Systematic Review of AI Driven Personalized Learning Models for Enhancing Student Engagement

Abstract

Artifical Intelligence (AI) has been a significant transformative force in modern education especially through providing personalized learning pathways and increasing student engagement. In this systematic review we will synthesize current developments in AI based personalized learning models and assess how these AI based models support student engagement on behavioural, emotional, and cognitive levels. We used a structured literature search over four databases: IEEE, Scopus, Springer, and ScienceDirect; to find all peer reviewed articles written since 2023 until 2025. Our results show that machine learning models specifically XGBoost, deep neural networks and reinforcement learning are very effective at identifying students\' needs and providing dynamic, adaptive instruction. Recommendation systems, based on AI, and behavioural analytics also enhance students\' motivation, encourage active participation in class, and improve academic achievement. The results also highlight several limitations such as explainability of AI decision making processes; ethically responsible personalized learning; ability of AI to adapt in real time; and limited utilization of multi modal student engagement data. Finally, our review includes recommendations for future research directions and outlines an integrative framework for using AI to provide personalized learning that is both scalable and equitable.

Introduction

AI has rapidly transformed education by enabling adaptive, data-driven, and personalized learning experiences that go beyond traditional “one-size-fits-all” instruction. Leveraging vast data from LMS, digital classrooms, and assessments, AI can continuously monitor, predict, and respond to student behaviors, supporting behavioral, emotional, and cognitive engagement. Machine learning techniques—including XGBoost, neural networks, transformer models, and reinforcement learning—have been applied to predict student performance, identify at-risk learners, and recommend individualized resources.

This systematic review examined 135 studies (2023–2025) across major academic databases to assess AI-driven personalized learning models and their impact on engagement. Key findings include:

• AI enables real-time adaptive interventions, proactive and reactive engagement, and scaffolding of learning based on individual needs.

• Supervised learning (e.g., XGBoost, Random Forest, SVM) is commonly used to tailor content, pace, and feedback.

• AI-powered tools, including intelligent tutoring systems and chatbots, enhance engagement and improve learning outcomes.

• Ethical, social, and practical considerations—such as data privacy, transparency, algorithmic bias, and teacher-student interactions—remain critical.

The review highlights gaps in theory-driven design, ethical personalization, explainable AI, and scalable implementation, offering guidance for educators, researchers, and system designers to build more adaptive, equitable, and engaging learning environments.

Conclusion

This systematic review highlights the growing significance of AI driven personalized learning models in improving student engagement and academic outcomes. The reviewed studies demonstrate that AI technologies—particularly machine learning algorithms such as XGBoost, Random Forest, LSTM, and transformer-based models—can effectively predict student performance, identify disengagement patterns, and deliver targeted interventions. These advancements have enabled more responsive and adaptive learning environments, supporting better engagement, motivation, and individualized learning pathways. Despite these promising outcomes, several limitations persist across existing literature. The majority of studies rely on single dimensional datasets, lack real world classroom evaluations, and pay insufficient attention to higher order thinking dimensions such as creativity, problem solving, and ethical AI use.

