Stocktwits dataset

Our dataset comes from StockTwits, a social media platform designed specifically for investors and traders to share insights. Users can post messages related to individual stocks, each of which has its own discussion thread for comments. These messages are often tagged with sentiments like ``bullish" or ``bearish," reflecting the user’s outlook on the stock. The collected data contains a vast amount of valuable information, with each tweet corresponding to a user’s opinion about a specific stock. For each tweet (record) in our dataset, the data includes the tweet content, the stock it refers to (including its associated industry, sector, symbol, and the exchange where it is listed), the creation time of the tweet, the user ID, the number of likes it received, the sentiment expressed by the user (either bullish or bearish), and the retweet information.

我们的数据集源自StockTwits——一款专为投资者与交易员搭建的投资见解分享社交媒体平台。用户可发布与个股相关的内容，每支个股均设有专属的评论讨论串。这类内容通常会被标记为「看多（bullish）」或「看空（bearish）」，以体现用户对该个股的后市展望。本次采集的数据集蕴含大量高价值信息，每条推文（tweet）均对应一位用户对特定个股的观点。针对数据集中的每条推文（记录），其包含的字段包括：推文正文内容、所关联的个股信息（含该个股所属行业、板块、股票代码及上市交易所）、推文发布时间、用户ID、获赞数量、用户表达的多空情绪（仅分为看多或看空两类）以及转发相关信息。