基于个性化遗忘建模的知识追踪方法

doi:10.19678/j.issn.1000-3428.0069739

计算机工程 ›› 2025, Vol. 51 ›› Issue (8): 120-130. doi: 10.19678/j.issn.1000-3428.0069739

基于个性化遗忘建模的知识追踪方法

张昭理¹, 李家豪¹, 刘海¹^,²^,*(), 石佛波¹, 何嘉文¹

1. 华中师范大学人工智能教育学部, 湖北武汉 430000
2. 华中师范大学深圳研究院, 广东深圳 518000

收稿日期:2024-04-15 修回日期:2024-07-01 出版日期:2025-08-15 发布日期:2025-08-15
通讯作者: 刘海
基金资助:
国家自然科学基金面上项目(6247077114); 国家自然科学基金面上项目(62377037); 国家自然科学基金面上项目(62277041); 国家自然科学基金面上项目(62173286); 国家自然科学基金面上项目(62177019); 国家自然科学基金面上项目(62177018); 湖北省自然科学基金面上项目(2022CFB971); 湖北省自然科学基金面上项目(2022CFB529); 深圳市自然科学基金面上项目(JCYJ20230807152900001); 江西省自然科学基金青年项目(20242BAB2S107); 江西省自然科学基金青年项目(20232BAB212026); 江西省高等学校教育教学改革研究课题(JXJG-23-27-6); 湖北省自然科学基金-创新发展联合基金项目(2025AFD621); 广东省基础与应用基础研究基金(2025A1515010266); 华中师范大学中央高校基本科研业务费项目(CCNU25ai012)

Personalized Forgetting Modeling for Knowledge Tracing via Transformers

ZHANG Zhaoli¹, LI Jiahao¹, LIU Hai¹^,²^,*(), SHI Fobo¹, HE Jiawen¹

1. Faculty of Artificial Intelligence in Education, Central China Normal University, Wuhan 430000, Hubei, China
2. Shenzhen Research Institute of Central China Normal University, Shenzhen 518000, Guangdong, China

Received:2024-04-15 Revised:2024-07-01 Online:2025-08-15 Published:2025-08-15
Contact: LIU Hai

摘要/Abstract

摘要：

针对传统知识追踪(KT)模型难以有效建模学习者在长交互序列下的知识状态变化的问题, 引入以Transformer为代表的注意力机制模型能够实现对学习者长交互序列中潜在信息的捕获, 并展现出良好的性能。然而, 现有模型在建模学习过程时, 往往忽视了学习者的能力差异, 且主要关注知识掌握状态的累加, 未能充分建模学习者的遗忘效益。因此, 提出一种基于个性化遗忘建模的知识追踪方法(PFKT), 通过引入额外的特征信息来建模学习者的答题能力, 并进一步探究学习者差异化的记忆遗忘能力。从学习者的历史交互序列出发, 综合考虑知识点的获取与遗忘现象, 以捕获学习者的真实知识掌握状态。同时, 结合额外的特征信息, 实现更为准确的个性化遗忘现象建模。实验结果显示, 所提出的PFKT模型在ASSISTments2017和Algebra 2005-2006数据集上均取得了较现有模型更优的性能。

关键词: 知识追踪, 注意力机制, 遗忘建模, Transformer, 特征信息

Abstract:

It is very difficult for traditional Knowledge Tracing (KT) models to model learners' knowledge state changes in long interaction sequences. This study introduces an attention mechanism model represented by a Transformer to capture potential information in learners' long interaction sequences that exhibits good performance. However, when modeling the learning process, existing models often ignore the differences in learners' abilities and focus mainly on the accumulation of knowledge mastery states, failing to fully model the forgetting benefit of learners. In this study, a Knowledge Tracing Method based on Personalized Forgetting Modeling (PFKT) is proposed that models learners' answering ability by introducing additional characteristic information and further explores learners' differentiated memory-forgetting ability. Specifically, this method starts with the historical interaction sequence of learners and comprehensively considers the acquisition and forgetting of knowledge points to capture the state of the learners' real knowledge mastery. Simultaneously, combined with additional characteristic information, personalized forgetting phenomenon modeling is realized more accurately. Experimental results demonstrate that the proposed PFKT model achieves better performance than existing models on the ASSISTments2017 and Algebra 2005-2006 datasets.