References

[1] C. Yuan, L. Zhang, and T. Chen, “Enhancing Student Learning Outcomes through AI-Driven Educational Interventions: A Comprehensive Study of Classroom Behavior and Machine Learning Integration,” Int. Trans. J. Pedagogical Human–Social Sci., vol. 2, no. 2, pp. 33–45, 2025. DOI: 10.70693/itphss.v2i2.216 [2] R. Bedi and M. Kapoor, “Adaptive Feedback and Personalization in AI-Enabled Tutoring Systems,” Computers & Education Open, vol. 6, no. 1, pp. 102–118, 2024. [3] Y. Li and H. Sun, “Fostering Creativity and Critical Thinking through AI-Enhanced Project-Based Learning,” BMC Psychology, vol. 12, art. no. 497, 2024. DOI: 10.1186/s40359-024-01979-0 [4] M. Rahman, P. Das, and F. Hossain, “Hybrid Reinforcement and Rule-Based Personalization for Adaptive Learning Environments,” IEEE Access, vol. 11, pp. 125431–125442, 2023. [5] S. Kumar, N. Patel, and R. Chauhan, “AI Ethics as a Pedagogical Imperative: Teachers’ Perspectives on Responsible Integration in Higher Education,” Int. J. Educ. Technol. High. Educ., vol. 21, no. 4, pp. 75–89, 2024. DOI: 10.1186/s41239-024-00496-9 [6] C. Yuan, L. Zhang, and T. Chen, “Enhancing Student Learning Outcomes through AI-Driven Educational Interventions: A Comprehensive Study of Classroom Behavior and Machine Learning Integration,” Int. Trans. J. Pedagogical Human–Social Sci., vol. 2, no. 2, pp. 33–45, 2025, doi: 10.70693/itphss.v2i2.216. [7] R. Bedi and M. Kapoor, “Adaptive Feedback and Personalization in AI-Enabled Tutoring Systems,” Comput. Educ. Open, vol. 6, pp. 102–118, 2024. [8] Y. Li and H. Sun, “Fostering Creativity and Critical Thinking through AI-Enhanced Project-Based Learning,” BMC Psychol., vol. 12, art. no. 497, 2024, doi: 10.1186/s40359-024-01979-0. [9] [M. Rahman, P. Das, and F. Hossain, “Hybrid Reinforcement and Rule-Based Personalization for Adaptive Learning Environments,” IEEE Access, vol. 11, pp. 125431–125442, 2023. [10] S. Kumar, N. Patel, and R. Chauhan, “AI Ethics as a Pedagogical Imperative: Teachers’ Perspectives on Responsible Integration in Higher Education,” Int. J. Educ. Technol. High. Educ., vol. 21, no. 4, pp. 75–89, 2024, doi: 10.1186/s41239-024-00496-9. [11] S. Sakri, L. I. Kori, D. N. Joshi, A. S. Nayak, and S. Marigoudar, “AI-Enabled Transformation of Online Learning through Personalization,” J. Eng. Educ. Transform., vol. 39, Special Issue 1, 2025, doi: 10.16920/jeet/2025/v39is1/25131. [12] L. Fletscher, J. Mercado, Á. Gómez, and C. Mendoza-Cardenas, “Innovating Personalized Learning in Virtual Education Through AI,” Multimodal Technol. Interact., vol. 9, no. 7, art. no. 69, 2025, doi: 10.3390/mti9070069. [13] N. Katiyar, V. K. Awasthi, R. Pratap, K. Mishra, N. Shukla, and M. Tiwari, “AI-Driven Personalized Learning Systems: Enhancing Educational Effectiveness,” Educ. Admin. Theory Pract., vol. 30, no. 5, 2024, doi: 10.53555/kuey.v30i5.4961. [14] J. Anderson, “AI in Education: Transforming Learning and Personalized Instruction,” J. Arts Soc. Educ. Stud., vol. 4, no. 3, 2022. [15] A. Singh and G. Kaur, “The Role of AI in Education and Personalized Learning,” Int. Res. J. Modern. Eng. Technol. Sci., vol. 7, no. 4, Apr. 2025, doi: 10.56726/IRJMETS73571. [16] B. Kim, H. Suh, J. Heo, and Y. Choi, “AI-Driven Interface Design for Intelligent Tutoring Systems Improves Student Engagement,” arXiv preprint, arXiv:2009.08976, 2020. [17] S. S. Lodhi and S. Lodhi, “Integration of AI in STEM Education: Addressing Ethical Challenges in K–12 Settings,” arXiv preprint, arXiv:2510.19196, 2025. [18] Z. Hao, J. Cao, R. Li, J. Yu, Z. Liu, and Y. Zhang, “Mapping Student–AI Interaction Dynamics in Multi-Agent Learning Environments: Supporting Personalized Learning and Reducing Performance Gaps,” arXiv preprint, arXiv:2506.02993, 2025. [19] Z. Hao, J. Jiang, J. Yu, Z. Liu, and Y. Zhang, “Student–AI Interaction in an LLM-Empowered Learning Environment: A Cluster Analysis of Engagement Profiles,” arXiv preprint, arXiv:2503.01694, 2025. [20] A. Iyer and S. Subramanian, “Personalized Learning Systems that Adjust to Individual Needs: Interactive Learning Environments & AI Real-Time Feedback,” Biomed J. Sci. Tech. Res., vol. 59, no. 1, 2024, doi: 10.26717/BJSTR.2024.59.009246.

Copyright

Copyright © 2025 Chetna Masih, Dr. Swati Namdev. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.