Social media is a large but still rapidly growing means of communication and efficient global interaction. Many of these conversations discuss companies, stocks and market activity, which makes them ripe for sentiment analysis in trading.

This is especially the case for Twitter, the lingua franca of social media.

Social Market Analytics, a software company, produces a family of metrics designed to capture the signature of financial market sentiment as expressed through tweets.

For each stock in SMA’s universe, its servers poll Twitter’s API to capture “indicative tweets” about the stock observed during a time sample window. “Indicative” means tweets that pas SMA’s filtering processes and are used in sentiment estimates.

The collected tweets are then filtered for financial trading relevance and scored for market sentiment content.

“Our process is unique in this emerging field both in its approach to filtering social media data and in the analytical methodology used to develop our proprietary metrics,” said Joe Gits, CEO of SMA.

SMA computes Sentiment Factors (S-Factors) for a large number of securities. Positive S-Scores levels are associated with favorable market sentiment, while negative levels are unfavorable.

For example, S-Core levels greater than 2.0 indicate that Twitter users are posting comments and sharing information with high positive content about a stock, such as might be expected from rumors of and reaction to an exceptionally strong stock buyback offer or acquisition.

“We expect changes in market sentiment, as measured by changes in S-Score, to be reflected in stock price changes over some time horizon,” Gits said. “Predictive analytics applied to social media is a new, rapidly evolving technology, offering opportunities for innovation.”

MIT Media Lab researcher Yaniv Altshuler, an expert in collective intelligence methods, has developed a tool for social financial trading that helps guide users to make better decisions by improving the information flow within the networks.

This is accomplished by diverting the traders’ attention away from certain links, and drawing their attention to others, changing the dynamics of the network.

Working with the eToro investment network, Altshuler distributed $20 trading coupons to 500 active financial traders out of the more than two million eToro members. Matches between traders and recommendations were based on an innovative algorithm designed to optimize information flow within the network.

“This study demonstrates how an efficient collaborative trading community can be formed by carefully balancing the complex mixture of ‘trend setters’ and ‘bellwethers’ who govern the behavior of the crowd,” said Altshuler.

Even this small number of coupons was enough to move the entire network away from dangerously high levels of “groupthink,” and as a consequence, the entire trading community–not just the 500 coupon users–saw a significant increase to their rate of return.

The increase in return was more than 10 percent compared to those who traded without guidance from the social network, and 4 percent higher than those who only followed the highest-performing gurus.