EMOTIVE can analyse up to 3,000 tweets a second, using complex software to extract from each tweet a direct expression of one of eight basic emotions: anger, disgust, fear, happiness, sadness, surprise, shame and confusion.

In the past it has been used to monitor people’s reactions to big events, such as the Paris terrorist attacks, and in 2015 the system correctly predicted the outcome of the UK General Election.

For the US Presidential campaign the research team – from the School of Business and Economics – are also hoping to use EMOTIVE to predict who will be taking up office in the White House.

Professor Tom Jackson, who heads up the EMOTIVE team, said: “Twitter is a very concise platform through which users express publicly how they feel about a particular event, be that a criminal act, an election or even a change in the weather.

“We have already shown what a fantastic tool EMOTIVE is in capturing through Twitter the public’s mood. The system gives us, in real time, a snapshot of how people are really feeling and from this, when looking at an election, we can make a prediction of how these feelings will be reflected in voting.

“It also enables to see how people’s emotions towards each candidate change, minute by minute. From our data so far it is a very close fought campaign, which the traditional polls have not shown!”

Dr Martin Sykora, who is also part of the EMOTIVE added: “The system we created takes the eight emotions and gives them a rich linguistic context so that we can chart the strength of emotions expressed in ordinary language and also in slang.  We can view how reactions grow and diminish over time towards Trump and Clinton.”

There is a dedicated EMOTIVE website for the US elections, where you can track in real time people’s feelings to the presidential race and which candidate is currently in the lead.

The system is tracking and analysing tweets with the following hashtags: #Trump, #Trump2016, #Hillary, Hillary2016, #MakeAmericaGreatAgain, #ImWithHer. In the last 24 hours it has looked at more than 40,000 tweets.