소셜미디어 빅데이터를 활용한 여성 피트니스대회의 토픽모델링분석

초록

This study aims to identify the major issues surrounding women’s fitness competitions by utilizing social media big data and applying text mining techniques. Data were collected using the keyword “women’s fitness competition” from blogs, online communities, news articles, and web documents provided by major Korean portals such as Naver, Daum, and Google. Text mining techniques including Term Frequency (TF) analysis, word cloud visualization, TF-IDF analysis, and topic modeling were applied to the collected data. First, the TF analysis identified the top 60 frequently mentioned keywords, such as “world championship,” “fitness competition,” and “workout volume,” while a word cloud visualization was created based on the top 500 keywords. Second, TF-IDF analysis revealed the top 30 keywords, including “workout volume,” “athlete,” and “bodybuilding,” which highlight distinctive and meaningful terms. Third, topic modeling analysis extracted five major topics, which were categorized and named as follows: (1) “International Competitions and Professional Female Athletes’ Activities,” (2) “Domestic Competitions and Celebrity Making of Female Fitness Athletes,” (3) “Body Management Strategies of Female Fitness Athletes,” (4) “Challenges and Success of Female Fitness Athletes,” and (5) “The Beautiful Body Image of Female Fitness Athletes.” These findings provide a comprehensive understanding of the discourse and perceptions surrounding women’s fitness competitions as observed in social media spaces.

키워드