Slashdot Mirror
Why Sarcasm Is Such a Problem In Artificial Intelligence (thestack.com)
An anonymous reader writes: A new paper from researchers in India and Australia, "Automatic Sarcasm Detection: A Survey," highlights one of the strangest and ironically most humorous facets of the problems in machine learning and humour. The paper outlines ten years of research efforts from groups interested in detecting sarcasm in online sources. It details the ways that academia has approached the sarcasm problem, including flagging authors and ring-fencing sarcastic data. However, the report concludes that the solution to the problem is not necessarily one of pattern recognition [PDF], but rather a more sophisticated matrix that has some ability to understand context.
Wake me up when we solve the problem of deterministically detecting sarcasm with human intelligence.
The reverse is also true. Because it's so hard to discern sarcasm in online text, it can also be difficult to express sincerity. One can write something truly sincere, only to have others interpret it as sarcastic, flippant, or derisive. Adding emphasis, such as "I'm being absolutely serious here" only worsen the problem, since over-emphasis is a hallmark of sarcasm.
There is also sometimes intended ambiguity; people write things that could be interpreted as either sincere or sarcastic/joking, and wait until after the fact to claim the intended meaning. For instance, one can send a text/email that is flirty, and decide (based on the recipient's response) whether to claim it was sincere or a joke.
In short, this is not just a hard problem for AI, it is a hard problem for intelligence more generally. Sarcasm is--quite intentionally--sitting right on the edge between credibility and exaggeration.
The real problem is they are looking at written data. Sarcasm is based on auditory and visual cues of the person. Detecting sarcasm online is like looking for a needle in a haystack when you don't know what a needle or hay is.
Brilliance without wisdom, power without conscience. Ours is a world of nuclear giants and ethical infants.
To some degree yes, but there are still satirical works of literature throughout history such as Swift's A Modest Proposal that would pose a similar problem. The problem in understanding sarcasm or satire without the visual or vocal cues relates to understanding meaning (a difficult problem in its own right) as well as why a particular response is absurd given the context, which means you also have to know what the expected or typical answer should look like.
For example, if people were discussing world hunger online and someone proposed a solution whereby half of the starving ate the other half, which would not only reduce their hunger, but would also effectively cut the problem size in half, most people would pick up on the sarcasm because the proposed solution is utterly absurd. Cannibalism isn't a typical answer to food shortages.
Other techniques involve detecting flawed logic or intentionally faulty reasoning which is often used in sarcasm or satire. For example, if someone posits that eliminating gasoline taxes would lead to flying cars due to a deterioration and lack of upkeep on roads necessitating alternatives for traversing them, most people would again recognize that an illogical leap has been made.
An interesting way of going about this might be to study how people with autism spectrum disorders process information as many of those individuals often lack the ability to detect or fully process sarcasm or satire, even in the presence of visual or auditory cues. Understanding why that happens or by what mechanism other individuals are able to process that information correctly could allow us to understand how to program computers to detect sarcasm. I