Designers should target investment platforms, design author rankings using comment counts, as it effectively identifies top analysts.
About this paper
The author studies the identification of experts in SeekingAlpha and StockTwits by analyzing datasets from these platforms.
They find that while general content has minimal correlation to stock performance, specific expert content can be identified through user interactions and significantly outperform broader markets.
Here are some methods used in this study:
Which part of the paper did the design guideline come from?
“We evaluate trading strategies generated using a combination of author ranking heuristics and long vs. long/short trading strategies. For our author ranking heuristics, we use average return per article (PerA), average return per stock (PerS), number of total comments (AllCom), and average comments per article (AvgCom). As described above, we choose 500 stocks mentioned by the top ranked authors, and split the funds of a hypothetical portfolio evenly among them. Each week, we trade them based (...)” (‘Empirical Evaluation’ section)
Wang, G., Wang, T., Wang, B., Sambasivan, D., Zhang, Z., Zheng, H., & Zhao, B. Y. (2015). Crowds on Wall Street. Proceedings of the 18th ACM Conference on Computer Supported Cooperative Work & Social Computing.