Become a fan of Slashdot on Facebook

 


Forgot your password?
Close
typodupeerror
×
Technology

Tech Companies Looking Into Sarcasm Detection 167

Posted by Soulskill from the meta-comment-challenge:-complete-sincerity dept.
Nerval's Lobster writes "Now here's the greatest thing ever: French tech firm Spotter has apparently devised an analytics platform capable of identifying sarcastic comments, according to the BBC. Spotter's platform scans social media and other sources to create reputation reports for clients such as the EU Commission and Air France. As with most analytics packages that determine popular sentiment, the software parses semantics, heuristics and linguistics. However, automated data-analytics systems often have a difficult time with some of the more nuanced elements of human speech, such as sarcasm and irony — an issue that Spotter has apparently overcome to some degree, although company executives admit that their solution isn't perfect. (Duh.) Spotter isn't alone: IBM, Salesforce, and other IT vendors are hard at work on analytics software that can more perfectly determine when you're mouthing off, you little punks. In theory, sarcasm detection can help with customer service, and judging how well products are doing on the open market... and we all know it's going to work perfectly, right? Nothing could possibly go wrong with automated platforms built to assess the nuances of human speech."
This discussion has been archived. No new comments can be posted.

Tech Companies Looking Into Sarcasm Detection More

Tech Companies Looking Into Sarcasm Detection

Comments Filter:

  • Great! (Score:3)

    by nospam007 ( 722110 ) * on Friday July 05, 2013 @04:14PM (#44198787)

    I hope they get 75% of it right. My personal guess is that around 25% of humans are unable to detect any sort of sarcasm, perhaps not quite as bad as Sheldon, but quite bad.

    • Re:Great! (Score:4, Funny)

      by ebno-10db ( 1459097 ) on Friday July 05, 2013 @04:22PM (#44198889)

      My personal guess is that around 25% of humans are unable to detect any sort of sarcasm

      And most of them seem to post at Slashdot.

      P.S. Irony, satire and facetiousness don't fare too well either.

      • Re:Great! (Score:4, Informative)

        by mcrbids ( 148650 ) on Friday July 05, 2013 @04:28PM (#44198935) Journal

        But Poe's law [wikipedia.org] predicted a long time ago, that such detection is, in many cases, actually impossible to accomplish.

        • Re: (Score:2)

          by ackthpt ( 218170 )

          But Poe's law [wikipedia.org] predicted a long time ago, that such detection is, in many cases, actually impossible to accomplish.

          Nothing prevents people from selling stock in a venture, particularly if the listening audience isn't already rolling up it's pants cuffs.

        • Re: (Score:2)

          by Kjella ( 173770 )

          But this only deals with one specific instance, if you have a poster with a history of tinfoil hat posts say "Sure, I totally believe NSA has only my best interests at heart" then that has a lot higher probability of being irony than a poster that is fully in the "if you have nothing to hide, you have nothing to fear" corner. Given all the defective sarcasm and irony detectors out there, the bar of out-detecting a human is pretty damn low.

          • Re:Great! (Score:5, Interesting)

            by plover ( 150551 ) on Friday July 05, 2013 @06:13PM (#44199695) Homepage Journal

            The problem comes with professional violators of Poe's Law, such as Stephen Colbert's character, "Steven Colbert of the Colbert Report". He's a parody of every right wing nut job talk show host. His schtick is to take a right-wing agenda item and push it beyond its obvious short term benefits to its logical but socially detrimental conclusion, where he continues to defend it even more vigorously using Republican platform talking points, ad hominem attacks, and every other logical fallacy [wikipedia.org] he can throw at it. He does this consistently without ever breaking character. And he has a flock of brilliant writers who are able to help him pull this off night after night.

            As a matter of fact, he is so consistent that he was mistaken for an actual right wing comedian, and was invited to speak at the White House Correspondent's Dinner in 2006 where he lampooned George W. Bush to his face for fifteen straight minutes. Very few of the faithful present laughed at the routine. President Bush turned red almost from the get-go, politely grimaced out a smile, sat through the entire speech, and left the stage immediately after Colbert finished. I have no doubt that heads rolled within five minutes. ( My favorite joke from the event went something like, " 'Those naysayers claim that this administration is just rearranging the deck chairs on the Titanic.' That is a terrible metaphor. This administration is not sinking. This administration is soaring! If anything, they are rearranging the deck chairs on the Hindenburg!" )

            • Re:Great! (Score:4, Informative)

              by JustOK ( 667959 ) on Friday July 05, 2013 @06:36PM (#44199815) Journal

              Oh, come on. Everyone knows it's not an act.

