Page 1 of 1
A sarcasm detector, that's a real useful invention.
PostPosted: Wed May 19, 2010 6:49 pm
by Cognitive Gear
Link to full articleDevised by computer scientists at The Hebrew University in Jerusalem, the algorithm has been programmed to recognise sarcasm in lengthy texts by analysing patterns of phrases and punctuation often used to indicate irony.
In tests on 66,000 product reviews posted on the Amazon shopping website, the algorithm had an impressive 77 per cent success rate in picking out sarcastic comments – arguably higher than some humans.
The researchers "trained" the algorithm to recognise sarcasm by teaching it nearly 5,500 sentences from Amazon reviews that human volunteers had marked as either sarcastic or non-sarcastic.
The sarcastic phrases from the pool of Amazon reviews used for the research included "Great for insomniacs", "Are these iPods designed to die after two years?" and "Defective by design".
From its learned list of sarcastic phrases, the algorithm was taught to recognise patterns of words commonly used by writers to show that they do not mean to be taken literally.
PostPosted: Wed May 19, 2010 7:25 pm
by steenajack
Okay, this is awesome! This really could come in handy.
PostPosted: Wed May 19, 2010 8:02 pm
by Midknight74012
People like me and many others I can name don't need this.
PostPosted: Wed May 19, 2010 8:08 pm
by Davidizer13
Oh, sure, like that'll ever work.
PostPosted: Wed May 19, 2010 8:14 pm
by Alcuinus
HOLD IT! I think I detect sarcasm in the responses thread!
PostPosted: Wed May 19, 2010 8:14 pm
by MightiMidget
Uh. Huh. Sure. Okay. That wouldn't be embarrassing to carry around or anything.
PostPosted: Wed May 19, 2010 8:25 pm
by Tsukuyomi
Haha!
That's pretty creative XD?
PostPosted: Wed May 19, 2010 8:36 pm
by Lynna
Well....shouldn't people be relying on thier own instincts to figure these things out rather than relying on computers? I mean, if all we ever do is go crying to technology to save us, how will we ever learn? I think new technology is great, and this probably WILL come in handy in some way or form, but for an ordinairy person in everyday life?
PostPosted: Wed May 19, 2010 9:16 pm
by TGJesusfreak
I agree Lynna. I don't see the use of a sarcasm detector XD It's like " *gasp* I think he's being sarcastic. Let me go get my sarcasm detector!" XD
PostPosted: Wed May 19, 2010 9:19 pm
by steenajack
I'm sorry, but all the sarcasim in the thread made me think of this clip for some reason. (at about 1:17 and on, but the whole thing is funny)
http://www.youtube.com/watch?v=kBJQIaWk8uk
PostPosted: Thu May 20, 2010 5:16 am
by Warrior4Christ
PostPosted: Thu May 20, 2010 6:44 am
by Mr. Rogers
Is this an invention for people who were born without wit?
PostPosted: Thu May 20, 2010 7:27 am
by Rusty Claymore
No, it will be released in the form of a sonic screwdriver, so when you feel someone is being sarcastic you can point it at them and go, "AHA! You thought you could get that by me?!? OBJECTION!! The screwdriver NEVER LIES!! Unlike you, it seems. Pitiful fool, you're no match for my tool! Bua-hahahahaaha!" ...or something like that.
Sarcasm- n. The gap between what is said, and what is meant.
PostPosted: Thu May 20, 2010 7:31 am
by blkmage
Serious business time.
Stuff like this is never intended for end-users to use. This is AI research; it's natural language processing in particular. Sarcasm was probably chosen for this paper because it's a tricky and interesting problem from a computational standpoint. I mean, some
people have a hard enough time recognizing.
Rather, these algorithms are likely to be used for stuff like data mining. Even if the sarcasm recognition is proven to have no usefulness at all, the techniques described in the paper can be extended to some other application of natural language processing. The ultimate goal of natural language processing is for a computer to be able to understand human language without us having to bend over backwards to phrase it in a way that the computer can understand.
Anyhow, if you're interested in current AI research, you can have fun and read the paper:
"A Great Catchy Name: Semi-Supervised Recognition of Sarcastic Sentences in Online Product Reviews".
PostPosted: Thu May 20, 2010 7:56 am
by Nate
Awesome, I'll just drop everything I'm doing to read that article because it interests me so much.
PostPosted: Thu May 20, 2010 1:36 pm
by uc pseudonym
In tests on 66,000 product reviews posted on the Amazon shopping website, the algorithm had an impressive 77 per cent success rate in picking out sarcastic comments – arguably higher than some humans.
Forget "arguably." I want someone to do an actual test to verify this statement.
PostPosted: Thu May 20, 2010 4:11 pm
by Lynna
PostPosted: Fri May 21, 2010 4:53 am
by Warrior4Christ
Rusty Claymore (post: 1395485) wrote:No, it will be released in the form of a sonic screwdriver, so when you feel someone is being sarcastic you can point it at them and go, "AHA! You thought you could get that by me?!? OBJECTION!! The screwdriver NEVER LIES!! Unlike you, it seems. Pitiful fool, you're no match for my tool! Bua-hahahahaaha!" ...or something like that.
Sarcasm- n. The gap between what is said, and what is meant.
No, that's sarchasm.
AI strategies sound so general when the mechanism is explained.. so it's probably not too much more difficult to find different applications for it with little modification.
PostPosted: Sat May 22, 2010 2:08 am
by Nikolai Melodie
...I wish we had one of those, but, for the internet. For text. 'Cause, really, I can think of at least five people in my immediate life (not here on CAA) who could reallllly use a blinking pop up window to tell them, 'CALM DOWN FOOL; I WAS BEING SARCASTIC!' every once in awhile.