Knowledge tracking model based on concept association memory network with graph attention

doi:10.11772/j.issn.1001-9081.2025010065

Journal of Computer Applications ›› 2026, Vol. 46 ›› Issue (1): 43-51.DOI: 10.11772/j.issn.1001-9081.2025010065

• Artificial intelligence • Previous Articles Next Articles

Knowledge tracking model based on concept association memory network with graph attention

Fan HE¹^,², Li LI¹^,²(), Zhongxu YUAN¹^,², Xiu YANG¹^,², Dongxuan HAN¹^,²^,³

^1.School of Computer Science and Technology，Southwest University of Science and Technology，Mianyang Sichuan 621010，China
^2.Sichuan Industrial Autonomous and Controllable Artificial Intelligence Engineering Technology Research Center，Mianyang Sichuan 621010，China
^3.State Key Laboratory of Cognitive Intelligence （University of Science and Technology of China），Hefei Anhui 230088，China

Received:2025-01-15 Revised:2025-03-27 Accepted:2025-03-27 Online:2026-01-10 Published:2026-01-10
Contact: Li LI
About author:HE Fan， born in 2001， M. S. candidate. His research interests include smart education， knowledge tracing.
YUAN Zhongxu， born in 2001， M. S. candidate. His research interests include natural language processing， large model fine-tuning.
YANG Xiu， born in 2001， M. S. candidate. Her research interests include smart healthcare.
HAN Dongxuan， born in 2001， Ph. D. candidate. His research interests include smart education， cognitive diagnosis.
Supported by:
Key Support Program of National Natural Science Foundation of China(U21A20157)

融合图注意力的概念关联记忆网络知识追踪模型

何凡¹^,², 李理¹^,²(), 苑中旭¹^,², 杨秀¹^,², 韩东轩¹^,²^,³

^1.西南科技大学计算机科学与技术学院，四川绵阳 621010
^2.四川省工业自主可控人工智能工程技术研究中心，四川绵阳 621010
^3.认知智能全国重点实验室（中国科学技术大学），安徽合肥 230088

通讯作者: 李理
作者简介:何凡（2001—），男，四川绵阳人，硕士研究生， CCF会员，主要研究方向：智慧教育、知识追踪
苑中旭（2001—），男，河南驻马店人，硕士研究生， CCF会员，主要研究方向：自然语言处理、大模型微调
杨秀（2001—），女，四川成都人，硕士研究生， CCF会员，主要研究方向：智慧医疗
韩东轩（2001—），男，四川绵阳人，博士研究生，主要研究方向：智慧教育、认知诊断。
基金资助:
国家自然科学基金重点支持项目(U21A20157)

Abstract

Abstract:

Tracking students' historical interactions to predict their future performance is a critical research focus in the field of Knowledge Tracing （KT）. Recent KT methods aim to explore students' learning patterns and evolving knowledge states to provide personalized learning guidance， but ignore the richness of exercises themselves. Additionally， with the emergence of new disciplines and interdisciplinary fields， Graph Neural Network （GNN） -based KT methods face challenges in the issues such as broadening the scope of concept associations and modeling students' learning behaviors. To address these challenges， a novel Knowledge Tracing （KT） model was proposed， termed the Knowledge Tracking model based on concept association Memory network with Graph Attention （GAMKT）. The GAMKT is capable of modeling students' exercise interaction sequences， tracking their knowledge states， and capturing global features of related concepts from the exercise-concept graph. Moreover， a forgetting gate and higher-order information extraction were incorporated into the model to realistically simulate students' exercise-solving processes. Experimental results on the Junyi， ASSIST09， and Static2011 datasets demonstrate that， compared with seven baseline models including Graph-based Knowledge Tracing （GKT）， GAMKT achieves average improvements of approximately 2.1% in Area Under the Curve （AUC） and 2.4% in Accuracy， indicating that GAMKT outperforms the baseline methods on datasets with well-structured knowledge.

