Learning State Prediction and Learning Path Generation Using Hidden Markov Models

Song-Hwan Kwon; Jong-Nam Rim; Yun-Ho Ko; Yun-Sok Ham; Ha-Gyong Kim

doi:doi:10.11648/j.scif.20260202.13

Research Article |

| Peer-Reviewed

Learning State Prediction and Learning Path Generation Using Hidden Markov Models

Song-Hwan Kwon^*

, Jong-Nam Rim, Yun-Ho Ko, Yun-Sok Ham, Ha-Gyong Kim

Published in Science Futures (Volume 2, Issue 2)

Received: 21 September 2025 Accepted: 23 October 2025 Published: 20 January 2026

Views: Downloads:

Download PDF

Share This Article

Twitter
Linked In
Facebook

Abstract

The main mission of the intelligent tutoring system is to estimate and predict the students’ learning status accurately and to generate a learning path that is inherent to the students. There are many learning state estimation and prediction methods, and there are also many methods to generate learning paths using the results. We propose a method to predict the learning state of a student and generate a learning path using the Hidden Markov Model, a probabilistic reasoning over time. We propose and illustrate in detail how to formalize the Hidden Markov Model using an extended knowledge graph for learning state prediction, how to collect the evidence variable values of the Hidden Markov Model in intelligent tutoring systems, how to predict the learning state using smoothing algorithms, and how to generate the stochastic learning path. The experiment was conducted to compare students using an intelligent tutoring system with those who did not. In addition, a comparison experiment of a method using Hidden Markov model with a multi-dimensional linear prediction model such as PFM and AFM was also carried out. The results showed that the average performance of students using the intelligent tutoring system was much higher than that of students without the intelligent tutoring system. In addition, the experimental results showed that the average performance of students using Hidden Markov models was higher than that of students using other models.

Published in	Science Futures (Volume 2, Issue 2)
DOI	10.11648/j.scif.20260202.13
Page(s)	118-134
Creative Commons	This is an Open Access article, distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution and reproduction in any medium or format, provided the original work is properly cited.
Copyright	Copyright © The Author(s), 2026. Published by Science Publishing Group

Keywords

HMM, Learning State Prediction & Estimation, Learning Path Generation, Probabilistic Reasoning over Time

