A Framework for Developing Customized Museum Guide Systems Leveraging Large Language Models

The recent advancements in large language models (LLMs) and artificial intelligence (AI) technology have brought about a transformative change in the way information is searched, processed, and delivered. In the case of art museums, this has opened up the possibility of providing personalized information and services to visitors in a conversational manner. Existing IT-based museum guide systems have primarily relied on voice guidance, smartphone applications, and chatbots. However, these systems mostly provide pre-prepared information unilaterally or allow only very limited interaction with users. Additionally, individual museums have faced challenges in terms of cost and technology in implementing and maintaining such systems. This study aims to address these issues by proposing a museum guide system framework based on customized LLM technology. The framework is designed to provide visitors with exhibition information, personalized recommendations, and supplementary information for a comprehensive cultural experience in a conversational manner. The development of the framework employed the design science research methodology, and a virtual museum was constructed to validate and demonstrate the framework’s potential using ChatGPT 4.0 and the GPTs platform. The results suggest that this framework can overcome the limitations of existing museum guide systems and provide visitors with an enhanced cultural and artistic experience.