A Framework for Enhancing Personalized Learning of Computer Programming using Large Language Models

DOI: https://doi.org/10.25147/ijcsr.v10i0.837

Abstract

Purpose – This study presents a framework that utilizes prompt engineering to refine AI-generated content, ensuring its coherence, accuracy, and educational relevance. It explores the integration of LLMs to enhance personalized learning in introductory computer programming courses, focusing on the C programming language.

Method – The quasi-experimental design was adopted to compare traditional teaching methods with LLM-supported learning. We evaluated adaptive learning algorithms to track changes in student performance over time using metrics like reduced error rates, average time spent on tasks, and task completion rates

Results – The experimental group recorded a pretest mean of 57.16 and an improved posttest mean of 66.24, yielding a gain of +9.08 marks. The control group began with a pretest mean of 60.94 and achieved a posttest mean of 63.08, resulting in a comparatively lower gain of +2.14 marks. To determine statistical significance between the groups, a Mann-Whitney U test was conducted on the posttest scores.  The test yielded a U-value of 1,725.5, a Z-score of -2.634, and a p-value of 0.008, indicating that the difference between the experimental and control groups was statistically significant at the p < 0.01 level.

Conclusion – Effective prompt engineering strategies reduce hallucinations and align generated content with curriculum objectives, thereby enabling scalable, adaptive, and student-centered support in programming education.

Recommendations – Learning through LLM frameworks should adopt effective prompt engineering strategies to achieve a practical impact on student engagement, content understanding, and the overall learning experience.

Practical Research Implications – The findings reinforce the pivotal role of prompt engineering in LLM-based learning. Effective GenAI frameworks for education must go beyond novelty and automation and instead embed educationally grounded principles that support learners' development through intelligent, responsive, and context-aware interactions.

Author Biographies

Erick Odhiambo Omuya, Department of Computing and Information Technology, Machakos University, Kenya

Dr. Erick Omuya is a researcher and AI-Machine Learning lecturer at Machakos University, Kenya. Dr. Omuya holds a PhD in Information Technology from the Jomo Kenyatta University of Agriculture and Technology (JKUAT), Kenya. His current research and publications deal with developing AI models for improving the performance of classification, clustering, and prediction using Machine/deep learning algorithms and Natural Language Processing as applied in sentiment analysis, market segmentation, among others.

Fredrick Muema Mboya, Department of Information Technology, Murang’a University of Technology, Kenya

Mr. Fredrick Muema Mboya is a researcher and Assistant Lecturer at Zetech University, Kenya. He holds a Master of Science in Information Technology from Murang'a University of Technology, Kenya. His research interests include Artificial Intelligence, Machine Learning, and Natural Language Processing, with a focus on intelligent systems and data-driven solutions.

Geoffrey Mariga Wambugu, Department of Information Technology, Murang’a University of Technology, Kenya

Dr. Geoffrey Mariga Wambugu is a Senior Lecturer in Information Technology at Murang'a University of Technology, Kenya. He holds a PhD in Information Technology - Machine Learning from Jomo Kenyatta University of Agriculture and Technology (JKUAT). His research focuses on machine learning, deep learning, data science, and natural language processing, among others. He has supervised numerous master's and doctoral students, led curriculum development initiatives, and published extensively in reputable journals.

Joyce Wangui Gikandi, Department of Educational Management and Curriculum Studies, Mount Kenya University, Kenya

Dr. Joyce W. Gikandi is a specialist in Information Systems and Educational Technology. She is currently a Senior Lecturer at Mount Kenya University (MKU). Her current research interests include E-Learning developments, ICT application in social and health sciences, AI Integration in education for adaptive learning, and digital skills for youth and women empowerment. Another area of interest is promoting innovative adaptation and use of open-source software and content for educational inclusivity.

Faith Mueni Musyoka, Department of Computing and Information Technology, University of Embu, Kenya

Dr. Faith Musyoka is a Senior Lecturer in the Computing and Information Technology Department at the University of Embu, Kenya. She holds a PhD in Information Technology from Kabarak University, Kenya. Her research interests span Artificial Intelligence, Responsible AI, Internet of Things (IoT), and digital transformation. She has published articles in high-impact peer-reviewed journals, and her current research focuses on developing trustworthy AI systems, AI governance frameworks, and digital health solutions.

Published
2026-07-02
How to Cite
OMUYA, Erick Odhiambo et al. A Framework for Enhancing Personalized Learning of Computer Programming using Large Language Models. International Journal of Computing Sciences Research, [S.l.], v. 10, p. 4266-4291, july 2026. ISSN 2546-115X. Available at: <//www.stepacademic.net/ijcsr/article/view/837>. Date accessed: 12 july 2026. doi: https://doi.org/10.25147/ijcsr.v10i0.837.
Citation Formats
Issue
Vol 10 (2026): Volume 10
Section
Articles
Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 International License.

This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited.

Most read articles by the same author(s)