AI-DRIVEN PERSONALIZED LEARNING SYSTEM FOR TERTIARY INSTITUTION STUDENTS

This project presents the design and development of an AI-Driven Personalized Learning System aimed at improving comprehension among tertiary institution students. Many Nigerian students struggle to understand complex academic materials due to the technical nature of textbooks and lecture notes. The system addresses this by allowing students to upload materials in text, PDF, or image formats, which are then simplified into clearer explanations using Optical Character Recognition (OCR) and Natural Language Processing (NLP) models. An Agile development approach guided the process, ensuring iterative refinement through user feedback. The system was implemented with Flask (Python) on the backend and HTML, CSS, and JavaScript on the frontend, supported by a hybrid AI architecture that combines local transformer models (Flan-T5) with fallback cloud APIs such as Gemini or ChatGPT. This design ensures both flexibility and reliability even under limited hardware or internet constraints. User evaluation involving selected tertiary students showed that over 85% found the simplified explanations significantly easier to understand, confirming the system’s effectiveness in enhancing learning outcomes. Despite challenges like limited computing power and occasional API restrictions, the prototype proved functional and impactful. The project demonstrates that hybrid AI solutions can bridge comprehension gaps in education by providing simplified, personalized learning support. Future improvements should focus on offline deployment, multilingual capability, and adaptive learning features to further localize and expand the system’s educational impact.