Key words: Knowledge Tracing (KT), Graph Neural Network (GNN), self-attention, concept association, forgetting mechanism

摘要：

追踪学生的历史互动以预测他们的未来表现是知识追踪（KT）领域的关键研究重点。最近的KT方法旨在研究学生的学习模式和不断变化的知识状态，以提供个性化学习指导，而忽略了习题本身的丰富性。此外，随着新专业和跨学科领域的兴起，基于图神经网络（GNN）的KT方法面临着一些挑战：如何扩展概念之间关联的视野并建模学生的学习行为。为了应对这些挑战，提出一种新的KT模型——融合图注意力的概念关联记忆网络知识追踪模型（GAMKT）。GAMKT可以建模学生的习题记录，追踪他们的知识状态，并从习题-概念图中捕捉相关概念的全局特征；同时，将遗忘门和高阶信息提取纳入模型，以真实模拟学生的习题过程。在Junyi、ASSIST09和Static2011数据集上的对比实验结果表明，与基于图神经网络的知识追踪（GKT）等7种模型相比，GAMKT的曲线下面积（AUC）和准确率分别平均至少提升了约2.1%和2.4%，表明GAMKT在成熟的知识结构数据集上优于基线模型。

关键词: 知识追踪, 图神经网络, 自注意力, 概念关联, 遗忘机制

CLC Number:

TP391

Fan HE, Li LI, Zhongxu YUAN, Xiu YANG, Dongxuan HAN. Knowledge tracking model based on concept association memory network with graph attention[J]. Journal of Computer Applications, 2026, 46(1): 43-51.

何凡, 李理, 苑中旭, 杨秀, 韩东轩. 融合图注意力的概念关联记忆网络知识追踪模型[J]. 《计算机应用》唯一官方网站, 2026, 46(1): 43-51.

