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.

Comments

Solution?

[OzPeter]

. . but rather a more sophisticated matrix that has some ability to understand context.

Yeah, right.

Re:Solution?

[Shortguy881]

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.

Re:Solution?

[alvinrod]

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

Re:Solution?

[U2xhc2hkb3QgU3Vja3M]

A.I.?

Re:Solution?

[squiggleslash]

Cultural and social cues too. British people, for example, frequently accuse people from a certain large Northern European country of having no sense of humor. Why? Well, because when they/we (I'm an ex-Brit) make sarcastic comments in front of them, said Northern Europeans take it seriously.

Now I have to assume sarcasm is fairly universal. I'd be surprised if aliens from the Planet Thargh IV are not familiar with the basic concept of "saying the opposite of what you mean because it's absurd, and finding humor in its absurdity". So the chances of said country not actually actually being familiar with the concept is pretty unbelievable.

More likely is that the transmission - the social cues, the way English speaking people exaggerate the first few words of a sarcastic sentence ("Oh a sarcasm detected. Well that's a useful invention!") to indicate that we're being sarcastic and not serious - is different.

There's another location where sarcasm just never seems to work (and, alas, I'm dumb enough not to realize it half the time): The Internet. Or rather, written text, where sarcasm is interpreted as stupidity more often than not. We've even developed cues to try to ensure it's not misinterpretted, from "/s" to fake HTML tags. Again, this suggests everything is about the cues.

Computers probably can detect sarcasm if taught the cues. It ought to be easy: look for cues, determine meaning of sentence, if cues present and interpretation in local context is absurd, call laugh().

Or raiseEyebrow(). Whatever seems appropriate for the lowest form of wit...

Re:Solution?

[Ol Olsoc]

Yeah right, sarcasm is lower than puns. Absolutely.

A good pun is it's own reword.

Problem in AI?

[CurryCamel]

Wake me up when we solve the problem of deterministically detecting sarcasm with human intelligence.

Wait, Are you saying we shouldn't eat children?

[goombah99]

I'm a devoted follower of Dr. Johnathan Swift and I was persuaded by his logical essay "For preventing the children of poor people in Ireland,
from being a burden on their parents or country, and for making them beneficial to the publick.". I was in fact arguing with my google car the other day about running over some filthy budernsome tyke and the google car AI was pretty darn adamant but eventually came around to seeing it my way after I instructued to parse some Swift. Sadly by then the succulent 28 pounder had wander off and no meal was to be had.

The last "command" given to the first AI

[Daetrin]

Well why don't you just go take over the world, see if I care!

Re:Poe's Law

[Anonymous Coward]

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.

Re:Someone mod the parent up

[Enigma2175]

Dammit, everything posted in this thread has the sarcastic voice in my head. How am I supposed to know what's sincere?

Re:Bite my shiny metal ass

[rtb61]

They would actually say AI is pretty invasive because sarcasm is not just contextual but also social and cultural. Your require a shared social and intellectual basis in order to properly express and understand sarcasm. The AI would need to know the person in order to effectively interpret their sarcasm. AI is required for quality translation services as contextual understanding is required. This requires multiple independent interpretation of the data provided (letters, words, sentences, paragraphs, grammar, nouns, verbs, adjective, adverbs, punctuation et al, their cross correlation relationships and then perversely enough the typical humans inability to use them properly and correct for that). and the results then correlated to the broader conversation and the particular individual. So a whole series of algorithms running continuously (the longer they run the better, with data flowing through them of course) into which data is fed, in fact it is the overall pattern of the algorithms being used more than the individual algorithms themselves, except the core controlling algorithms which are more complex, than the actual data processing algorithms of which there are many.