Designing AI-Based Personalized Viewing Paths for Art Exhibitions : focusing on the Alphonse Mucha Exhibition

This study proposes a method for designing and generating personalized viewing paths for art exhibitions using artificial intelligence (AI). A structured, dual-table knowledge base, composed of a 'Visual Element Deconstruction Table' and a 'Path Node Logic Table', was constructed by deeply deconstructing the visual elements and inherent meanings within the works of Alphonse Mucha. Based on this, a strategy was designed for an AI to generate personalized paths for five core themes by intelligently combining knowledge nodes according to user personas and specific prompts. The research confirmed that the proposed framework can effectively generate viewing paths that are logically coherent, thematically clear, and rich in content. The significance of this study lies in its presentation of a concrete methodology for the contemporary museum field that integrates AI, art analysis, and personalized guidance, exploring the applicability of AI in enhancing curatorial efficiency and deepening the visitor experience.