References

[1]	MURTAZA, M., AHMED, Y., SHAMSI, J. A., SHERWANI, F., & USMAN, M. (2022). AI-Based Personalized E-Learning Systems: Issues, Challenges, and Solutions. IEEE Access, VOLUME 10, pp 81323-81326.
[2]	Ouyang, Y., Zhou, Y., Zhang, H., Rong, W., and Xiong, Z. (2021). PAKT: A Position-Aware Self-attentive Approach for Knowledge Tracing. Artificial Intelligence in Education-2021, 22nd International Conference, AIED 2021 Utrecht, The Netherlands, June 14–18, Proceedings, Part II, Springer.
[3]	Su, Y., Liu, Q., Liu, Q., Huang, Z., Yin, Y., Chen, E., Ding, C., Wei, S., & Hu, G. (2018). Exercise-enhanced sequential modeling for Student performance prediction. in Proc. AAAI Conf. Artif. Intell., 2018, vol. 32, no. 1, pp. 1–9.
[4]	Liu, Q., Huang, Z., Yin, Y., Chen, E., Xiong, H., Su, Y., & Hu, G. (2021). EKT: Exercise-aware knowledge tracing for Student performance prediction. IEEE Trans. Knowl. Data Eng., vol. 33, no. 1, pp. 100–115, Jan.
[5]	Rosen, Y., Rushkin, I., Rubin, R., Munson, L., Ang, A., Weber, G., Lopez, G., and Tingley, D. (2018). Adaptive Learning Open Source Initiative for MOOC Experimentation. © Springer International Publishing AG, part of Springer Nature 2018, C. Penstein Rosé et al. (Eds.): AIED 2018, LNAI 10948, pp. 307–311, https://doi.org/10.1007/978-3-319-93846-2_57
[6]	Lipton, Z. C., Berkowitz, J., and Elkan, C. (2015). A critical review of recurrent neural networks for sequence learning. arXiv: 1506.00019.
[7]	Chen, P., Lu, Y., Zheng, V. W., and Pian, Y. (2018). ‘Prerequisite-driven deep knowledge tracing. in Proc. IEEE Int. Conf. Data Mining (ICDM), Nov. pp. 39–48.
[8]	Liu, Q., Shen, S., Huang, Z., Chen, E., and Zheng, Y. (2021). A survey of knowledge tracing. arXiv: 2105.15106.
[9]	Nakagawa, H., Iwasawa, Y., and Matsuo, Y. (2019). Graph-based knowledge tracing: Modeling Student proficiency using graph neural network. in Proc. IEEE/WIC/ACM Int. Conf. Web Intell., Oct., pp. 156–163.
[10]	Yang, Y., Shen, J., Qu, Y., Liu, Y., Wang, K., Zhu, Y., Zhang, W., and Yu, Y. (2020). GIKT: A graph-based interaction model for knowledge tracing. in Proc. Joint Eur. Conf. Mach. Learn. Knowl. Discovery Databases. Cham, Switzerland: Springer, pp. 299–315.
[11]	Song, X., Li, J., Lei, Q., Zhao, W., Chen, Y., and Mian, A. (2022). Bi-CLKT: Bi-graph contrastive learning based knowledge tracing. arXiv: 2201.09020.
[12]	Tong, S., Liu, Q., Huang, W., Hunag, Z., Chen, E., Liu, C., Ma, H., and Wang, S. (2020). Structure-based knowledge tracing: An influence propagation view. in Proc. IEEE Int. Conf. Data Mining (ICDM), Nov. pp. 541–550.
[13]	Russell, S. J., & Norvig, P. (2022). Artificial Intelligence: A Modern Approach. Prentice Hall. Retrieved from http://www.amazon.com/Artificial-Intelligence-ApproachStuart-Russell/dp/0131038052
[14]	Pardos, Z. A., Heffernan, N. T. (2022). Using HMMs and bagged decision trees to leverage rich features of user and skill from an intelligent tutoring system dataset. the Journal of Machine Learning Research W & CP, In Press.
[15]	Masun, H., Rania, L., Rosa Maria, C., and Ghias, B. (2007). Student Modeling Using NN-HMM for EFL Course. ICTTA08, Syria, Abril.
[16]	Masun, H., Rania, L., Rosa Maria, C., and Ghias, B. (2007). Building adaptive web-based educational system using Fuzzy-ART2 for EFL course. Hiroshima, Japan, November.
[17]	Masun, H., L. Ghias, R. B. (2006). Adaptive web-based educational system using neural networks in EFL Course. IEEE Xplore Release 2.2, 1, pp. 622- 625.
[18]	Xuejing, G. (2003). Research on modeling artificial emotion based on HMM and techniques correlated with virtual human. Univ of Science and Technology of Beijing. pp. 66-72.
[19]	Xuejing, G. (2010) Design of Emotional Intelligent Tutor System Based on HMM. 2010 Sixth International Conference on Natural Computation (ICNC 2010).
[20]	Boyer, K. E., Phillips, R., Wallis, M., Vouk, M., and Lester, J. (2008). Balancing cognitive and motivational scaffolding in tutorial dialogue. Proceedings of the 9th International Conference on Intelligent Tutoring Systems: 239-249.
[21]	Boyer, K. E. et al, (2020). Inferring Tutorial Dialogue Structure with Hidden Markov Modeling.
[22]	Chevalère, J. (2023). A sequence of learning processes in an intelligent tutoring system from topic-related appraisals to learning gains. Learning and Instruction. Volume 87, October, ScienceDirect.

Cite This Article

Plain Text BibTeX RIS

APA Style

Kwon, S., Rim, J., Ko, Y., Ham, Y., Kim, H. (2026). Learning State Prediction and Learning Path Generation Using Hidden Markov Models. Science Futures, 2(2), 118-134. https://doi.org/10.11648/j.scif.20260202.13

Copy | Download

ACS Style

Kwon, S.; Rim, J.; Ko, Y.; Ham, Y.; Kim, H. Learning State Prediction and Learning Path Generation Using Hidden Markov Models. Sci. Futures 2026, 2(2), 118-134. doi: 10.11648/j.scif.20260202.13

Copy | Download

AMA Style

Kwon S, Rim J, Ko Y, Ham Y, Kim H. Learning State Prediction and Learning Path Generation Using Hidden Markov Models. Sci Futures. 2026;2(2):118-134. doi: 10.11648/j.scif.20260202.13

