Can social media be used to predict the future? Read the special issue of Internet Research

BRADFORD, U.K. - Oct. 25, 2013 - PRLog -- Every move we make on social media is being documented somewhere on machine readable formats – and now computer scientists are using this data to predict the future.

In the latest issue of Internet Research, researchers have explored how social media is providing computer and social scientists with an instantaneous, cheap and unstructured way of collecting data on a large scale.  By studying our online behaviour, we can now forecast - or even “nowcast” –  health scares, movie success or even results in political elections. This issue can be read for free until 30 November by visiting www.emeraldinsight.com/tk/prediction

The four articles in this special issue examine the different types of forecasting models and how they are being adapted to social media to help better understand (and predict) societal behaviours.  Citing as one example the 2008 bird flu outbreak, researchers in the US and the UK were able to build a statistical model that predicted the number of cases one week in advance based on the contents and locations of tweets associated with the epidemic at the time.

Other scenarios include how the evolution of blog posts over time exhibited positive correlation with book sales, weeks in advance of publication, and how predicting election outcomes using Twitter is becoming increasingly popular.

Guest editor, Harald Schoen, Professor at the University of Bamberg, Germany, comments: “If there is anything human experience has taught us, it is that predicting the future is both highly desirable and extremely difficult.  Social media provides an impressive amount of data about users and their interactions, thereby offering computer and social scientists, economists, and statisticians –among others– new opportunities for research”.

Published by Emerald Group Publishing, Internet Research is an international, refereed journal that aims to describe, assess and foster understanding of the role of wide-area, multi-purpose computer networks such as the Internet.  This special issue is published as Volume 23 Issue 5, and is freely available to read online until 30 November by visiting www.emeraldinsight.com/tk/prediction

For more information about the journal, visit http://www.emeraldinsight.com/intr.htm