Big media is getting even bigger, or should we say, wiser. Today, Reuters launched a new “sentiment analysis” service in which computers read the news, judge whether it is positive or negative, and trigger trades based on the results. The service is being marketed to algorithmic traders, including hedge funds.

Reuters claims the system will enable customers to analyze news about thousands of companies before humans can even read the first headline of a newspaper. The information provider has already been offering a product called NewsScope, which allows its clients to use news content to drive automatic trading and respond to market moving events, but according to Peter Moss, who heads Reuters global enterprise solutions group, the new release allows the machines to “interpret the sentiment of news stories as they are published.”

For now, the system will only scan Reuters’ own articles, but there are plans to add additional news sources.

According to Reuters, the new system works by assigning numerical “sentiment scores” to words or phrases which are then processed to give an overall positive, neutral or negative score to the company in the news article. These scores can be added together to calculate the prevailing sentiment for a company, a sector, an index, or even to assess global market sentiment.
“Imagine a machine scanning hundreds of stories on companies’ results, measuring the sentiment around them and incorporating that into algorithmic trading strategies,” said Moss.

Reuters collaborated with U.K.-based linguistics and software developer Corpora to develop the new system.