AI-powered system for predicting student performance in python programming within the Zimbabwean O-level computer science curriculum.
- Author
- Matanga, Takudzwa
- Title
- AI-powered system for predicting student performance in python programming within the Zimbabwean O-level computer science curriculum.
- Abstract
- This research presents the development and evaluation of an AI-powered system designed to predict student performance in Python programming within the Zimbabwean O-Level Computer Science curriculum. Addressing challenges prevalent in under-resourced educational settings, such as limited teacher proficiency and inadequate digital learning support, the system employs a Random Forest algorithm to analyse student interactions with multiple-choice quizzes. The methodology integrates a mixed-methods, quasi-experimental research design with an agile, user centred system development approach, combining quantitative data from quiz results with qualitative feedback from students and teachers. The machine learning model achieved an 88% accuracy in predicting student performance, effectively identifying at-risk learners and highlighting challenging topics like 'Loops' and 'Data Structures'. Qualitative findings affirmed the system's usability, effectiveness, and practical value, with teachers appreciating data-driven insights and students reporting enhanced motivation. The study successfully demonstrated the potential of AI to offer predictive insights and analytical tools tailored for local contexts, thereby empowering educators and fostering data-informed learning. Recommendations include integrating the system into the curriculum, developing teacher training programs, and exploring future research avenues such as longitudinal impact studies and adaptive learning path generation. While acknowledging limitations such as sample size and reliance on quiz data, this research contributes significantly to the field of technology-enhanced learning in developing countries.
- Date
- June 2025
- Publisher
- BUSE
- Keywords
- AI-Powered
- Python Programming
- Supervisor
- Mr. G. Mhlanga
- Item sets
- Department of Computer Science
- Media
-
Matanga, Takudzwa.pdf