Copy | Download

@article{10.11648/j.scif.20260202.13,
  author = {Song-Hwan Kwon and Jong-Nam Rim and Yun-Ho Ko and Yun-Sok Ham and Ha-Gyong Kim},
  title = {Learning State Prediction and Learning Path Generation Using Hidden Markov Models},
  journal = {Science Futures},
  volume = {2},
  number = {2},
  pages = {118-134},
  doi = {10.11648/j.scif.20260202.13},
  url = {https://doi.org/10.11648/j.scif.20260202.13},
  eprint = {https://article.sciencepublishinggroup.com/pdf/10.11648.j.scif.20260202.13},
  abstract = {The main mission of the intelligent tutoring system is to estimate and predict the students’ learning status accurately and to generate a learning path that is inherent to the students. There are many learning state estimation and prediction methods, and there are also many methods to generate learning paths using the results. We propose a method to predict the learning state of a student and generate a learning path using the Hidden Markov Model, a probabilistic reasoning over time. We propose and illustrate in detail how to formalize the Hidden Markov Model using an extended knowledge graph for learning state prediction, how to collect the evidence variable values of the Hidden Markov Model in intelligent tutoring systems, how to predict the learning state using smoothing algorithms, and how to generate the stochastic learning path. The experiment was conducted to compare students using an intelligent tutoring system with those who did not. In addition, a comparison experiment of a method using Hidden Markov model with a multi-dimensional linear prediction model such as PFM and AFM was also carried out. The results showed that the average performance of students using the intelligent tutoring system was much higher than that of students without the intelligent tutoring system. In addition, the experimental results showed that the average performance of students using Hidden Markov models was higher than that of students using other models.},
 year = {2026}
}

Copy | Download

TY  - JOUR
T1  - Learning State Prediction and Learning Path Generation Using Hidden Markov Models
AU  - Song-Hwan Kwon
AU  - Jong-Nam Rim
AU  - Yun-Ho Ko
AU  - Yun-Sok Ham
AU  - Ha-Gyong Kim
Y1  - 2026/01/20
PY  - 2026
N1  - https://doi.org/10.11648/j.scif.20260202.13
DO  - 10.11648/j.scif.20260202.13
T2  - Science Futures
JF  - Science Futures
JO  - Science Futures
SP  - 118
EP  - 134
PB  - Science Publishing Group
UR  - https://doi.org/10.11648/j.scif.20260202.13
AB  - The main mission of the intelligent tutoring system is to estimate and predict the students’ learning status accurately and to generate a learning path that is inherent to the students. There are many learning state estimation and prediction methods, and there are also many methods to generate learning paths using the results. We propose a method to predict the learning state of a student and generate a learning path using the Hidden Markov Model, a probabilistic reasoning over time. We propose and illustrate in detail how to formalize the Hidden Markov Model using an extended knowledge graph for learning state prediction, how to collect the evidence variable values of the Hidden Markov Model in intelligent tutoring systems, how to predict the learning state using smoothing algorithms, and how to generate the stochastic learning path. The experiment was conducted to compare students using an intelligent tutoring system with those who did not. In addition, a comparison experiment of a method using Hidden Markov model with a multi-dimensional linear prediction model such as PFM and AFM was also carried out. The results showed that the average performance of students using the intelligent tutoring system was much higher than that of students without the intelligent tutoring system. In addition, the experimental results showed that the average performance of students using Hidden Markov models was higher than that of students using other models.
VL  - 2
IS  - 2
ER  -

Copy | Download

Author Information

Song-Hwan Kwon

Department of Information Science, University of Science, Pyongyang, Democratic People’s Republic of Korea

Contact Email

http://orcid.org/0009-0005-2162-081X
Jong-Nam Rim

Department of Information Science, University of Science, Pyongyang, Democratic People’s Republic of Korea
Yun-Ho Ko

Department of Information Science, University of Science, Pyongyang, Democratic People’s Republic of Korea
Yun-Sok Ham

Institute of Information Technology, University of Science, Pyongyang, Democratic People’s Republic of Korea
Ha-Gyong Kim

Department of Information Science, University of Science, Pyongyang, Democratic People’s Republic of Korea

