掌握情况对学生进行建模,忽视了学生的个人能力在答题中的作用. 因此,本文提出了一种融合知识和能力的深度知识追踪模型(FKA-DKT). 首先利用DKT模型构建基于知识的答题预测网络(KAPN),从知识层面预测学生答案的正确性. 然后提出基于能力的答案预测(AAPN)网络对学生的能力进行建模,从能力层面预测学生答案的正确性. 最后,将 KAPN 和 AAPN 的预测结果进行线性组合,使模型能够融合知识和能力两个方面的信息来预测学生的作答结果. 在4个公开的数据集上的实验结果表明,相较于现有的主流方法,FKA-DKT在AUC指标上取得了显著的性能提升.
Abstract:
Knowledge tracing(KT)is an important research problem in intelligent education,which predicts students' future answering behaviors by analyzing their historical interactions. Existing mainstream KT models only model students based on their knowledge mastery,neglecting the role of students' personal abilities in answering questions. Therefore,this paper proposes a deep knowledge tracing model(FKA-DKT)that integrates both knowledge and ability. First,we use the DKT model to construct a Knowledge-based Answer Prediction Network(KAPN),which predicts student answer correctness at the knowledge level. Then,we propose an Ability-based Answer Prediction Network(AAPN)to model students' abilities and predict answer correctness at the ability level. Finally,we linearly combine the predictions from KAPN and AAPN to integrate both knowledge and ability information for answer prediction. Experimental results on four publicly available datasets show that compared to existing mainstream methods,FKA-DKT achieves significant performance improvements in terms of the AUC metric.

参考文献/References:

[1]CORBETT 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.
[2]刘铁园,陈威,常亮,等. 基于深度学习的知识追踪研究进展[J]. 计算机研究与发展,2022,59(01):81-104.
[3]PIECH C,BASSEN J,HUANG J,et al. Deep knowledge tracing[J]. Advances in neural information processing systems,2015,28:505-513.
[4]ZHANG J,SHI X,KING I,et al. Dynamic key-value memory networks for knowledge tracing[C/OL]//Proceedings of the 26th International Conference on World Wide Web. Republic and Canton of Geneva,CHE:International World Wide Web Conferences Steering Committee,2017:765-774[2022-02-20]. https://doi.org/10.1145/3038912.3052580.
[5]PANDEY S,KARYPIS G. A self-attentive model for knowledge tracing[J/OL]. ArXiv:1907.06837[cs,Stat],2019[2022-02-21]. http://arxiv.org/abs/1907.06837.
[6]GHOSH A,HEFFERNAN N,LAN A S. Context-aware attentive knowledge tracing[C/OL]//Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. New York,NY,USA:Association for Computing Machinery,2020:2330-2339[2022-02-21]. https://doi.org/10.1145/3394486.3403282.
[7]BAKER R S J D,CORBETT A T,ALEVEN V. More accurate student modeling through contextual estimation of slip and guess probabilities in bayesian knowledge tracing[C]//Intelligent Tutoring Systems. Berlin,Heidelberg:Springer,2008:406-415.
[8]PARDOS Z A,HEFFERNAN N T. KT-idem:introducing item difficulty to the knowledge tracing model[C]//User Modeling,Adaption and Personalization. Berlin,Heidelberg:Springer,2011:243-254.
[9]KÄSER T,KLINGLER S,SCHWING A G,et al. Dynamic bayesian networks for student modeling[J]. IEEE transactions on learning technologies,2017,10(4):450-462.
[10]李晓光,魏思齐,张昕,等. LFKT:学习与遗忘融合的深度知识追踪模型[J]. 软件学报,2021,32(03):818-830.
[11]HOCHREITER S,SCHMIDHUBER J. Long short-term memory[J]. Neural computation,1997,9(8):1735-1780.
[12]CHUNG J,GULCEHRE C,CHO K,et al. Empirical evaluation of gated recurrent neural networks on sequence modeling[J/OL]. ArXiv:1412.3555[Cs],2014[2022-01-28]. http://arxiv.org/abs/1412.3555.
[13]MINN S,YU Y,DESMARAIS M C,et al. Deep knowledge tracing and dynamic student classification for knowledge tracing[C]//2018 IEEE International Conference on Data Mining(ICDM). IEEE,2018:1182-1187.
[14]YEUNG C K,YEUNG D Y. Addressing two problems in deep knowledge tracing via prediction-consistent regularization[Z/OL]. London,United Kingdom:Association for Computing Machinery,2018:1-10(2018). https://doi.org/10.1145/3231644.3231647.
[15]CHOI Y,LEE Y,CHO J,et al. Towards an appropriate query,key,and value computation for knowledge tracing[C/OL]//Proceedings of the Seventh ACM Conference on Learning@Scale. New York,NY,USA:Association for Computing Machinery,2020:341-344[2022-06-09]. https://doi.org/10.1145/3386527.3405945.
[16]ZHOU Y,LI X,CAO Y,et al. LANA:towards personalized deep knowledge tracing through distinguishable interactive sequences:arXiv:2105.06266[Z/OL]. arXiv,2021(2021-04-20)[2023-03-20]. http://arxiv.org/abs/2105.06266.
[17]LIU S,YU J,LI Q,et al. Ability boosted knowledge tracing[J]. Information sciences,2022,596:567-587.
[18]GABRIELLA C,GRILLI L,LIMONE P,et al. Deep learning for knowledge tracing in learning analytics:an overview[C]//CEUR Workshop Proceedings. Foggia,Italy:CEUR-WS,2021:1-10.
[19]WANG W,MA H,ZHAO Y,et al. Tracking knowledge proficiency of students with calibrated q-matrix[J]. Expert systems with applications,2022,192:116454.
[20]VASWANI A,SHAZEER N,PARMAR N,et al. Attention is all you need[J/OL]. ArXiv:1706.03762[Cs],2017[2021-11-26]. http://arxiv.org/abs/1706.03762.
[21]LING C X,HUANG J,ZHANG H. AUC:a statistically consistent and more discriminating measure than accuracy[C]//International Joint Conference on Intelligence. San Francisco,CA,USA:Morgan Kaufmann Publishers Inc,2003:519-524.
[22]KINGMA D P,BA J. Adam:a method for stochastic optimization:arXiv:1412.6980[Z/OL]. arXiv,2017(2017-01-29)[2022-06-24]. http://arxiv.org/abs/1412.6980.
[23]LIU Q,SHEN S,HUANG Z,et al. A survey of knowledge tracing[J/OL]. ArXiv:2105.15106[Cs],2021[2022-02-16]. http://arxiv.org/abs/2105.15106.

备注/Memo

备注/Memo:
收稿日期:2023-06-01.
基金项目:国家自然科学基金面上项目(61872168)、江苏省教育科学十四五规划项目(d/2021/01/112)、江苏师范大学研究生科研与实践创新计划项目(2022XKT1527).
通讯作者:董永权,博士,教授,研究方向:数据集成、数据挖掘、群体智能、教育信息化等. E-mail:tomdyq@jsnu.edu.cn
更新日期/Last Update: 2024-06-15