During the Olympics, the London Eye was transformed into a stunning light show. The spectacle was driven by the sentiment of tweets related to the London 2012 Games for the Energy of the Nation project, commissioned by EDF Energy. Every tweet was analysed for its sentiment, whether it was positive or negative, and filtered through a computer program which converted them into a light show using the London Eye. The colours reflected the mood of people tweeting – purple for negative and yellow for positive.
The project – the world’s first social media driven light show - was made possible by the expertise of Professor Mike Thelwall from the University of Wolverhampton. Mike is one of the leading experts worldwide in infometrics - the science of measuring information to assess the importance of a set of documents - and Mike specialises in the newest area, measuring the value of web pages. For the light show, Mike created a dictionary of 2,750 terms using a system he developed called SentiStrength, which can determine the positivity or negativity of tweets.
He explained: “The system basically scored every tweet according to a host of keywords and assigned it an overall sentiment. It happened in a millionth of a second.
“I was quite worried in the run up to the Games as there was a lot of negativity beforehand, particularly about transport. I thought we might get bad results if people couldn’t get to work for example! But that didn’t happen – there was a lot of positivity and enthusiasm.
“I went to see the London Eye and it was fantastic. It was so huge and impressive against the dark sky. It was wonderful to be part of the Games in such a way.”
As well as the half an hour light show every evening during the Games, a barge with a big screen was set up on the Thames showing up sentiments expressed on Twitter in real-time, with a control room inside explaining how it was all done.
This year saw the SentiStrength program used again in the lead-up to the Superbowl final – and this time it was the Empire State Building in New York that became a fantastic light show. Fans were encouraged to tweet in support of their team and negative comments about their opponents and each night the building was lit up in the colours of the team with the strongest sentiments.
Professor Thelwall’s work has been rated as world leading in the Research Excellence Framework (REF 2014), which evaluates the quality and impact of research submitted by UK universities across all disciplines.
The SentiStrength program is used by Yahoo Answers, to filter for good intentions based on the feedback given to responders to previous questions. Those with positive feedback will appear higher on future answers, as their responses are considered helpful. It is also used in market research to enable companies to detect what their customers are saying about their products. This means they can monitor reaction in real-time, so if a problem develops with a product they should know about it quickly.