Big data sentiment analysis and experience optimization of health tourism based on deep learning and NLP

With the rise of health tourism, tourists’ requirements for tourism experience are increasing daily, and sentiment analysis has become an important means to optimize the experience. Based on deep learning and natural language processing (NLP) technology, this study constructs a big data sentiment analysis model of health tourism, aiming to provide data support for experience optimization by analyzing tourists’ online reviews, mining emotional tendencies, and providing data support for experience optimization. The study first collected millions of reviews related to health tourism from major travel platforms. After data cleaning and preprocessing, deep learning models such as Word2Vec and BERT were used for feature extraction, and emotion dictionaries and machine learning algorithms were combined for emotion classification. The experimental results show that the emotion classification accuracy of the proposed model reaches 92.3%, which is nearly five percentage points higher than that of the traditional method. Furthermore, through the results of sentiment analysis, the research identifies the key factors that affect tourists’ experience, such as service quality, environmental facilities, and health effects, and puts forward targeted optimization suggestions. In addition, the study also found that by implementing optimization measures, tourists’ satisfaction increased by 8.7%, and their willingness to revisit increased by 15.2%. This study provides an effective sentiment analysis tool for the health and wellness tourism industry and a scientific basis for improving the tourist experience and promoting the development of health and wellness tourism.