Comparative Study of AI Techniques for Personalized Learning

Author

Smita Mangesh Junnarkar, Karuna Tripathi

Abstract

 Learners differ in prior knowledge, pace, goals, and motivation, and a single sequence of lessons seldom suits all of them. Artificial intelligence offers many ways to adapt instruction, but the techniques rest on different assumptions, need different data, and fail in different ways. This paper compares eight families of AI technique used for personalized learning: rule-based expert tutors, Bayesian knowledge tracing, item response theory and cognitive diagnosis, collaborative filtering and matrix factorisation, deep knowledge tracing, reinforcement learning and bandit methods, knowledge-graph and content-based methods, and large language model tutors. For each family the paper explains what the technique models, what data it needs, what it does well, and where it breaks down. A comparison framework is then proposed, with a technique inventory, formal models of mastery, item difficulty, and learning gain, and metrics for prediction, recommendation, learning outcome, interpretability, cost, and fairness. The paper closes with design guidelines and ethical considerations. The work is analytical and reports no new empirical results. Its purpose is to help students, researchers, and developers choose and combine techniques that fit their learners, data, and constraints. 

Keywords

Personalized learning, adaptive learning, intelligent tutoring systems, knowledge tracing, recommender systems, reinforcement learning, large language models, educational data mining, learner modelling

Full Text:

Download Paper PDF

References

[1]B. S. Bloom, "The 2 sigma problem: The search for methods of group instruction as effective as one-to-one tutoring," Educational Researcher, vol. 13, no. 6, pp. 4--16, 1984. 

[2]K. VanLehn, "The relative effectiveness of human tutoring, intelligent tutoring systems, and other tutoring systems," Educational Psychologist, vol. 46, no. 4, pp. 197--221, 2011. 

[3]J. A. Kulik and J. D. Fletcher, "Effectiveness of intelligent tutoring systems: A meta-analytic review," Review of Educational Research, vol. 86, no. 1, pp. 42--78, 2016. 

[4]J. F. Pane, B. A. Griffin, D. F. McCaffrey, and R. Karam, "Effectiveness of Cognitive Tutor Algebra I at scale," Educational Evaluation and Policy Analysis, vol. 36, no. 2, pp. 127--144, 2014. 

[5]O. Zawacki-Richter, V. I. Marín, M. Bond, and F. Gouverneur, "Systematic review of research on artificial intelligence applications in higher education -- where are the educators?" International Journal of Educational Technology in Higher Education, vol. 16, art. 39, 2019. 

[6]L. Chen, P. Chen, and Z. Lin, "Artificial intelligence in education: A review," IEEE Access, vol. 8, pp. 75264--75278, 2020. 

[7]K. Zhang and A. B. Aslan, "AI technologies for education: Recent research and future directions," Computers and Education: Artificial Intelligence, vol. 2, art. 100025, 2021. 

[8]A. T. Corbett and J. R. Anderson, "Knowledge tracing: Modeling the acquisition of procedural knowledge," User Modeling and User-Adapted Interaction, vol. 4, no. 4, pp. 253--278, 1994. 

[9]R. S. J. d. Baker and P. S. Inventado, "Educational data mining and learning analytics," in Learning Analytics: From Research to Practice. New York, NY, USA: Springer, 2014, pp. 61--75. 

[10]C. Romero and S. Ventura, "Educational data mining and learning analytics: An updated survey," WIREs Data Mining and Knowledge Discovery, vol. 10, no. 3, art. e1355, 2020. 

[11]T. Gervet, K. Koedinger, J. Schneider, and T. Mitchell, "When is deep learning the best approach to knowledge tracing?" Journal of Educational Data Mining, vol. 12, no. 3, pp. 31--54, 2020. 

[12]M. V. Yudelson, K. R. Koedinger, and G. J. Gordon, "Individualized Bayesian knowledge tracing models," in Artificial Intelligence in Education (AIED), Lecture Notes in Computer Science, vol. 7926, 2013, pp. 171--180. 

[13]S. Hochreiter and J. Schmidhuber, "Long short-term memory," Neural Computation, vol. 9, no. 8, pp. 1735--1780, 1997. 

[14]C. Piech et al., "Deep knowledge tracing," in Advances in Neural Information Processing Systems (NIPS), vol. 28, 2015, pp. 505--513. 

[15]J. Zhang, X. Shi, I. King, and D.-Y. Yeung, "Dynamic key-value memory networks for knowledge tracing," in Proc. International World Wide Web Conference (WWW), 2017, pp. 765--774. 

[16]A. Vaswani et al., "Attention is all you need," in Advances in Neural Information Processing Systems (NIPS), vol. 30, 2017, pp. 5998--6008. 

