Enhancing a Hybrid Machine Components Course With a 24/7 AI Chatbot for Improved Student Support

Abstract

Abstract This study investigated the impact of integrating an AI chatbot into a hybrid flipped classroom environment to support student learning in a Machine Components course. The intervention aimed to address challenges common in online learning, such as student isolation and lack of immediate access to instructor support. A commercially available AI chatbot platform (Poe) was trained on specific concepts from the course textbook, focusing on failure theories and fatigue. A sub-study was conducted to optimize the chatbot’s training by comparing the effects of different training materials (explanations vs. limited solutions) on its problem-solving guidance approach. The sub-study revealed the chatbot’s tendency to provide direct answers to calculation-based problems regardless of the training set. The main study compared the performance of students who had access to the AI chatbot (intervention group) with those who did not (control group) on an exam question related to fatigue. Results showed that the intervention group achieved significantly higher grades on the fatigue-related question. This suggests that the AI chatbot effectively supported student learning by providing immediate access to explanations and guidance, potentially mitigating the challenges of isolation and limited instructor availability. Integrating AI chatbots holds promise for enhancing learning outcomes in engineering courses within online and hybrid environments.

Ähnliche Arbeiten