Figures/Tables 11

References 37

[1]	赵元棣，刘永欣，于槐松.智慧学习环境下的在线学习行为分析［J］.科技风， 2025（3）： 163-165.
	ZHAO Y D， LIU Y X， YU H S. Online learning behavior analysis in intelligent learning environment ［J］. Technology Wind， 2025（3）： 163-165.
[2]	HUO Y， WONG D F， NI L M， et al. Knowledge modeling via contextualized representations for LSTM-based personalized exercise recommendation ［J］. Information Sciences， 2020， 523： 266-278.
[3]	刘坤佳，李欣奕，唐九阳，等.可解释深度知识追踪模型［J］.计算机研究与发展， 2021， 58（12）： 2618-2629.
	LIU K J， LI X Y， TANG J Y， et al. Interpretable deep knowledge tracing ［J］. Journal of Computer Research and Development， 2021， 58（12）： 2618-2629.
[4]	CHEN P， LU Y， ZHENG V W， et al. Prerequisite-driven deep knowledge tracing ［C］// Proceedings of the 2018 IEEE International Conference on Data Mining. Piscataway： IEEE， 2018： 39-48.
[5]	熊余，张健，王盈，等.基于深度学习的演化知识追踪模型［J］.电化教育研究， 2022， 43（11）： 23-30.
	XIONG Y， ZHANG J， WANG Y， et al. An evolutionary knowledge tracing model based on deep learning ［J］. e-Education Research， 2022， 43（11）： 23-30.
[6]	SONG X， LI J， LEI Q， et al. Bi-CLKT： bi-graph contrastive learning based knowledge tracing ［J］. Knowledge-Based Systems， 2022， 241： No.108274.
[7]	YIN Y， DAI L， HUANG Z， et al. Tracing knowledge instead of patterns： stable knowledge tracing with diagnostic Transformer ［C］// Proceedings of the ACM Web Conference 2023. New York： ACM， 2023： 855-864.
[8]	PARK S， LEE D， PARK H. Enhancing knowledge tracing with concept map and response disentanglement ［J］. Knowledge-Based Systems， 2024， 302： No.112346.
[9]	HAMILTON W L， YING Z， LESKOVEC J. Inductive representation learning on large graphs ［C］// Proceedings of the 31st International Conference on Neural Information Processing Systems. Red Hook： Curran Associates Inc.， 2017： 1025-1035.
[10]	ZHANG H， LU G， ZHAN M， et al. Semi-supervised classification of graph convolutional networks with Laplacian rank constraints ［J］. Neural Processing Letters， 2022， 54（4）： 2645-2656.
[11]	BECK M， PÖPPEL K， SPANRING M， et al. xLSTM： extended long short-term memory ［C］// Proceedings of the 38th International Conference on Neural Information Processing Systems. Red Hook： Curran Associates Inc.， 2024： 107547-107603.
[12]	ABDELRAHMAN G， WANG Q. Deep graph memory networks for forgetting-robust knowledge tracing ［J］. IEEE Transactions on Knowledge and Data Engineering， 2023， 35（8）： 7844-7855.
[13]	郑浩东，马华，谢颖超，等.融合遗忘因素与记忆门的图神经网络知识追踪模型［J］.计算机应用， 2023， 43（9）： 2747-2752.
	ZHENG H D， MA H， XIE Y C， et al. A graph neural network knowledge tracing model incorporating forgetting factors and memory gates ［J］. Journal of Computer Applications， 2023， 43（9）： 2747-2752.
[14]	CHEN Z， SHAN Z， ZENG Y. Informative representations for forgetting-robust knowledge tracing ［J］. User Modeling and User-Adapted Interaction， 2024， 34（4）： 1227-1249.
[15]	DWIVEDI V P， BRESSON X. A generalization of Transformer networks to graphs ［EB/OL］. ［2021-06-24］. .
[16]	LeCUN Y， BENGIO Y， HINTON G. Deep learning ［J］. Nature， 2015， 521（7553）： 436-444.
[17]	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.
[18]	KHAJAH M， WING R W， LINDSEY R V， et al. Integrating latent-factor and knowledge-tracing models to predict individual differences in learning ［EB/OL］. ［2024-06-30］. .
[19]	PIECH C， BASSEN J， HUANG J， et al. Deep knowledge tracing ［C］// Proceedings of the 29th International Conference on Neural Information Processing Systems， Volume 1. Cambridge： MIT Press， 2015： 505-513.
[20]	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： ACM， 2020： 1205-1214.
[21]	ZHANG J， SHI X， KING I， et al. Dynamic key-value memory networks for knowledge tracing ［C］// Proceedings of the 26th International Conference on World Wide Web. Republic and Canton of Geneva： International World Wide Web Conferences Steering Committee， 2017： 765-774.
[22]	PANDEY S， KARYPIS G. A self-attentive model for knowledge tracing ［EB/OL］. ［2024-07-16］. .
[23]	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： ACM， 2021： 490-496.
[24]	LU Y， WANG D， MENG Q， et al. Towards interpretable deep learning models for knowledge tracing ［C］// Proceedings of the 2020 International Conference Artificial Intelligence in Education， LNCS 12164. Cham： Springer， 2020： 185-190.
[25]	LIU S， YU J， LI Q， et al. Ability boosted knowledge tracing ［J］. Information Sciences， 2022， 596： 567-587.
[26]	CUI J， CHEN Z， ZHOU A， et al. Fine-grained interaction modeling with multi-relational transformer for knowledge tracing ［J］. ACM Transactions on Information Systems， 2023， 41（4）： No.104.
[27]	NAKAGAWA H， IWASAWA Y， MATSUO Y. Graph-based knowledge tracing： modeling student proficiency using graph neural network ［C］// Proceedings of the 2019 International Conference on Web Intelligence. Piscataway： IEEE， 2019： 156-163.
[28]	YANG Y， SHEN J， QU Y， et al. GIKT： a graph-based interaction model for knowledge tracing ［C］// Proceedings of the 2020 European Conference on Machine Learning and Knowledge Discovery in Databases， LNCS 12457. Cham： Springer， 2021： 299-315.
[29]	ZHAO Z， LIU Z， WANG B， et al. Research on deep knowledge tracing model integrating graph attention network ［C］// Proceedings of the 2022 Conference on Prognostics and Health Management. Piscataway： IEEE， 2022： 389-394.
[30]	WU Z， HUANG L， HUANG Q， et al. SGKT： session graph-based knowledge tracing for student performance prediction ［J］. Expert Systems with Applications， 2022， 206： No.117681.
[31]	CUI C， YAO Y， ZHANG C， et al. DGEKT： a dual graph ensemble learning method for knowledge tracing ［J］. ACM Transactions on Information Systems， 2024， 42（3）： No.78.
[32]	VELIČKOVIĆ P， CUCURULL G， CASANOVA A， et al. Graph attention networks ［EB/OL］. ［2025-02-04］. .
[33]	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. Red Hook： Curran Associates Inc.， 2017： 6000-6010.
[34]	CHANG H S， HSU H J， CHEN K T. Modeling exercise relationships in e-learning： a unified approach ［EB/OL］. ［2024-06-30］. .
[35]	FENG M， HEFFERNAN N， KOEDINGER K. Addressing the assessment challenge with an online system that tutors as it assesses ［J］. User Modeling and User-Adapted Interaction， 2009， 19（3）： 243-266.
[36]	KOEDINGER K R， BAKER R S J， CUNNINGHAM K， et al. A data repository for the EDM community： the PSLC DataShop ［M］// ROMERO C， VENTURA S， PECHENIZKIY M， et al. Handbook of educational data mining. Boca Raton， FL： CRC Press， 2010： 43-56.
[37]	KINGMA D P， BA J L. Adam： a method for stochastic optimization ［EB/OL］. ［2024-06-30］. .