[17]S. Pandey and G. Karypis, "A self-attentive model for knowledge tracing," in Proc. International Conference on Educational Data Mining (EDM), 2019, pp. 384--389. 

[18]A. Ghosh, N. Heffernan, and A. S. Lan, "Context-aware attentive knowledge tracing," in Proc. ACM SIGKDD Conf. Knowledge Discovery and Data Mining (KDD), 2020, pp. 2330--2339. 

[19]G. Abdelrahman, Q. Wang, and B. P. Nunes, "Knowledge tracing: A survey," ACM Computing Surveys, vol. 55, no. 11, art. 224, 2023. 

[20]Y. Koren, R. Bell, and C. Volinsky, "Matrix factorization techniques for recommender systems," Computer, vol. 42, no. 8, pp. 30--37, 2009. 

[21]X. He, L. Liao, H. Zhang, L. Nie, X. Hu, and T.-S. Chua, "Neural collaborative filtering," in Proc. International World Wide Web Conference (WWW), 2017, pp. 173--182. 

[22]F. Ricci, L. Rokach, and B. Shapira, Eds., Recommender Systems Handbook, 2nd ed. New York, NY, USA: Springer, 2015. 

[23]R. S. Sutton and A. G. Barto, Reinforcement Learning: An Introduction, 2nd ed. Cambridge, MA, USA: MIT Press, 2018. 

[24]V. Mnih et al., "Human-level control through deep reinforcement learning," Nature, vol. 518, pp. 529--533, 2015. 

[25]B. Clement, D. Roy, P.-Y. Oudeyer, and M. Lopes, "Multi-armed bandits for intelligent tutoring systems," Journal of Educational Data Mining, vol. 7, no. 2, pp. 20--48, 2015. 

[26]T. B. Brown et al., "Language models are few-shot learners," in Advances in Neural Information Processing Systems (NeurIPS), vol. 33, 2020, pp. 1877--1901. 

[27]E. Kasneci et al., "ChatGPT for good? On opportunities and challenges of large language models for education," Learning and Individual Differences, vol. 103, art. 102274, 2023. 

[28]Z. Ji et al., "Survey of hallucination in natural language generation," ACM Computing Surveys, vol. 55, no. 12, art. 248, 2023. 

[29]E. M. Bender, T. Gebru, A. McMillan-Major, and S. Shmitchell, "On the dangers of stochastic parrots: Can language models be too big?" in Proc. ACM Conf. Fairness, Accountability, and Transparency (FAccT), 2021, pp. 610--623. 

[30]R. S. Baker and A. Hawn, "Algorithmic bias in education," International Journal of Artificial Intelligence in Education, vol. 32, no. 4, pp. 1052--1092, 2022. 

[31]Government of India, "The Digital Personal Data Protection Act, 2023 (Act No. 22 of 2023)," 2023. 

[32]A. Hogan et al., "Knowledge graphs," ACM Computing Surveys, vol. 54, no. 4, art. 71, 2021. 

[33]S. E. Embretson and S. P. Reise, Item Response Theory for Psychologists. Mahwah, NJ, USA: Lawrence Erlbaum Associates, 2000. 

[34]T. N. Kipf and M. Welling, "Semi-supervised classification with graph convolutional networks," in Proc. International Conference on Learning Representations (ICLR), 2017. 

[35]S. Doroudi, V. Aleven, and E. Brunskill, "Where's the reward? A review of reinforcement learning for instructional sequencing," International Journal of Artificial Intelligence in Education, vol. 29, no. 4, pp. 568--620, 2019. 

[36]P. Lewis et al., "Retrieval-augmented generation for knowledge-intensive NLP tasks," in Advances in Neural Information Processing Systems (NeurIPS), vol. 33, 2020, pp. 9459--9474. 

[37]R. R. Hake, "Interactive-engagement versus traditional methods: A six-thousand-student survey of mechanics test data for introductory physics courses," American Journal of Physics, vol. 66, no. 1, pp. 64--74, 1998. 

[38]K. Järvelin and J. Kekäläinen, "Cumulated gain-based evaluation of IR techniques," ACM Transactions on Information Systems, vol. 20, no. 4, pp. 422--446, 2002. 

[39]B. McMahan, E. Moore, D. Ramage, S. Hampson, and B. A. y Arcas, "Communication-efficient learning of deep networks from decentralized data," Proceedings of Machine Learning Research, vol. 54, pp. 1273--1282, 2017. 

[40]C. Dwork and A. Roth, "The algorithmic foundations of differential privacy," Foundations and Trends in Theoretical Computer Science, vol. 9, no. 3--4, pp. 211--407, 2014. 

Share your valuable work from Social Media Buttons