UK and US researchers looked at how often pages describing 30 companies listed in the Dow Jones Industrial Average - such as Procter & Gamble, Bank of America, and The Walt Disney Company - were viewed between December 2007 and April 2012.

They found that an increase in the number of page views for a firm was followed by a decline in its share price. A simple trading strategy tested, based on changes in the frequency of views, would have led to significant profits of up to 141%, claims the paper.

The researchers also devised a trading strategy based on the number of page views for 285 more general finance-related topics, such as 'macroeconomics', 'capital' and 'wealth'. This proved even more successful, generating potential profits of up to 297%

In contrast, measuring how often the Wikipedia pages were edited provided no indication of future market movements while devising strategies based on non-finance related pages - instead using actors and film makers - was also fruitless.

Suzy Moat, co-author, says: "These results provide evidence that online data may allow us to gain a new understanding of the early stages of decision making, giving us an insight into how people gather information before they decide to take action in the real world."

Last month, a paper from the same team found that analysing how often key finance-related words are searched for using Google could help predict stock market moves.

Previous studies have also shown that Twitter chatter can also be used to predict the ups and downs of the markets, with a $25 million hedge fund based on one sentiment algorithm launched in 2011, although it closed down within weeks.

You can read the full paper, from Moat and Tobias Preis, both of Warwick Business School, and Chester Curme, Adam Avakian, Dror Kenett and Eugene Stanley, of Boston University, here.