Researchers help BBC track hottest trends

A new BBC Radio 5 Live show is using cutting-edge Big Data research to find the week's hottest social media trends.

Sunday evening's '5 Live Hit List' will feature a weekly list of 'hot topics' generated by the Collaborative Online Social Media Observatory (COSMOS) a consortium of Cardiff, Warwick and St. Andrews social and computer scientists

The COSMOS platform was launched a month ago - cosmosproject.net . It allows social media data to be collected and analysed and is free of charge to UK academics, NGOs and public sector organisations.

Dr Pete Burnap, School of Computer Sciences & Informatics, technical lead on COSMOS, said "We expected there to be demand from academics interested in using this data in their research, and around 100 universities internationally are now using COSMOS, but we did not expect to see the amount of download requests from the public sector and NGOs".

The Office for National Statistics, the Food Standards Agency, the Department for Work and Pensions, the Foreign and Commonwealth Office and Citizens Foundation Iceland all requested access to the platform within the first month of release.

Dr Matthew Williams, School of Social Sciences and principal investigator on the grant that founded COSMOS said: "public sector departments and charities are now recognising the value in using social media data to understand how 'digital publics' are organising in new ways online. Increasingly 'online citizens' are engaging with government services via Twitter and there is a pressing need to engage with new social media technologies to capture and respond to trends in opinion in order to improve service delivery."

The COSMOS team are currently working with the Office for National Statistics and the Food Standards Agency to incorporate big data analytics into working practices in order to meet departmental objectives.

Dr Luke Sloan, School of Social Sciences, the project's data integration and demographics lead said: "We are working closely with government in order to increase the usability of social media data.  In particular we are developing algorithms that estimate the gender, location, age and occupation of twitter users, allowing departments to link this new form of data with existing administrative datasets."