Decoding Public Sentiments: A Text Mining Analysis of National Fitness Empowering by Digital Intelligence

Abstract

This study focuses on the public sentiments of national fitness empowering by digital intelligence (DI), and analyzes the data of 14086 comments on social media platforms Weibo. By combining the comprehensive research methods of TF-IDF, Naive Bayes sentiment classification model and LDA topic model, the results show that positive sentiments (accounting for 85.1%) focus on the technological advantages (such as AI, data function) and service upgrading of intelligent fitness, neutral sentiments (3.0%) focus on rural promotion, consumption costs and other popularization issues, and negative sentiments (12.0%) expose facility function defects and management contradictions. This study expands the field of digital intelligence empowerment of national fitness, aiming to provide support for the industry's precise optimization of services and policy design, and help the high-quality development of the digital intelligence fitness industry.