Negative Public Opinion Data on Chinese Marathons from 2020 to 2025 (LDA Topic Analysis, Sentiment Analysis, Web Crawling, etc.)

A web crawler program was developed using the Scrapy framework, targeting keywords such as "marathon events," "marathon event evaluations," and "marathon public sentiment" to scrape relevant topics and comments from the Sina Weibo platform. The data collection spanned from January 1, 2020, to December 31, 2025, resulting in a total of 11,045 raw text data entries. Subsequently, negative public sentiment content was filtered out through sentiment analysis. Finally, cluster analysis was conducted using the LDA model, with the number of topics determined to be seven.