Anime Short Text Data Set

The dataset for this study was acquired through a Python web scraping program developed by the authors. The dataset primarily originates from the Tencent platform, encompassing comments and danmaku (bullet screen) data from two animated works: "Sword Dynasty: The Fifth Season" and "Recognizing the Night, Depicting the Silver." These short text data include a rich array of information fields, specifically:1. Basic identifiers: id, userid, username2. Temporal information: posttime3. Content data: msgcontent4. Geographical location: country, province, city5. Interaction metrics: comments, forwards, exposures, likes To facilitate subsequent analysis and processing, all collected data were stored in a structured format as xlsx files. This storage method not only ensures data integrity and readability but also provides convenience for subsequent data analysis and topic identification tasks. Through this systematic data collection approach, we successfully constructed a comprehensive and valuable short text dataset. This dataset not only contains user-generated core content but also records relevant metadata and interaction information, providing a solid data foundation for in-depth research on audience reactions and thematic distribution in animated works.