RESEARCHERS AT THE VENERABLE MIT have come up with a system that can read the emojis in a message and gauge whether the author is being sarcastic.
We aren't impressed. Sarcasm is pretty easy to spot and anyone that uses a emoji as a means of being slyly clever in a post or email probably isn't writing anything that we need to read. The INQUIRER is cynical though, so we were never going to be particularly moved by this news.
MIT has used artificial intelligence (AI) to filter the messages and identify what it assumes is sarcasm. It does that by looking at the emojis and the overall content and context of the message. It is called DeepMoji, and MIT calls it 'emotional artificial intelligence'.
"We use millions of texts on Twitter containing emojis for training a deep learning model that understands many nuances of how language is used to express emotions. For instance, it does well at capturing sarcasm and slang. We beat state-of-the-art algorithms across many benchmarks datasets," says the official skinny.
"Disclaimer: Note that the model has learned about language from the raw, uncurated expressions of individuals on social media. We do not endorse in any way the emotional interpretation that the model has of any particular content."
MIT says that research has done this sort of thing already, just in a particularly crappy way. Its system is much better, apparently, and knows that bad can mean good.
"For instance", says researcher Bjarke Felbo, "the model can capture slang such as ‘this is the shit' being a positive statement as well as very varied usage of the word ‘love'...
"We teach our model an understanding of emotions by finding millions of tweets containing one of the top 64 emojis and ask the model to predict them in context. Just by examining the predictions of our model on the test set it is clear that the model does have an understanding of how the emojis are related.
"The model learns to group emojis into overall categories associated with e.g. negativity, positivity or love. Similarly, the model learns to differentiate within these categories, mapping sad emojis in one subcategory of negativity, annoyed in another subcategory and angry in a third one."
[Shrug emoji]. µ
More wise words from Mountain View
Firm failed to say that launch prices were only an 'introductory offer'
We think we all have an airbag problem
Give us a break