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.
. . but rather a more sophisticated matrix that has some ability to understand context.
Yeah, right.
I am Slashdot. Are you Slashdot as well?
So, sarcasm research in AI doesn't have a sarcasm problem, it has a sarcasm filter problem?
Wake me up when we solve the problem of deterministically detecting sarcasm with human intelligence.
They're essentially trying to develop a machine that can understand Poe's Law. Considering how much trouble people have with that, they've set a pretty high bar for themselves.
I'm sure automatic sarcasm detection's totally going to help those poor marketers sift through their pittance of data in even more ways.
I'm pretty sure they will sort it out very soon
Heck, many humans don't get sarcasm... it depends on what environment you grew up with, I think.
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.
Some drink at the fountain of knowledge. Others just gargle.
All jokes aside, this is a real issue because it is incredibly complex. First, you have to have enough skepticism to disbelief a statement. Then you have to enough confidence to acknowledge that you don't believe that statement. Third, you have to know enough about the speaker to realize that the statement they gave is not consistent with their own internal beliefs. This is made more complex because good sarcasm requires that someone somewhere in the world actually believe the. statement.
Those are all very high level thought processes. You are not just judging a statement as true/false, but judging what others believe. Any child psychologist can tell you how hard that is.
excitingthingstodo.blogspot.com
When AI finally does understand saracsm, Bender won't seem anywhere near as funny.
Some drink at the fountain of knowledge. Others just gargle.
Well why don't you just go take over the world, see if I care!
This Space Intentionally Left Blank
If a conversational bot can get to the level of Aspergers (I know, I know, who the hell cares if it's in DSM-5) then I'll be awfully happy.
Though...if you're using it less for conversation, and more for fact scraping, I guess that makes things way more important. Trying to think of how people recognize sarcasm, and it seems to be "a person is stating--without expressing uncertainty or hedging--something that I have good reason to think they don't believe". It seems like every culture also has their own shorthand for sarcastically agreeing with someone, but the shorthand isn't always recognizable cross-culturally (Americans visiting England have a big problem here).
The big AI problem here is, you have to be able to model the writer's thoughts and guess what their opinions are. Ask your stereotypical left-winger and right-winger about gun control and they might both say "Oh, yeah, that's exactly what we need", but you need to be able to predict their actual opinions in order to know if there's dissonance or not. Makes it hard to detect sarcasm in people you don't know anything about, unless they're using shorthand ("yeah right", "great idea--let's discuss it later")
Well, this research wasn't a waste of time and money.
Judging from the anti-science anti-government of the GOP and the nihilistic religious wars spreading across multiple continents, i'd say that high-science has reached its high water mark and the tide is flowing out. Trump's popularity is outside of scientific explanation. And Ted Cruz's state of the union address will be the epitaph of government funded science in the USA.
Just tell people the AI is American, and it will still pass the Turing Test. At least it will if you only test British people.
Well, duh!
https://www.youtube.com/watch?...
Nope, no sig
https://www.youtube.com/watch?...
It was sarcasm before, now it is freedom of expression or "trolling". All the neckbeards here (wink) should know it already.
This has already been solved humans just need to include :-) or say "I'm only kidding" every time they say something sarcastic such as:
Vote donald trump. :-)
or
News flash: Republicans against torture. I'm only kidding.
Minimum threshold fixed. Thanks!
Was this modded +5 because the Berniacs missed the sarcasm or did everyone get it and find the news link I cited informative?
Yes.
Nope, no sig
Well that's fucking useful isn't it
A sarcasm detector? Oh that's a really useful invention...
There's probably a joke in there about how hard it is for Germans to detect sarcasm too, but I would never make such a comment.
-Styopa
Doesn't this effectively demonstrate that much so-called "artificial intelligence" has nothing whatsoever to do with intelligence?
To properly understand sarcasm requires a very high level of understanding of the language, the situation being commented on, and human nature. The current levels of AI would only be able to understand context-free phrases like "Yeah, right" as sarcasm and maybe guess at others. Of course, sometimes sarcasm is also indistinguishable from sincerity, either on purpose or due to the speaker's incompetence, and even more so on the internet (Poe's Law).
Don't waste your vote! Vote for whoever you want, unless you live in a swing state it won't matter anyways
Modern AI systems actually "get" very, very little - witness Siri, Google Now et al., whose understanding remains extremely limited and narrow.
Surely when AI is at that stage it will be able to analyse vocal patterns to detect sarcasm. The main reason people don't detect it online it's because there's no tone of voice when written.
Wanna buy a shirt?
https://www.redbubble.com/people/stealthfinger/shop?asc=u
There's a ton of aspie freaks online that can't seem to figure out sarcasm or humor in general, either.
If you were me, you'd be good lookin'. - six string samurai
That's what humans do. 8-)
...there should be a font for sarcasm.
all you have to do is look for the /s tag. /s
Just make a reCAPTCHA system based on sarcasm and the bots will be better than an average human in a couple weeks.
Like me. Yes, I often don't get sarcasm. Unless it's really, really, REALLY obvious.
We used to have a Bill of Rights. Now, with the rights gone, all we have left is the bill.
