U.S. Intelligence Agency to 'Foresee the News'

Prophecy News Watch:  Some hedge funds already use Twitter and other social networks to 'predict' fluctuations in the stock market.

Now a U.S. government agency is to use the social network to attempt to predict revolutions.

Iarpa - the Intelligence Advanced Research Projects Activity - hopes to be able to predict 'unexpected' events such as revolutions using a wide range of data.

Iarpa is to spend three years analysing data taken from 'open' sources such as Wikipedia edits, Twitter posts and even publicly available data from traffic systems and webcams.

The data will be taken from 21 countries in mainland Latin America, chosen because they offer a wide variety of data, and a wide variety of reported events.

The goal is to allow Iarpa to 'predict' news before it happens.

Teams will compete to make the best predictions.

A statement from the proejct, called OSI - the Open Source Initiative - said, 'Teams choose sensors, data, and methods. Teams are rewarded for early and accurate warnings of as many newsworthy events as possible.'

The focus is on advanced computer modelling, rather than simply building up a reservoir of data to 'guess' events.

OSI has already investigated the idea of using such computer models to predict consumer behaviour and even disease outbreaks.
Using such a wide range of sources is a new approach, however.

'Few methods can anticipate "unexpected events". The program aims to fill this gap''Many significant societal events are preceded and/or followed by population-level changes in communication, consumption, and movement,' said IARPA's initial solicitation for teams to work with Open Source Initiative.

'Some of these changes may be indirectly observable from publicly available data, such as web search queries, blogs, micro-blogs, internet traffic, financial markets, traffic webcams, Wikipedia edits, and many others.

'Published research has found that some of these data sources are individually useful in the early detection of events such as disease outbreaks.

'But few methods have been developed for anticipating or detecting unexpected events. The IARPA’s Open Source Indicators (OSI) Program aims to fill this gap.'

Teams are currently competing for funding from OSI - including computer scientists from the Centre for Collective Intelligence at MIT.

Computer scientist Peter Gloor of the Centre for Collective Intelligence claims to have a working predictive model box office returns, using sources including social media.

Gloor told Nature that the model is already 90 per cent accurate.
IARPA hopes teams will be able to model real-world events within nine months of the programme launching.

'At kickoff, Government team will provide a large list of significant events in Latin America for the prior year, for which an early warning would have been valuable,' says the OSI website.
'After kickoff, Government team will provide monthly 'ground truth' – events for the last month, for which a warning would have been expected.'