New York, Dec 15 (IANS): Just by observing how users respond to online videos and other social media content, computers may be able to group them into marketing segments in real time, a new research has revealed.

"Computers used information from social media accounts to automatically build marketing personae. Marketing research professionals typically create these personae to help editors and marketers better understand the behaviours of specific consumer groups," said James Jansen, Professor at Penn State University.

But it is hard to make a decision looking at a bunch of complex numbers that most people do not understand.

To understand these complex numbers, researchers took a bunch of market data and condensed it into a fictitious person.

From 188,000 subscribers of a news website, the team developed algorithms to analyse data, such as demographic information, topics of interest and customer interactions. 

The researchers identified unique ways that groups of people were interacting with the information.

"News site editors could use this information to better collect and target content to these audiences," said Jansen, adding "Journalists can use that information to reach readers with better titles, content and article framing."

The researchers said that the method was transferable to other domains as well and could work at any consumer touch-point.

The method is being scaled up to millions of users where researchers could use other types of social media data from sites like Facebook or Twitter.

The study was presented at the second international workshop on online social networks technologies held in Agadir, Morocco, recently.