Scholarlink

User-Generated Content (UGC), such as reviews and comments posted on platforms like Tripadvisor, Google Reviews, and Instagram, can be a valuable tool for understanding tourist experiences and preferences. In this paper, we propose a recommender framework for Destination Management Organizations (DMOs) using a Large Language Model (LLM) to analyze UGC from 51 key attractions on the Greek island of Kefalonia from Tripadvisor. The dataset included the total number of reviews with a rating of 1–3 posted by users up to February 2025. Our primary aim was to identify areas of dissatisfaction. The LLM analyzed each review to create tailored, actionable suggestions for DMOs, based on visitor type, time, and location. The reviews were then grouped by place and theme, filtered for relevance, and combined into a detailed yet focused strategy to support more personalized destination management. The proposed framework demonstrates the potential of integrating user feedback into destination management practices to foster more proactive and tourist-centered services.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A Recommender Framework for Destination Management Organizations (DMOs) Based on User-Generated Content

  • 会议论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年

查看全文
  • Agisilaos Konidaris,
  • Athanasios Tsipis,
  • Aggeliki Sgora,
  • Erato Koustoumpardi

摘要

User-Generated Content (UGC), such as reviews and comments posted on platforms like Tripadvisor, Google Reviews, and Instagram, can be a valuable tool for understanding tourist experiences and preferences. In this paper, we propose a recommender framework for Destination Management Organizations (DMOs) using a Large Language Model (LLM) to analyze UGC from 51 key attractions on the Greek island of Kefalonia from Tripadvisor. The dataset included the total number of reviews with a rating of 1–3 posted by users up to February 2025. Our primary aim was to identify areas of dissatisfaction. The LLM analyzed each review to create tailored, actionable suggestions for DMOs, based on visitor type, time, and location. The reviews were then grouped by place and theme, filtered for relevance, and combined into a detailed yet focused strategy to support more personalized destination management. The proposed framework demonstrates the potential of integrating user feedback into destination management practices to foster more proactive and tourist-centered services.