数据集	学生数	问题数	概念数	总交互数	平均交互数
Junyi	10 000	568	399	136 163	13
ASSIST09	4 217	26 688	123	401 756	95
Static2011	333	300	1 223	194 947	69

数据集	学生数	问题数	概念数	总交互数	平均交互数
Junyi	10 000	568	399	136 163	13
ASSIST09	4 217	26 688	123	401 756	95
Static2011	333	300	1 223	194 947	69

模型	图结构	特性		练习嵌入方法
模型	图结构	注意力	遗忘	练习嵌入方法
GKT^［27］	√			GNN
DGMN^［12］	√	√	√	Key-value Memory
Bi-CLKT^［6］	√	√		Contrastive Learning
DTransformer^［7］		√		Cognitive Diagnosis
IFKT^［14］	√		√	GCN
CRKT^［8］	√			Exercise Options
GKT-FM^［13］	√		√	Heterogeneous Relational Graph
GAMKT	√	√	√	Graphormer

模型	图结构	特性		练习嵌入方法
模型	图结构	注意力	遗忘	练习嵌入方法
GKT^［27］	√			GNN
DGMN^［12］	√	√	√	Key-value Memory
Bi-CLKT^［6］	√	√		Contrastive Learning
DTransformer^［7］		√		Cognitive Diagnosis
IFKT^［14］	√		√	GCN
CRKT^［8］	√			Exercise Options
GKT-FM^［13］	√		√	Heterogeneous Relational Graph
GAMKT	√	√	√	Graphormer

数据集	平均节点数	评价指标	GKT	DGMN	Bi-CLKT	DTransformer	IFKT	CRKT	GKT-FM	GAMKT
Junyi	10	AUC/%	77.4±0.03	84.6±0.02	83.1±0.04	82.3±0.01	84.3±0.02	82.6±0.03	84.2±0.03	86.4±0.02
		ACC/%	74.7±0.02	82.2±0.01	81.0±0.02	80.2±0.01	80.4±0.02	81.4±0.01	83.7±0.02	85.8±0.02
		RMSE	0.463	0.441	0.423	0.383	0.387	0.395	0.378	0.371
ASSIST09	5	AUC/%	72.3±0.01	85.9±0.03	83.3±0.04	81.4±0.03	84.5±0.02	85.7±0.02	84.1±0.02	86.1±0.01
		ACC/%	70.6±0.02	82.6±0.01	81.2±0.03	76.2±0.02	80.2±0.02	83.5±0.03	78.5±0.06	84.7±0.01
		RMSE	0.474	0.458	0.398	0.401	0.453	0.404	0.401	0.392
Static2011	15	AUC/%	76.9±0.01	85.4±0.01	85.6±0.01	83.7±0.02	86.2±0.02	84.8±0.01	85.3±0.04	87.1±0.01
		ACC/%	75.3±0.02	82.7±0.01	83.0±0.01	84.1±0.02	83.5±0.03	82.2±0.02	77.8±0.03	86.2±0.03
		RMSE	0.419	0.388	0.367	0.370	0.365	0.368	0.362	0.360

Knowledge tracking model based on concept association memory network with graph attention

