This paper presents an Artificial Intelligence (AI)-driven framework designed to bridge the gap between raw user feedback and strategic decision-making. Moving beyond traditional sentiment analysis, which often overlooks the specific “why” behind visitor dissatisfaction, this research utilizes a sophisticated dual approach. By integrating the contextual precision of Bidirectional Encoder Representations from Transformers (BERT) with the generative reasoning of Large Language Models (LLMs) like Gemini, the system extracts fine-grained, aspect-based insights and actionable recommendations. The framework’s effectiveness is demonstrated through a case study of the Archaeological Site of Mystras. Ultimately, this work offers a scalable solution for tourism professionals and policymakers to listen more deeply to the authentic voice of the traveler.
Gkaripis et al. (Thu,) studied this question.