【1976 Archives】

There's a new stock market rhythm emerging: Donald Trump tweets about a company976 Archives and its stock goes straight down.

The entire thing happens in a matter of seconds, which on Wall Street is plenty of time for thousands of shares to be bought and sold. It happened to Boeing and biotech companies last week. On Monday, it was Lockheed Martin and its F-35 military jet program that ended up in Trump's crosshairs.

In about an hour, the company was worth about $4 billion dollars less.

This Tweet is currently unavailable. It might be loading or has been removed.

To stock traders, that's a market opportunity. If you can see a tweet and know a stock is about to go down (or up), you can make a nice chunk of change. The question is just how fast can you see the tweet, understand it and then execute the right strategy.

This is where the computers come in. Combine cutting-edge artificial intelligence that can (sort of) understand human language with advanced trading algorithms and you can conceivably make some money off of Trump's tweets.

Which is not to say that it would be a good idea.

"If one of my algo strategists came to me and told me he wanted to build a computerized strategy to trade based on Trump tweets, I'd respond 'you're fired,'" said Manoj Narang, a pioneer in high-speed trading, told Mashablein an email.

"What I'm saying is that tweets in general, even market-moving ones by authority figures like Trump, are very challenging for machines to interpret," Narang added.

Not that people aren't going to try. Using Twitter to get an edge in the stock market is a concept that has been around for at least a few years. Twitter is already one of many information sources that are culled by programs that have been developed to read text and gain a basic understanding of the content therein.

Some investment firms go a step further, eliminating much of the human touch and allowing programs to read tweets and then execute trades. The most high-profile example of this practice came in April 2013 when the Twitter account of the Associated Press sent out a tweet saying that a bomb had gone off at the White House. The account had been hacked, but nobody at the time knew that. The tweet sent the stock market plummeting, a move blamed primarily on this computerized trading.

The computer programs that have been developed to parse context and sentiment from human language have come a ways since then, with some able to comb through massive troves of documents and identify complex concepts such as fraud.

This Tweet is currently unavailable. It might be loading or has been removed.

These programs excel in dealing with large volumes of data in part because they rely on being able to hone their systems over and over again. When it comes to something short with a relatively small sample size — like the tweets of one president-elect — limitations start to emerge.

"There is this research phase that needs to happen overall, and I suppose that's maybe part of the reason that Trump hasn't been calibrated in models yet. You need the data to analyze," said Dan Joldzic, president of Alexandria Technologies, a firm that uses artificial intelligence to classify news for financial analysts. "Trump's negative statement on Boeing negatively reflected on Boeing's stock, but that could be one of a thousand statements he's made on Boeing."

There's also the actual tweets, which can sometimes even be difficult for a person to understand let alone a computer.

"NLP (natural language processing) is not in a particularly advanced stage, and the kinds of cryptic things you see on Twitter make even less sense to computer algorithms than ordinary language," Narang said.

That won't stop some people from trying, Joldzic noted, particularly if Trump continues to tweet about companies in a negative way.

"It could very well be that they hadn't calibrated their models for him or they hadn't filtered him out as someone to follow more closely, but it's likely going to happen going forward," Joldzic said.

Just because it might not be easy to make money off Trump's Twitter tirades doesn't mean there isn't a way to benefit. As inauguration day approaches, some, like CNBC's resident stock picker Jim Cramer, are wondering if the only way to play this game is to not play at all.

This Tweet is currently unavailable. It might be loading or has been removed.