French company Spotter is going to help the British Home Office identify sarcastic comments posted online.
Spotter has developed an algorithm driven analytical tool that can scan the social web and dig up snark. It uses a combination of linguistics, semantics and heuristics to create these algorithms and generate reports about online reputation. The company claims an 80% accuracy rate.
A company spokesman told the BBC, the source of the original report:
“One of our clients is Air France. If someone has a delayed flight, they will tweet, ‘Thanks Air France for getting us into London two hours late’ – obviously they are not actually thanking them,” he said.
“We also have to be very specific to specific industries. The word ‘virus’ is usually negative. But if you’re talking about virus in the context of the medical industry, it might not be.”
The service is not cheap, and comes at the cost of a minimum of £1,000 ($1,500) per month for its services.
However, Simon Collister, a lecturer at the London College of Communication who is an expert in PR and social media, to the BB that, well, basically, “these tools are often next to useless.”
Mr. Collister doesn’t think that machines are much good at handling and comprehending tone, context, and the nuances of language. People are still good at picking those things up. Except Sheldon Leonard. He has no clue about sarcasm.
Mr Collister had the final words in the BBC article, and quite apt they are:
“The challenge that governments and businesses have is whether to rely on automated tools that are not that effective or to engage a huge amount of human effort.”
So, we can probably rest assured that government spying on its citizens is going to miss all the nuance, snark, and bitchiness of the haters online and we should be prepared for a million trolls in our prisons by the year 2020.