News sentiment analysis tools still not yet ready to embrace smatter

Photo courtesy of Robin Hawkes on Flickr

Photo courtesy of Robin Hawkes on Flickr

Rob Swystun, Pristine Advisers

Tools that monitor news to gauge general sentiment about companies continue to gain popularity, but they’re not quite ready to jump into the murky social media pool just yet.

With no actual accountability on social media, it just makes it too untrustworthy at this point to gather information from for the purposes of news sentiment.

In a nutshell, news sentiment tools gather information from trusted news sources and analyze the data to find out if there is a general positive or negative sentiment about a company in the news, making them useful as a trading indicator for hedge funds and asset managers.

The tools already have a difficult enough time distinguishing between Apple the company and apple the fruit, reports Becca Lipman on InformationWeek’s WallStreet & Technology, so wading into the information free-for-all of social media is a daunting task. And what’s making it even more daunting is that news sentiment tools are getting more sophisticated, meaning they are able to pick up more than just a general positive or negative sentiment about a company. They soon will be able to identify emerging relationships among tradable assets.

They’ll do this by flagging instances when firms, people or brands that are normally not associated with each other get mentioned together in the news. These flags will tell users there is a situation worth analyzing to find out why the mentions are happening, how frequently these mentions are becoming and in what context they are happening.

“Most importantly, advances in event detection capabilities are enabling quant analysts and data scientists to add context to a story, anticipate market reaction to news, or predict creation of a business contract,” Lipman says.

An example: news sentiment software may flag announcements of layoffs, and while layoffs are typically seen as a bad thing, if the software also flags mentions of factors leading to layoffs, analysis of the data may reveal that layoffs done under certain conditions causes the market price to go down while layoffs done under certain other conditions cause it to go up. Over time, theoretically, there should be enough data to show typical correlations, which will allow a person looking at the data to understand how important news is and to make better forecasts for a specific firm or country.

Oh, and all this is done fast, as in: 200 milliseconds to read, analyze, and distribute information in real-time across 1 million to 2 million news stories per day fast.


As more hedge funds and asset managers look to add buzz, abnormalities, and sentiment to their multi-factor trading models, news sentiment analysis tools are reaping the rewards.

Armando Gonzalez, CEO of real-time financial news analysis service RavenPack, says he has seen a growth rate of 30 – 40% per year for news sentiment products for the past three years. RavenPack also has a 90% renewal rate.

“It continues to grow,” Gonzales says, “because this type of factor is becoming more accepted as a trading tool.”

In order to ensure that buy-side analysts can trust the findings of news sentiment services, services like RavenPack stick to trusted sources like The Wall Street Journal, PR Newswire, Bloomberg, and Zero Hedge. They tend not to bother with smatter (social media chatter), which has zero accountability and a dodgy reputation as far as actual news goes.

“I have strong views on why not to use Twitter and Facebook yet,” Gonzalez says. “Customers want accountability. We can take a bunch of news from Barrons, from reporters who are investing resources to report and will be held accountable for that news, people who will be there tomorrow to back it up. With Twitter and Facebook there is no accountability, no trust, you’re not able to prove the information.”

Although Twitter is often touted as being a good source for breaking news, Gonzales points out that anyone can put up false information and see it spread quickly, but finding and verifying the original source can be tricky, plus computer algorithms can misinterpret data.

For example, people on social media use the word “earthquake” far more than earthquakes actually happen in messages like “was that an earthquake?” or “feels like an earthquake.” So, a Twitter search for earthquakes in New York City would pull up a lot of tweets about them, but checking geological records would not reveal any correlation between the amount of times people talk about an earthquake and how often they actually happen there.

While social media can’t yet be trusted for gathering news sentiment, it is the next logical step, Gonzales notes. And platforms like StockTwits — essentially Twitter for the stock market crowd — do hold a lot of promise for news sentiment services. StockTwits users have incentive to keep their content valid, as they want to gain followers and trust themselves.

“It brings a bit more validity to the table,” Gonzalez says. “We haven’t yet found quantitative value in it, but it’s an exciting example of people sharing and being held accountable for opinions. I value that from a microblogging prospective.”


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s