|
|
 |
| |
|
|
|
Adaptive Gamification in Python Learning: Integrating Machine Learning Algorithms for Educational Process Personalization |
|
|
|
PP: 519-542 |
|
|
doi:10.18576/jsap/150311
|
|
|
|
Author(s) |
|
|
|
Nurkasym Arkabaev,
Topchubay Isakov,
Imad Yagoub Hamid,
Halla Elziber Elemam,
Sami Elsir Mohamed,
Sulima M. Awad Yousif,
Abdelgalal O. I. Abaker,
|
|
|
|
Abstract |
|
|
| Traditional approaches to programming education often face challenges with declining student motivation and high dropout rates, particularly when learning fundamental concepts. This paper presents an innovative adaptive gamification model that employs machine learning algorithms for dynamic personalization of the educational process in Python programming instruction. The developed system combines reinforcement learning and collaborative filtering methods to continuously analyze student behavioral patterns and automatically adjust game mechanics. Unlike static gamified platforms, the proposed approach adapts not only task difficulty but also types of motivational incentives, narrative elements, and reward systems according to each student’s individual learning style. Experimental validation was conducted with 120 first-year students divided into experimental and control groups. The study was carried out at Osh State University (Kyrgyz Republic, Osh) over 3 semesters from 2023-2025 academic years with students majoring in ”Computer Science and Engineering.” The application of adaptive gamification resulted in statistically significant increases in student intrinsic motivation by 68%, code quality improvement by 45%, and reduction in time to master basic Python concepts by 35%. Longitudinal observation showed sustained positive effects throughout subsequent semesters. The system demonstrated the ability to automatically identify different types of learners and apply optimal gamification strategies for each group. Log analysis revealed four main student behavioral patterns, for which specialized adaptive mechanics were developed. The theoretical significance of this research lies in creating a new model for integrating artificial intelligence with pedagogical principles, expanding understanding of personalization possibilities in education. The practical value is confirmed by developing a ready-to-implement solution that can be adapted for various programming languages and educational contexts. |
|
|
|
|
|
|
|
