ICHEC and NewsWhip study predicts how viral a story will be in its first hour

3 Jul 2014

A study by a research team at the Irish Centre for High End Computing (ICHEC) and new media start-up NewsWhip has found that 70pc of viral stories can be accurately identified within their first hour of publication.

The “social velocity” of news stories were measured by a metric developed by NewsWhip.

The research was conducted to test the effectiveness of the early signals used in NewsWhip’s Spike platform to identify big emerging stories.

NewsWhip’s Spike platform is used by news companies big and small, including Bloomberg, BBC and BuzzFeed. 

ICHEC collected scores for over 123,022 stories, analysing the data performance to track the impact of each story over time, using Spike’s 24 hour scoring as the benchmark. 

The ICHEC researchers divided stories into big successes, middling successes, and failures, based on how far stories spread over 24 hours, using NewsWhip’s tracking data.

The findings show that stories that attract significant levels of sharing within the first hour of publication are likely to end up achieving a high level of “virality” over 24 hours, and the level of early sharing is predictive of how far they will go.

Social velocity

“They calculated threshold ‘social velocity’ scores for each category of content,” data architect Dr Bruno Voisin explained.

“This allowed the ICHEC to determine how many of the stories that have been picked out as ‘viral stories’ in Spike 1 Hour did make it to 24 hour.”

The ICHE’s study was supported by Enterprise Ireland.

“Our active SME programme enables us to collaborate with companies from various sectors on innovative projects,” said Sufian Al Aswad, business development and industry liaison at ICHEC.

“The programme provide SMEs access to cutting edge technology and high level of expertise, in addition to the funding and business mentoring provided by Enterprise Ireland, all of which sets the desired environment for Irish companies to achieve their objectives and targets for growth.”

Social data image via Shutterstock

John Kennedy is a journalist who served as editor of Silicon Republic for 17 years

editorial@siliconrepublic.com