Poe's Law Much?
Our new computer overlords think they're so smart...
Sarcasm and satire have only superficially similar.
The thing with satire is to create incongruity between the package and its contents.
Some dishevelled fellow shows up, claims he's a world-class chef from Syria, and offers to cook you a five-star dinner in exchange for his own meal. He doesn't look especially Syrian. And you can't make out his accent.
So you say, "well, I'm not sure whether to believe you. I made this pate yesterday, and I'd like your opinion. Just one second." Then you duck around the corner, dip a knife into the kitty litter box, smoosh the fresh excavation onto a nice Wheat Thin that Jr. left uneaten on his plate after dropping it on the garage floor (use the dishwasher, people!), which you then—in a flash of inspiration—decorate with two fish eyes from the fish carcass in the fridge that is now mired in cold gelatin and really should have been turned into fish stock three days ago. Oh, what the heck—let's do this right!—so you add a tablespoon slab of the aging fish gelatin.
At this point, it looks like fancy French cooking (looks can be deceiving). It really looks like French cooking when you extend it with the utmost graciousness on your whitest French serving cloth.
"Syrian chef" picks it up, opens his mouth, slides it trustingly under his nostrils, and is about to bite down, then freezes in an eyebrow-raising display of alarm and disgust.
"Don't be angry! I had to make sure. Do you still want to cook dinner? Oh well, better luck next time. "
"Sheesh. I think he called me a racist bastard in some foreign gibberish. Did he really mean it? Surely he could see my predicament and my efficient solution. Hmmm. I suppose it did look a lot like I was serving him shit on a cracker, Gallic style, from his point of view."
Sarcasm is exactly the same thing, except you can't be bothered with the cracker, the fish eyes, the extra slab of fish gelatin, or the white napkin. You just hold up a dry cat turd with your bare fingers and call it a cheesy, because you really wouldn't want to have to eat a fancy dinner prepared by the colour blind, not even if he really was a great chef in his own land.
That's a really good point.
Nope, no sig
If an automatic system can detect sarcasm, then it is arguable (put aside Searle for just a moment) that it may understand sarcasm.
If so, how do we know it's not just taking the piss when it provides feedback?
Next, they'll be telling us seven years of research on PhD students has proved that five times out of six telling the ones with Asperger's to honk on Bobo will send them out looking for a clown to blow.
I've calculated my velocity with such exquisite precision that I have no idea where I am.
Honestly, you guys kill each other so often, how the hell was I supposed to know you were serious about this one. Geez, sarcasm mode off, sorry.
TL;DR but maybe AI is doing a good job and the researchers just do not understand the difference between sarcasm and irony
Sarcasm in spoken language often relies on subtle inflections in speech (deadpan) and facial expression/movement, and even with cues is still missed a lot of the time. Online you can either rely on smilies or accept that a good 75% of people just won't get it. Or, conversely, if they do it might just be that they have a bias to see sarcasm where it isn't - a trap I've fallen into often enough when I see spouting nonsense and wrongly conclude they're using sarcasm when, in fact, they turn out to be 100% serious.
That said... give me some grant money and I'll work on this difficult and very important problem.
See what I mean.
Pharque,
a computer with a sense of humour
context is the wrong direction,
have to start at a much more human level,
isn't sarcasm the highest form of wit,
start the process off with knock, knock, who's there.
Go well
Not surprised to see this come out of Australia, we're a people that regularly use the phrase "yeah, nah".
Calling someone a "hater" only means you can not rationally rebut their argument.
The authors of the paper are mystified as to why everyone is taking this paper so seriously. They were just being sarcastic.
I really need this. I have a very difficult time knowing what is sarcasm and what isn't online. Hell, I even have a problem with this face to face.
Who would have guessed that you need context to understand sarcasm?
How embarrassing. I have to explain my sarcasm all the time. Obviously, I and the bots are all who are left on the internet.
Such as "artificial intelligence."
They just need to scan for the winking emoji ;-)
Difficult as it is for machines to divine humour, especially sarcasm, from ordinary discourse, it is even more difficult for those firmly entrenched in the regressive left to see anything other than sexism, misogyny, homophobia (sic), Islamophobia (sic), trans*-phobia (sic), ableism (sic), microagreesions (sic), patriarchal oppression, and a plethora of other supposed and manufactured ills even in something as innocuous as wishing someone "good morning." Radical, third wave, intersectional (sic) feminists, social justice (sic) warriors (screaming juvenile whingers/whiners or SJWs), and their fellow travelers are all notoriously dour and humourless. To those who are hopelessly immersed in this crazy world of regressive propaganda and group think (a very funny term that really means unthinking masses) humour that satires them and their vicious, illogical and insane ideology and behaviour is tantamount to a hate crime.
Compared to that, AI humour recognition is a piece of cake.
When most humans cannot detect sarcasm, except in small in-groups where it is stylized and standardized, then it is useless to expect computers to do it.
Unless of course, you have a mystical faith in the perfection of computers. Which some people do...
P.S., Never use sarcasm in internet communications, no-one will understand you. And they will just think you are a fool.