Key words: Knowledge Tracing (KT), attention mechanism, forgetting modeling, Transformer, feature information

张昭理, 李家豪, 刘海, 石佛波, 何嘉文. 基于个性化遗忘建模的知识追踪方法[J]. 计算机工程, 2025, 51(8): 120-130.

ZHANG Zhaoli, LI Jiahao, LIU Hai, SHI Fobo, HE Jiawen. Personalized Forgetting Modeling for Knowledge Tracing via Transformers[J]. Computer Engineering, 2025, 51(8): 120-130.

https://www.ecice06.com/CN/Y2025/V51/I8/120

图/表 9

图1 PFKT模型整体架构

Fig.1 Overall structure of the PFKT model

图2 不同变体模型在2个数据集上的ACC值和AUC值

Fig.2 ACC and AUC values of different variant models on two datasets

图3 不同变体模型在2个数据集上的MAE值和RMSE值

Fig.3 MAE and RMSE values of different variant models on two datasets

图4 不同数量的编解码层数的结果对比

Fig.4 Comparison of results for different numbers of encoding and decoding layers

图5 学习者的真实答题交互情况可视化展示

Fig.5 Visual display of the interaction of learners' real answers

参考文献 35

1	王志锋, 熊莎莎, 左明章, 等. 智慧教育视域下的知识追踪: 现状、框架及趋势. 远程教育杂志, 2021, 39 (5): 45- 54.
	WANG Z F , XIONG S S , ZUO M Z , et al. Knowledge tracing from the perspective of smart education: current situation, framework and trend. Journal of Distance Education, 2021, 39 (5): 45- 54.
2	PIECH C , BASSEN J , HUANG J , et al. Deep knowledge tracing. Advances in Neural Information Processing Systems, 2015, 28, 505- 513.
3	HOCHREITER S , SCHMIDHUBER J . Long short-term memory. Neural Computation, 1997, 9 (8): 1735- 1780. doi: 10.1162/neco.1997.9.8.1735
4	PANDEY S, KARYPIS G. A self-attentive model for knowledge tracing[EB/OL]. [2024-03-10]. https://arxiv.org/abs/06837.
5	LIU Q , HUANG Z Y , YIN Y , et al. EKT: exercise-aware knowledge tracing for student performance prediction. IEEE Transactions on Knowledge and Data Engineering, 2021, 33 (1): 100- 115. doi: 10.1109/TKDE.2019.2924374
6	CHOI Y, LEE Y, CHO J, et al. Towards an appropriate query, key, and value computation for knowledge tracing[C]//Proceedings of the 17th ACM Conference on Learning @ Scale. New York, USA: ACM Press, 2020: 341-344.
7	MURRAY R C, RITTER S, NIXON T, et al. Revealing the learning in learning curves[M]//LANE H C, YACEF K, MOSTOW J, et al. Artificial intelligence in education. Berlin, Germany: Springer, 2013: 473-482.
8	KYLLONEN P C , TIRRE W C . Individual differences in associative learning and forgetting. Intelligence, 1988, 12 (4): 393- 421. doi: 10.1016/0160-2896(88)90004-9
9	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. 软件学报, 2021, 32 (3): 818- 830.
	LI X G , WEI S Q , ZHANG X , et al. LFKT: a deep knowledge tracking model based on the fusion of learning and forgetting. Journal of Software, 2021, 32 (3): 818- 830.
10	SHEN S H, LIU Q, CHEN E H, et al. Learning process-consistent knowledge tracing[C]//Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. New York, USA: ACM Press, 2021: 1452-1460.
11	SHIN D, SHIM Y, YU H, et al. SAINT+: integrating temporal features for EdNet correctness prediction[C]//Proceedings of the 11th International Learning Analytics and Knowledge Conference. New York, USA: ACM Press, 2021: 490-496.
12	CORBETT A T , ANDERSON J R . Knowledge tracing: modeling the acquisition of procedural knowledge. User Modeling and User-Adapted Interaction, 1994, 4 (4): 253- 278.
13	KHAJAH M M, WING R, LINDSEY R V, et al. Integrating latent-factor and knowledge-tracing models to predict individual differences in learning[C]//Proceedings of the 17th International Conference on Educational Data Mining. Washington D. C., USA: IEEE Press, 2014: 99-106.
14	LEE J I, BRUNSKILL E. The impact on individualizing student models on necessary practice opportunities[EB/OL]. [2024-03-10]. https://cs.stanford.edu/people/ebrun/pdfs/lee2012impact.pdf.
15	EMBRETSON S E , REISE S P . Item response theory. [S. l.]: Psychology Press, 2013.
