Using the social media to predict the stock market sentiments is a bizarre task but it is not possible to ascertain the moods or emotions of investors by mining the huge chunk of data sets, especially from Twitter.

But, Indiana University researchers have recently said using the collective mood of the Twitter community – as measured by millions of twitter updates, to predict the upcoming moves in Dow Jones Industrial Average is possible. Citing the research by Indiana University Professor Johan Bollen, they said their algorithms can predict the direction of Dow with 87 percent accuracy, up to four days before the actual move takes place in the market.

The stock market sentiments are primarily driven by human emotions like greed and fear but the market movements are not solely governed by normal beliefs. If the public is excessively bullish, it's time to be cautious. If it's not, it might be time to snap up a bargain.

The captured emotions on social media like Twitter cannot be used as a real-time reflection of investor sentiment as the vast majority of accounts aren't even verified. Most of the verified accounts reveal the inappropriate or irrelevant data which will not be helpful to get any investment gains.

Dow Jones Industrial Average Index
A screen displays the Dow Jones Industrial Average on the trading floor of the New York Stock Exchange.

In 2009, Brunswick Group conducted an online survey to trace out the role of social media in investment decisions. Out of the surveyed 500 institutional investors in the US and Europe, only 4 percent acknowledged that social media had an influence in their investment decisions or recommendations.

A handful of fund managers are betting on emotions and recommending that it is going to be the future of trading which is a startling mechanism and it is quite illogical that an individual investor can make any sense out of the collective tweets of 100 million active users.

Mining the relevant tweets to gauge the global mood from billions of tweets in real time requires highly complex mathematical models and algorithms.

Sole dependence on twitter signals without at least comparing with other trading models that are already built and proven would not be a prudent decision while investing.