Scholarlink

<p>Tourist satisfaction in wellness destinations is increasingly shaped by user-generated content on social media, yet limited evidence exists on how specific aspects of wellness experiences drive sentiment in particular locations. This study uses aspect-based sentiment analysis to evaluate tourist satisfaction and the perceived importance of wellness treatments, cultural experiences, and service-related attributes in the wellness destinations of Granada (Spain) and Algarve (Portugal). A dataset of 2613 X posts, 7712 Instagram posts, and 1850 TripAdvisor reviews collected between 2018 and 2024 was pre-processed for grammar-based extraction of aspect–opinion pairs, which were then analysed using the VADER sentiment model and grouped with k-means clustering in the aspect embedding space. Model performance was quantified through accuracy, macro-precision, macro-recall, and macro-F1 and compared on this dataset with alternative sentiment analysis approaches based on TextBlob, a bidirectional LSTM classifier, and a transformer architecture. The VADER–k-means configuration achieved an accuracy of 96.02%, with macro-precision of 0.9847, macro-recall of 0.9149, and macro-F1 of 0.9434, outperforming the baseline models in classifying sentiments related to wellness tourism. Clustered aspects revealed coherent groupings of spa and wellness services, natural landscapes, cultural heritage, and hospitality, with spa quality, staff responsiveness, and environment-related features most frequently associated with positive sentiment. These results demonstrate the value of social media–based sentiment analysis for complementing traditional survey approaches and provide data-driven guidance for destination managers seeking to enhance wellness tourism offerings in Granada and Algarve.</p>

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

Social media sentiment analysis of tourist satisfaction in the wellness destinations of Granada and Algarve

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年2月6日

查看全文
  • Shi Yong Fei

摘要

Tourist satisfaction in wellness destinations is increasingly shaped by user-generated content on social media, yet limited evidence exists on how specific aspects of wellness experiences drive sentiment in particular locations. This study uses aspect-based sentiment analysis to evaluate tourist satisfaction and the perceived importance of wellness treatments, cultural experiences, and service-related attributes in the wellness destinations of Granada (Spain) and Algarve (Portugal). A dataset of 2613 X posts, 7712 Instagram posts, and 1850 TripAdvisor reviews collected between 2018 and 2024 was pre-processed for grammar-based extraction of aspect–opinion pairs, which were then analysed using the VADER sentiment model and grouped with k-means clustering in the aspect embedding space. Model performance was quantified through accuracy, macro-precision, macro-recall, and macro-F1 and compared on this dataset with alternative sentiment analysis approaches based on TextBlob, a bidirectional LSTM classifier, and a transformer architecture. The VADER–k-means configuration achieved an accuracy of 96.02%, with macro-precision of 0.9847, macro-recall of 0.9149, and macro-F1 of 0.9434, outperforming the baseline models in classifying sentiments related to wellness tourism. Clustered aspects revealed coherent groupings of spa and wellness services, natural landscapes, cultural heritage, and hospitality, with spa quality, staff responsiveness, and environment-related features most frequently associated with positive sentiment. These results demonstrate the value of social media–based sentiment analysis for complementing traditional survey approaches and provide data-driven guidance for destination managers seeking to enhance wellness tourism offerings in Granada and Algarve.