Download PDF

Submit an Article

Sections

Plain Text BibTeX RIS

APA Style

Kwon, S., Rim, J., Ko, Y., Ham, Y., Kim, H. (2026). Learning State Prediction and Learning Path Generation Using Hidden Markov Models. Science Futures, 2(2), 118-134. https://doi.org/10.11648/j.scif.20260202.13

Copy | Download

ACS Style

Kwon, S.; Rim, J.; Ko, Y.; Ham, Y.; Kim, H. Learning State Prediction and Learning Path Generation Using Hidden Markov Models. Sci. Futures 2026, 2(2), 118-134. doi: 10.11648/j.scif.20260202.13

Copy | Download

AMA Style

Kwon S, Rim J, Ko Y, Ham Y, Kim H. Learning State Prediction and Learning Path Generation Using Hidden Markov Models. Sci Futures. 2026;2(2):118-134. doi: 10.11648/j.scif.20260202.13

Copy | Download

@article{10.11648/j.scif.20260202.13,
  author = {Song-Hwan Kwon and Jong-Nam Rim and Yun-Ho Ko and Yun-Sok Ham and Ha-Gyong Kim},
  title = {Learning State Prediction and Learning Path Generation Using Hidden Markov Models},
  journal = {Science Futures},
  volume = {2},
  number = {2},
  pages = {118-134},
  doi = {10.11648/j.scif.20260202.13},
  url = {https://doi.org/10.11648/j.scif.20260202.13},
  eprint = {https://article.sciencepublishinggroup.com/pdf/10.11648.j.scif.20260202.13},
  abstract = {The main mission of the intelligent tutoring system is to estimate and predict the students’ learning status accurately and to generate a learning path that is inherent to the students. There are many learning state estimation and prediction methods, and there are also many methods to generate learning paths using the results. We propose a method to predict the learning state of a student and generate a learning path using the Hidden Markov Model, a probabilistic reasoning over time. We propose and illustrate in detail how to formalize the Hidden Markov Model using an extended knowledge graph for learning state prediction, how to collect the evidence variable values of the Hidden Markov Model in intelligent tutoring systems, how to predict the learning state using smoothing algorithms, and how to generate the stochastic learning path. The experiment was conducted to compare students using an intelligent tutoring system with those who did not. In addition, a comparison experiment of a method using Hidden Markov model with a multi-dimensional linear prediction model such as PFM and AFM was also carried out. The results showed that the average performance of students using the intelligent tutoring system was much higher than that of students without the intelligent tutoring system. In addition, the experimental results showed that the average performance of students using Hidden Markov models was higher than that of students using other models.},
 year = {2026}
}

Copy | Download

TY  - JOUR
T1  - Learning State Prediction and Learning Path Generation Using Hidden Markov Models
AU  - Song-Hwan Kwon
AU  - Jong-Nam Rim
AU  - Yun-Ho Ko
AU  - Yun-Sok Ham
AU  - Ha-Gyong Kim
Y1  - 2026/01/20
PY  - 2026
N1  - https://doi.org/10.11648/j.scif.20260202.13
DO  - 10.11648/j.scif.20260202.13
T2  - Science Futures
JF  - Science Futures
JO  - Science Futures
SP  - 118
EP  - 134
PB  - Science Publishing Group
UR  - https://doi.org/10.11648/j.scif.20260202.13
AB  - The main mission of the intelligent tutoring system is to estimate and predict the students’ learning status accurately and to generate a learning path that is inherent to the students. There are many learning state estimation and prediction methods, and there are also many methods to generate learning paths using the results. We propose a method to predict the learning state of a student and generate a learning path using the Hidden Markov Model, a probabilistic reasoning over time. We propose and illustrate in detail how to formalize the Hidden Markov Model using an extended knowledge graph for learning state prediction, how to collect the evidence variable values of the Hidden Markov Model in intelligent tutoring systems, how to predict the learning state using smoothing algorithms, and how to generate the stochastic learning path. The experiment was conducted to compare students using an intelligent tutoring system with those who did not. In addition, a comparison experiment of a method using Hidden Markov model with a multi-dimensional linear prediction model such as PFM and AFM was also carried out. The results showed that the average performance of students using the intelligent tutoring system was much higher than that of students without the intelligent tutoring system. In addition, the experimental results showed that the average performance of students using Hidden Markov models was higher than that of students using other models.
VL  - 2
IS  - 2
ER  -

Copy | Download