融合图注意力的概念关联记忆网络知识追踪模型

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 11

References 37

Related Articles 15

Recommended Articles

Metrics

[1]	Wen LI, Kairong LI, Kai YANG. Subgraph-aware contrastive learning with data augmentation [J]. Journal of Computer Applications, 2026, 46(1): 1-9.
[2]	Xiang WANG, Zhixiang CHEN, Guojun MAO. Multivariate time series prediction method combining local and global correlation [J]. Journal of Computer Applications, 2025, 45(9): 2806-2816.
[3]	Wei ZHANG, Zhongwei GONG, Zhixin LI, Peihua LUO, Lingling SONG. Learning behavior boosted knowledge tracing model [J]. Journal of Computer Applications, 2025, 45(9): 2747-2754.
[4]	Yilin DENG, Fajiang YU. Pseudo random number generator based on LSTM and separable self-attention mechanism [J]. Journal of Computer Applications, 2025, 45(9): 2893-2901.
[5]	Chao LIU, Yanhua YU. Knowledge-aware recommendation model combining denoising strategy and multi-view contrastive learning [J]. Journal of Computer Applications, 2025, 45(9): 2827-2837.
[6]	Yi WANG, Yinglong MA. Multi-task social item recommendation method based on dynamic adaptive generation of item graph [J]. Journal of Computer Applications, 2025, 45(8): 2592-2599.
[7]	Yinchuan TU, Yong GUO, Heng MAO, Yi REN, Jianfeng ZHANG, Bao LI. Evaluation of training efficiency and training performance of graph neural network models based on distributed environment [J]. Journal of Computer Applications, 2025, 45(8): 2409-2420.
[8]	Chen LIANG, Yisen WANG, Qiang WEI, Jiang DU. Source code vulnerability detection method based on Transformer-GCN [J]. Journal of Computer Applications, 2025, 45(7): 2296-2303.
[9]	Zimo ZHANG, Xuezhuan ZHAO. Multi-scale sparse graph guided vision graph neural networks [J]. Journal of Computer Applications, 2025, 45(7): 2188-2194.
[10]	Danyang CHEN, Changlun ZHANG. Multi-scale decorrelation graph convolutional network model [J]. Journal of Computer Applications, 2025, 45(7): 2180-2187.
[11]	Yuelan ZHANG, Jing SU, Hangyu ZHAO, Baili YANG. Multi-view knowledge-aware and interactive distillation recommendation algorithm [J]. Journal of Computer Applications, 2025, 45(7): 2211-2220.
[12]	Xiaoqiang ZHAO, Yongyong LIU, Yongyong HUI, Kai LIU. Batch process quality prediction model using improved time-domain convolutional network with multi-head self-attention mechanism [J]. Journal of Computer Applications, 2025, 45(7): 2245-2252.
[13]	Mingfeng YU, Yongbin QIN, Ruizhang HUANG, Yanping CHEN, Chuan LIN. Multi-label text classification method based on contrastive learning enhanced dual-attention mechanism [J]. Journal of Computer Applications, 2025, 45(6): 1732-1740.
[14]	Hui LI, Bingzhi JIA, Chenxi WANG, Ziyu DONG, Jilong LI, Zhaoman ZHONG, Yanyan CHEN. Generative adversarial network underwater image enhancement model based on Swin Transformer [J]. Journal of Computer Applications, 2025, 45(5): 1439-1446.
[15]	Weichao DANG, Xinyu WEN, Gaimei GAO, Chunxia LIU. Multi-view and multi-scale contrastive learning for graph collaborative filtering [J]. Journal of Computer Applications, 2025, 45(4): 1061-1068.

模型	AUC/%
模型	Junyi	ASSIST09	Static2011
GAMKT-T	79.2±0.03	83.1±0.02	81.1±0.04
GAMKT-M	83.4±0.01	83.2±0.03	84.2±0.04
GAMKT-G	85.3±0.01	85.2±0.01	86.3±0.01
GAMKT	86.4±0.01	86.1±0.01	87.1±0.01

模型	AUC/%
模型	Junyi	ASSIST09	Static2011
GAMKT-T	79.2±0.03	83.1±0.02	81.1±0.04
GAMKT-M	83.4±0.01	83.2±0.03	84.2±0.04
GAMKT-G	85.3±0.01	85.2±0.01	86.3±0.01
GAMKT	86.4±0.01	86.1±0.01	87.1±0.01