Generative AI-based ChatGPT and AI chatbots for learning AI education

• Main Author: Chin, Woon Ping
• Other Authors: Chee Wei Tan
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2024
• Subjects: Computer and Information Science; Artificial intelligence
• Online Access: https://hdl.handle.net/10356/176015
• Institution: Nanyang Technological University

Recent developments in large language models such as ChatGPT is becoming popular with businesses in different sectors as they seek to leverage on its capabilities to improve customer’s satisfaction. Most chatbots created for businesses today are pre-trained on a set of domain knowledge which are often engineered to cater for frequently asked questions in their specific domain. By applying the same idea to the education sector, the objective of this study is to examine how an existing chatbot can be further enhanced through integration with large language models to help students taking subjects in artificial intelligence and machine learning as a learning tool. Large language models such as ChatGPT and the machine learning algorithm of AlphaGo will be studied to understand how education in programming can be better taught through conversational and gamification methods to students. Experiments and research were done on ChatGPT to understand the limitations and potential use cases. The results of the experiments were impressive although there are still areas for fine tuning to cater for the education domain better. AlphaGo algorithm also gave possibility that gamification can be made available on the chatbot and will help students to learn better. The algorithm provides the opportunity to encourage students to derive better solutions to their problems. From this study, it is established that chatbot can serve as a viable educational tool for student gradually with fine tuning to the domain knowledge.