The purpose of this study is to provide empirical implications for satisfying users' demands and contribute to vitalization of public sports facilities by analyzing Google Map reviews on public sports facilities in various ways and understanding users' perceptions and evaluations of public sports facilities. In order to achieve this research purpose, text mining techniques were applied. Among public sports facilities, gymnasiums were selected for analysis. Using Python, 23,356 reviews of 533 living gyms nationwide were collected. The collected data went through pre-processing processes such as tokenization and terminology processing. Frequency analysis, semantic network analysis, topic modeling, and emotional analysis were conducted. As a result of analyzing 23,356 reviews of 522 gyms, 9,163 words were derived. As a result of conducting the frequency analysis, 'good' was analyzed to be the most, followed by swimming, facilities, exercise, cheap, parking, clean, use, wide, and program. Semantic network analysis was conducted to identify the network structure and meaning between major keywords. As a result of conducting topic modeling, reviews of gyms nationwide were classified into three topics: 'facility', 'swimming pool', and 'program and service'. As a result of conducting an emotional analysis, it was found that there were much more positive reviews with 89% of positive emotions and 11% of negative emotions.