Customer satisfaction analysis and preference prediction in historic sites through electronic word of mouth

Cultural tourism is a continuously rapidly developing product which the global travel sector has experienced. Cultural products are a vital part of the economy and post-modern society. Satisfaction is a prominent factor in tourism and marketing literature. This paper aims to analyze customer satisfaction in historic sites through electronic word-of-mouth. We developed a new method through text mining, clustering and supervised learning techniques. The method is developed through latent dirichlet allocation for customer online reviews analysis, learning vector quantization to find important customers segments and Adaptive Neuro-Fuzzy Inference System for customer preference prediction in historic sites. The data are collected from TripAdvisor which is a comprehensive online review system in tourism and hospitality. The results revealed that electronic word-of-mouth (eWOM) effectively reveals customer satisfaction in historic sites through data analytical approaches.