<![CDATA[Ruang Terbuka Hijau (RTH) play a crucial role in urban environments, not only supporting nature conservation but also fostering social interaction and contributing to economic growth. City parks, as representations of GOS, can be enhanced through thematic development and place branding to improve user engagement and functional value. This study aims to analyze visitor perceptions of thematic parks by identifying high-frequency keywords extracted from user-generated reviews. Text mining techniques were employed using Term Frequency-Inverse Document Frequency (TF-IDF) and Term Frequency-Relevance Frequency (TF-RF) methods, followed by text summarization using Cosine Similarity and Maximum Marginal Relevance (MMR). These methods effectively process large volumes of unstructured data to reveal meaningful insights. The analysis focused on three parks with the highest number of reviews: Alun-Alun Kota Malang (945 reviews), Taman Merjosari (552 reviews), and Alun-Alun Tugu (462 reviews). Keyword analysis showed prominent terms such as ‘tugu’, ‘olahraga’ (sports), and ‘anak’ (children) under both TF-IDF and TF-RF methods, with TF-RF emphasizing more context-specific vocabulary. Results indicate that Alun-Alun Tugu is perceived as a comfortable space near government offices featuring a lotus pond, Taman Merjosari is recognized for its sports facilities, and Alun-Alun Malang is identified as a child-friendly park with fountains. The study offers place branding recommendations by analyzing word associations in summarized user feedback. This study can contribute valuable insights for governments, architects, and urban designers in developing thematic parks that better reflect and accommodate user preferences.]]>
