So a pair of scientists have created a trading program that is fed massive quantities of financial text (business articles) and can trade based on this text in a methodical way. The machine trader is called The AZFinText system, and it has been programmed to parse the actual language contained in press releases and then act on this language by buying and selling stocks.
It turns out that certain verbs, contained in the text that the system analyzed, tended to lead to down stocks. Other specific verbs led to up stocks.
From MIT’s Technology Review:
The five verbs with highest negative impact on stock price are hereto, comparable, charge, summit and green. If the verb hereto were to appear in a financial article, AZFinText would discount the price by $0.0029. While this movement may not appear to be much, the continued usage of negative verbs is additive.
The five verbs with the highest positive impact on stock prices are planted, announcing, front, smaller and crude.
The rest of the piece is an interesting read. The gist of the article is that this artificial intelligence can outperform professional stockpickers, but that’s the type of silly headline that is used to get attention. The author of the story picks a random timeframe to compare AZFinText to some other index funds to make this claim. You can ignore that aspect of it.