My friend, @rwl4, who is reluctant to blog here for some reason, wrote:

> Seems like if you setup a trigger system that whenever Trump tweets, it runs tweet through a bayesian filter programmed to detect either hostile or friendly language, you could automatically buy or sell the relevant stock and make some decent gains. (Of course, in this case Nordstrom isn't a public company.)
> 
> It would see words like "terrible" and "unfair" and that would help classify it.

---

<div class="pull-right">
  <img src="https://external-ord1-1.xx.fbcdn.net/safe_image.php?d=AQCx4xShH42AFEWv&amp;w=158&amp;h=158&amp;url=https%3A%2F%2Fpbs.twimg.com%2Fprofile_images%2F1980294624%2FDJT_Headshot_V2_400x400.jpg&amp;cfs=1&amp;upscale=1&amp;sx=0&amp;sy=0&amp;sw=400&amp;sh=400&amp;_nc_hash=AQDKSTv6yVUVZqdx" />
</div>

> [**Donald J. Trump on Twitter**](https://twitter.com/realDonaldTrump/status/829356871848951809) - "My daughter Ivanka has been treated so unfairly by @Nordstrom. She is a great person -- always pushing me to do the right thing! Terrible!"
> &nbsp;
> &nbsp;

---

I wonder if you'd have to calibrate the bayesian filter to deal with misspellings and verbiage that is wrongly interpreted by Trump's assistant (who does all of the tweeting for him).