16	BARTON M A , LORD F M . An upper asymptote for the three-parameter logistic item-response model. ETS Research Report Series, 1981 (1): 1- 8.
17	CEN H, KOEDINGER K, JUNKER B. Comparing two IRT models for conjunctive skills[C]//Proceedings of the 9th International Conference on Intelligent Tutoring Systems. Montreal, Canada: [s. n. ], 2008: 796-798.
18	VIE J J , KASHIMA H . Knowledge tracing machines: factorization machines for knowledge tracing. Proceedings of the AAAI Conference on Artificial Intelligence, 2019, 33 (1): 750- 757.
19	VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need[C]//Proceedings of the 31st International Conference on Neural Information Processing Systems. New York, USA: ACM Press, 2017: 6000-6010.
20	YIN Y, DAI L, HUANG Z Y, et al. Tracing knowledge instead of patterns: stable knowledge tracing with diagnostic transformer[C]//Proceedings of the 2023 ACM Web Conference. New York, USA: ACM Press, 2023: 855-864.
21	NI Q , WEI T J , ZHAO J B , et al. HHSKT: a learner-question interactions based heterogeneous graph neural network model for knowledge tracing. Expert Systems with Applications, 2023, 215, 119334.
22	邵小萌, 张猛. 融合注意力机制的时间卷积知识追踪模型. 计算机应用, 2023, 43 (2): 343- 348.
	SHAO X M , ZHANG M . Temporal convolutional knowledge tracing model with attention mechanism. Journal of Computer Applications, 2023, 43 (2): 343- 348.
23	GHOSH A, HEFFERNAN N, LAN A S. Context-aware attentive knowledge tracing[C]//Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York, USA: ACM Press, 2020: 2330-2339.
24	RASCH G . Probabilistic models for some intelligence and attainment tests. [S. l.]: MESA Press, 1993.
25	ZHOU Y H, LI X H, CAO Y B, et al. LANA: towards personalized deep knowledge tracing through distinguishable interactive sequences[EB/OL]. [2024-03-10]. https://arxiv.org/abs/2105.06266?context=cs.LG.
26	PANDEY S, SRIVASTAVA J. RKT: relation-aware self-attention for knowledge tracing[C]//Proceedings of the 29th ACM International Conference on Information and Knowledge Management. New York, USA: ACM Press, 2020: 1205-1214.
27	SHEN S H, LIU Q, CHEN E H, et al. Convolutional knowledge tracing: modeling individualization in student learning process[C]//Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval. New York, USA: ACM Press, 2020: 1857-1860.
28	LONG T, QIN J R, SHEN J, et al. Improving knowledge tracing with collaborative information[C]//Proceedings of the 15th ACM International Conference on Web Search and Data Mining. New York, USA: ACM Press, 2022: 599-607.
29	ZHAO W Z , XIA J , JIANG X P , et al. A novel framework for deep knowledge tracing via gating-controlled forgetting and learning mechanisms. Information Processing & Management, 2023, 60 (1): 103114.
30	LIU T , ZHANG M , ZHU C , et al. Transformer-based convolutional forgetting knowledge tracking. Scientific Reports, 2023, 13 (1): 19112.
31	NAGATANI K, ZHANG Q, SATO M, et al. Augmenting knowledge tracing by considering forgetting behavior[C]//Proceedings of the World Wide Web Conference. New York, USA: ACM Press, 2019: 3101-3107.
32	CHEN Y Y, LIU Q, HUANG Z Y, et al. Tracking knowledge proficiency of students with educational priors[C]//Proceedings of the 2017 ACM Conference on Information and Knowledge Management. New York, USA: ACM Press, 2017: 989-998.
33	HUANG Z Y , LIU Q , CHEN Y Y , et al. Learning or forgetting? A dynamic approach for tracking the knowledge proficiency of students. ACM Transactions on Information Systems, 2020, 38 (2): 1- 33.
34	ABDELRAHMAN G , WANG Q . Deep graph memory networks for forgetting-robust knowledge tracing. IEEE Transactions on Knowledge and Data Engineering, 2022, 35 (8): 1- 13.
35	ZHANG J N, SHI X J, KING I, et al. Dynamic key-value memory networks for knowledge tracing[EB/OL]. [2024-03-10]. https://arxiv.org/abs/1611.08108v2.

