TweetTrader.net: Leveraging Crowd Wisdom in a Stock Microblogging Forum

TweetTrader.net is a stock microblogging forum that leverages the wisdom of crowds to aggregate the information contained in stock-related tweets. Based on insights from academic research on stock microblogs, the application integrates inputs from text classification, user voting and a proprietary Stock Game in order to extract the sentiment (i.e., the bullishness) of online investors with respect to all publicly traded companies of the S&P 500. A demo of TweetTrader.net is available at http://TweetTrader.net. Background of stock microblogging Popularity of stock microblogs Twitter has become a vibrant platform to exchange trading ideas and other stock-related information. Traders have adopted the convention of tagging stock-related messages (i.e. stock microblogs) with a dollar sign followed by the relevant company’s ticker symbol (e.g., “$AAPL“ for tweets related to Apple Inc.). The business press describes the conversations on Twitter as “the modern version of traders shouting in the pits“ (BusinessWeek 2009). There are investors who attribute their trading success to the information they find on social media websites and, moreover, financial professionals have developed Twitterbased trading systems to identify sentiment-based investment opportunities. As a result, the investor community has come to call Twitter and related third-party applications “a Bloomberg for the average guy“ (BusinessWeek 2009). Related academic research A few recent studies suggest that the information content of general Twitter messages, including those without a specific reference to the stock market, may help predict Copyright © 2011, Association for the Advancement of Artificial Intelligence (www.aaai.org). All rights reserved. macroeconomic indicators such as the Index of Consumer Sentiment (O’Connor et al. 2010) or stock market indices such as the Dow Jones Industrial Average (Bollen, Mao, and Zeng 2010) and the S&P 500 (Zhang, Fuehres, and Gloor 2010). Regarding the information content of microblogs with respect to individual stocks, Sprenger and Welpe (2010) find robust relationships between the sentiment of stock microblogs (i.e. their bullishness) and abnormal returns as well as message volume and trading volume. In addition, the authors offer an explanation for the mechanism leading to the efficient aggregation of information in microblogging forums by showing that users who provide above average investment advice are retweeted (i.e., quoted) more often and have more followers. In other words, retweets and followership may represent the Twittersphere's “currency“ and provide it with a mechanism to weigh information. The stock microblogging forum TweetTrader.net leverages the insights from this research and enables online investors to cut through the noise of thousands of daily messages. In contrast to a few related third-party applications such as StockTwits.com, which are limited to filtering the message stream by ticker symbol, TweetTrader.net taps the wisdom of crowds and aggregates the information in a meaningful fashion. Features of TweetTrader.net Tapping the wisdom of crowds TweetTrader.net uses the Twitter Search API to provide a Livestream of all tweets related to S&P 500 stocks. Users have a choice between all tweets or a subset of selected indices or industries. Bullish words (e.g., buy, upgrade, or growth) are marked in green, bearish words (e.g., sold, downgrade, or decline) are marked in red. Embedded within these basic functionalities are three main features that tap the collective wisdom of thousands of online investors: automatic text classification, user-driven sentiment voting and a Stock Game. 663 Proceedings of the Fifth International AAAI Conference on Weblogs and Social Media