Visitor Experience of the Grand Canal National Cultural Park Museum Based on Sentiment Analysis Algorithm

This study investigates visitor experiences at the Grand Canal National Cultural Park Museum through sentiment analysis of online reviews collected. The analysis aimed to assess visitor sentiments, identify key themes influencing visitor perceptions and evaluate the performance of sentiment analysis classifiers-Support Vector Machine (SVM), Naive Bayes, and Random Forest-in categorizing sentiments. A total of over 50,000 reviews were analyzed using natural language processing techniques, revealing that 60% of reviews expressed positive sentiments, highlighting the museum's success in providing a culturally enriching experience. Thematic analysis identified exhibit quality, visitor services, staff interactions, and ambience as critical themes shaping visitor experiences. Positive feedback predominantly praised the museum's well-curated exhibits, interactive displays, and educational value, while criticisms focused on exhibit maintenance and occasional service lapses. The SVM classifier demonstrated the highest accuracy of 87%, outperforming Naive Bayes (84%) and Random Forest (86%) in sentiment classification tasks. Precision, recall, and F1-score metrics further validated SVM's effectiveness in accurately categorizing sentiment from visitor reviews. The study's findings suggest opportunities for enhancing visitor satisfaction through improvements in exhibit maintenance, visitor services optimization, and ambience management. These insights provide actionable recommendations for museum management to sustain positive visitor experiences and foster a welcoming environment. Future research could explore longitudinal studies, cross-cultural comparisons, and the impact of digital engagement strategies on visitor perceptions and museum experiences.