[QUOTE]The pursuit of machine intelligence means we have to come up with ways to communicate with our computers in a way both entities can understand. But while computers process verbal commands in a straightforward fashion, humans tend to use more sophisticated speech forms, employing slang or symbols to convey an idea. So an Israeli research team has developed a machine algorithm that can recognize sarcasm.
SASI, a Semi-supervised Algorithm for Sarcasm Identification, can recognize sarcastic sentences in product reviews online with pretty astounding 77 percent precision. To create such an algorithm, the team scanned 66,000 Amazon.com product reviews, with three different human annotators tagging sentences for sarcasm. The team then identified certain sarcastic patterns that emerged in the reviews and created a classification algorithm that puts each statement into a sarcastic class.
The algorithms were then trained on that seed set of 80 sentences from the collection of reviews. These annotated sentences helped the algorithm learn what sorts of words and patterns distinguish sarcastic remarks – those that mean the opposite of what they literally convey, or that convey a sentiment inconsistent with the literal reading.
They then turned the algorithm loose on an evaluation set. Pattern evaluation efficiency scored accurately 81 percent of the time, while the overall precision of the pattern recognition/sarcasm categorizing algorithm was accurate in 77 percent of instances. Not bad for a computer’s first shot at interpreting the human sense of humor.
This isn’t all just so your Roomba gets the joke when you tell it it sucks. Computer programs that can recognize sarcastic statements could generate better personalized content and make better recommendations to human users by not mistaking a product review titled “keep your receipt” with a sound piece of online shopping advice. It could also benefit opinion-mining systems that troll the Web trying to measure public sentiment about a product or idea.
[img]http://www.popsci.com/files/imagecache/article_image_large/articles/sarcasm01.jpg[/img][/QUOTE]
Source: [url]http://www.popsci.com/technology/article/2010-05/computer-algorithm-can-recognize-sarcasm-which-soooo-cool[/url]
One step closer to fembots that can feel love :smith:
:q:
I'm sure.
what a useful discovery
This'll probably work
This is pretty incredible.
[QUOTE=Eudoxia;22087526]
:q:[/QUOTE]
I love that emote.
Not being sarcastic.
My computer detected sarcasm
ITT: people try to make funny sarcastic jokes and fail miserably.
[QUOTE=sloppy_joes;22087713]ITT: people try to make funny sarcastic jokes and fail miserably.[/QUOTE]
what a bummer
i hope youre being sarcastic
How can it tell?
Text is just text
Theres hardly ever emotion behind it
[QUOTE=InsanePyro;22087813]How can it tell?
Text is just text
Theres hardly ever emotion behind it[/QUOTE]
I don't know man, sometimes when reading posts on a forum the, er, mental voice thingee reads them with different tones and such, depending on the wording of a sentence and all that.
[QUOTE=InsanePyro;22087813]How can it tell?
Text is just text
Theres hardly ever emotion behind it[/QUOTE]
How can we tell?
Text is just text.
There's hardly ever emotion behind it.
[QUOTE=Lego399;22087574]This is pretty incredible.[/QUOTE]
[QUOTE=PrismatexV8;22087540]I'm sure.[/QUOTE]
[QUOTE=Triumph Forks;22087565]what a useful discovery[/QUOTE]
[QUOTE=deggemannen;22087567]This'll probably work[/QUOTE]
Warning - buffer overflow
cool bro
[media]http://www.youtube.com/watch?v=VRFZj3FWRrw[/media]
Felt it fitting
It won't work on me, my dead-pan is rock-solid.
[QUOTE=helpiminabox;22087877]How can we tell?
Text is just text.
There's hardly ever emotion behind it.[/QUOTE]
Humans have emotion. Computers do not. I think it will require much more coding and such then a simple algorithm can provide. And even then will it be true emotion?
Hopefully we can get a personal one, so sarcasm now will work over the internet!
[sarcasm]That would be the best thing ever![/sarcasm]
Haha, take that sarcmark.
Computers will never experience emotions. You can use neural nets to simulate emotion, but they will never truly have it. It's [b] Artificial[/b] Intelligence for a reason - it's not real intelligence.
[QUOTE=helpiminabox;22087877]How can we tell?
Text is just text.
There's hardly ever emotion behind it.[/QUOTE]
You really can't tell for sure.
[QUOTE=helpiminabox;22087877]How can we tell?
Text is just text.
There's hardly ever emotion behind it.[/QUOTE]
It's harder, but if bolds/italics/caps/other emphasis are used, or the right wording, to make it visible though text, but not obnoxiously so, then it can work.
[QUOTE=|FlapJack|;22089849]Computers will never experience emotions. You can use neural nets to simulate emotion, but they will never truly have it. It's [b] Artificial[/b] Intelligence for a reason - it's not real intelligence.[/QUOTE]
Yes. This is also why computers will never be able to generate truly random numbers. It's all an algorithm, there's nothing you can do to make a computer think or feel, it's all programmed commands.
[QUOTE=suslowim;22091097]It's harder, but if bolds/italics/caps/other emphasis are used, or the right wording, to make it visible though text, but not obnoxiously so, then it can work.[/QUOTE]
[b]oh[/b] [i]you[/i]
[QUOTE=Heroms;22091364]Yes. This is also why computers will never be able to generate truly random numbers. It's all an algorithm, there's nothing you can do to make a computer think or feel, it's all programmed commands.[/QUOTE]
Yes, but the reductionist theory states that the human brain is also reducible to a series of "programmed commands". (I personally don't believe it; I'm just playing devil's advocate.)
Not too fond of the mysticism surrounding human conciousness, personally. I firmly believe it is possible to create a synthetic duplicate to the human conciousness.
[QUOTE=InsanePyro;22087813]How can it tell?
Text is just text
Theres hardly ever emotion behind it[/QUOTE]
There's text patterns and such.
Yeah that's a real useful invention
Sorry, you need to Log In to post a reply to this thread.
