Skip to main content

Software tool tackles political spin

Oh, what a beautiful world it would be if traffic wardens had compassion, footballers could speak without clichés and politicians gave a straight, honest answers to simple questions.

Mr David Blair and Mr Tony Cameron (well, can you tell the difference?), your days of half-truths could be numbered thanks to a “spin-o-meter” developed at Queens University in Canada.

The software is based on a model developed by the University of Texas for finding deception in text. Seemingly, it is possible to see if a person is telling the truth based on the number of first person pronouns, exceptions words such as `however' and `unless' , and negative emotion and action words, which are used.

The software model was first applied to look for deception in the emails sent to and from Enron employees in the three and a half years before the collapse of the company.

According to the researchers:

We discovered that, while the model detects deception, it also detects other kinds of unusual text. The best way we know to categorise what the model detects is spin -- text whose apparent meaning isn't the true beliefs of the person saying or writing it.

With this discovery, the boffins analysed the speeches of Canada’s three main English-speaking political leaders and came up with an average spin rating per party.

For those of you with an interest in Canadian politics, the Liberals were deemed worst with 124, perhaps because being the party in power they have a track record to protect, then came the New Democratic Party (NDP) with 88 and the Conservatives with 73.

Whether the software model could make sense of a George Bush speech is, however, not yet clear.

You can read more about their work here.