Data-driven analysis of cultural tourism experience: emotional journey mapping of museum visitors in China

Amid deepening culture–tourism integration, optimizing museum visitor experience is urgent. Using 230,000 online reviews from 31 National First-Grade Museums in China, we propose a scalable Language–Touchpoint–Emotion–Process (LTEP) framework that fuses user-generated content (UGC) analytics with customer-journey mapping (CJM). Unlike existing CJM or sentiment models, LTEP automatically discovers touchpoints from unstructured language, models multi-stage emotion trajectories, and quantifies each touchpoint’s marginal effect via explainable machine learning. Application identifies 17 touchpoints across pre-visit, on-site, and post-visit stages. Results show an enduring asymmetry: content-core factors drive positive emotion on-site, while service-periphery factors depress overall experience. IPA locates high-importance/low-satisfaction gaps concentrated at the journey’s periphery. The framework supports continuous monitoring and cross-museum benchmarking, translating language into actionable journey metrics. These findings advance process-based theory of value co-creation in cultural services and offer data-driven guidance for aligning content attractiveness with operational effectiveness.