Hybrid Learning Dynamics
Said Benlakhdar, Karim Lkamel, Mohammed Rziza, Rachid Oulad Haj Thami · 2024
Large-scale language models (LLMs), such as Chat Generative Pre-trained Transformer (ChatGPT), have garnered significant attention from researchers globally. We have introduced an innovative hybrid approach that combines ChatGPT with teacher interaction. In this chapter, our aim was to evaluate the performance of our hybrid model in comparison to ChatGPT learning and the traditional learning model. The evaluation takes place within the framework of developing a hybrid educational system for intelligently designing learning materials. The analysis revealed noteworthy variations in overall scores among participants, highlighting significant differences between all pairs of groups. Furthermore, our results demonstrate that hybrid learning produces remarkable outcomes compared to other existing learning approaches. Thus, a thorough regression analysis provides confirmation regarding the significance of the chosen methods and the computer literacy of participants in predicting assessment scores. These findings advocate for a nuanced integration of artificial intelligence (AI) with human expertise, fostering dynamic interactions that reshape how we address challenges in education.