[1]	苗茹, 李祎, 周珂, 张俨娜, 常然然, 孟更. 一种改进的Faster R-CNN遥感图像多目标检测模型研究[J]. 计算机工程, 2025, 51(8): 292-304.
[2]	姬莉霞, 周洪鑫, 肖士杰, 陈允峰, 张晗. 一种基于邻域注意力的扩散模型训练方法研究[J]. 计算机工程, 2025, 51(8): 262-269.
[3]	闫建红, 刘芝妍, 王震. 融合时空注意力机制的多尺度卷积车辆轨迹预测[J]. 计算机工程, 2025, 51(8): 406-414.
[4]	倪源松, 韩军, 邹小燕, 胡广怡, 王文帅. 两阶段自适应分块输电线路螺栓缺陷检测方法[J]. 计算机工程, 2025, 51(8): 281-291.
[5]	郝宏达, 罗健旭. 基于多尺度区域特征融合的多器官语义分割模型[J]. 计算机工程, 2025, 51(8): 270-280.
[6]	彭菊红, 张弛, 高谦, 张光明, 谈栋华, 赵明俊. 基于改进的YOLOv8算法的钢材缺陷检测[J]. 计算机工程, 2025, 51(7): 152-160.
[7]	刘春霞, 孟吉星, 潘理虎, 龚大立. 融合RGB与IR图像的遥感小目标检测方法[J]. 计算机工程, 2025, 51(7): 326-338.
[8]	栾孟娜, 郑秋梅, 王风华. 基于DMC-YOLO的交通标志实时检测算法[J]. 计算机工程, 2025, 51(7): 90-99.
[9]	宋杰, 徐慧英, 朱信忠, 黄晓, 陈晨, 王泽宇. 基于YOLOv8改进的跌倒检测算法: OEF-YOLO[J]. 计算机工程, 2025, 51(7): 127-139.
[10]	单鹏畅, 高利剑, 董文龙, 毛启容. 基于显著目标追踪的行为检测方法[J]. 计算机工程, 2025, 51(6): 93-101.
[11]	赵小虎, 谢礼逊, 慕灯聪, 张悦. 基于TCM-YOLO网络的金属表面缺陷检测方法[J]. 计算机工程, 2025, 51(6): 338-348.
[12]	华家宝, 张京瑞, 朱福民, 陈璐. 基于路侧相机的自适应空间变换车辆检测方法[J]. 计算机工程, 2025, 51(6): 349-359.
[13]	刘凯, 任洪逸, 李蓥, 季怡, 刘纯平. 基于交叉模态注意力特征增强的医学视觉问答[J]. 计算机工程, 2025, 51(6): 49-56.
[14]	张睿, 张雪英, 陈桂军, 黄丽霞. 基于GC特征和脑区频段Transformer模型的EEG情感识别[J]. 计算机工程, 2025, 51(6): 311-319.
[15]	马月坤, 马铭佑. 基于全局与局部特征加权融合的隐喻识别模型[J]. 计算机工程, 2025, 51(5): 143-153.

选择文件类型/文献管理软件名称

选择包含的内容

基于个性化遗忘建模的知识追踪方法

Personalized Forgetting Modeling for Knowledge Tracing via Transformers

RichHTML

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

图/表 9

参考文献 35

相关文章 15

编辑推荐

Metrics

本文评价

模态框（Modal）标题

选择文件类型/文献管理软件名称

选择包含的内容

基于个性化遗忘建模的知识追踪方法

Personalized Forgetting Modeling for Knowledge Tracing via Transformers

RichHTML

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

图/表 9

参考文献 35

相关文章 15

编辑推荐

Metrics

本文评价