        • I propose that the very attempt to detect it is in fact a form of it.
          Especially when the output is, "No, he's being totally serious, really!" :)
      • I was about to post a snarky reply, but come on, it's so deep around here you can't even scuba dive to the bottom. ;)

        And yes, I do know the previous poster just added to the depth. :p
      • Sarcasm in general does not fare well in text media. As practiced in spoken language, there's a very significant component of intonation and body language. Of course it can be done and well in text media, but it's a skill most people don't have, like writing good expository prose.
      • They are doing humanity a great service. Every penny that went into this research is money well spent.

      • Were you being sarcastic?

    • Re: (Score:2)

      by gweihir ( 88907 )

      Cool! Finally a tool that lets me find out whether I am making factual statements, sarcastic comments, or indulging in irony! I have a hard time telling what it is so far.

      Seriously, this is just another piece of BS sold for a lot of money. It cannot work without working AI, and that is still completely out of reach.

  • Yeah... (Score:5, Funny)

    by ackthpt ( 218170 ) on Friday July 05, 2013 @04:14PM (#44198791) Homepage Journal

    Like that's going to work.

  • Sarcasm is nice, but what about detecting the effect of Poe's law?

    • Re: (Score:3)

      by icebike ( 68054 )

      Why would you expect the software to be any better at this than the humans?

  • If they make it work and ever point that at slashdot, the readings are gonna be flying off the charts!

    • If this thing works, it's going to be a new Turing Test, judging by the number of posts here that simply state: "Whoosh!"

    • Re: (Score:3)

      by ackthpt ( 218170 )

      If they make it work and ever point that at slashdot, the readings are gonna be flying off the charts!

      Slashdot would be a poor test bed for the project. Sarcasm is too easily detected on here to be useful, it's as subtle as being hit by a brick.

      Now ... if they pointed it at Faux Nooz, that would be pretty interesting to see how much the presenters don't believe of the garbage they're spewing to keep the market other broadcasters have neglected: the disenfranchised intelligentsia.

  • (My turn)

  • Until they run into this guy... (Score:4, Funny)

    by decipher_saint ( 72686 ) on Friday July 05, 2013 @04:17PM (#44198817)

    Oh! I'm not being sar-cas-tic.

    http://youtu.be/ziH9St7ajuw [youtu.be]

  • Gee, that's useful.

  • Fool's errand (Score:5, Insightful)

    by ericloewe ( 2129490 ) on Friday July 05, 2013 @04:17PM (#44198831)

    Sarcasm is very frequently indicated by nuances that aren't transmitted through text. If humans have trouble getting sarcasm out of text, why should an algorithm do any better with the same set of data?

    • Re: (Score:2)

      by mysidia ( 191772 )

      Perhaps we should have a new computer keyboard that senses biological changes in the typist, to infer things like sarcasm; typing rate, skin conductivity, pulse, body temp, combine with webcam, add facial expressions.

      or more EMoTICONS and markup such as [SARCASM]text here[/SARCASM]

    • Re: (Score:2)

      by gweihir ( 88907 )

      Their great technology will make it work! It is so much better than what puny humans can do and not bound by the limitations of text. In fact, I predict the core technology is an advanced quantum-bogon-detector, that will even be able to classify statements before they are made or if they are not made at all!

    • Re: (Score:2)

      by Guppy ( 12314 )

      Sarcasm is very frequently indicated by nuances that aren't transmitted through text. If humans have trouble getting sarcasm out of text, why should an algorithm do any better with the same set of data?

      :P

    • Re: (Score:2)

      by arth1 ( 260657 )

      I think the problem is the other way around. People spot sarcasm where there is none, mistaking irony for sarcasm.

    • Depends who writes the code. British folk will find sarcasm far more easily than, say, American folk (since that's pretty much their default mode). So if it's a British coder, it might work.

  • a sarcasm detector that a real useful invention (Score:5, Informative)

    by Joe_Dragon ( 2206452 ) on Friday July 05, 2013 @04:20PM (#44198861)

    simpsons did it

    • Oh here comes that cannonball guy. He's cool.
      Are you being sarcastic, dude?
      (Hangs head) I don't even know anymore.

  • There's no way they'll just leave it to languish once it's working. Corporations just love bad news and never ignore negative feedback.

  • Do we have a sarcasm sign?

  • Sarcasm detector (Score:5, Insightful)

    by mysidia ( 191772 ) on Friday July 05, 2013 @04:35PM (#44198985)

    What we really need is a lie detector, a spam detector, and a troll detector

    Extra points for the spam detector, THAT is what is most sorely needed, and what is so inadequately provided thus far.

    • It's an arms race. The better the detector, the better the spammer, the better the detector...

      But, can you imagine how good sarcasm would get if they started treating it the same way? Words so powerful they could melt your screen!

    • Why do you need a spam detector. Can't you find spam without it?

      • Re: (Score:2)

        by mysidia ( 191772 )

        Why do you need a spam detector. Can't you find spam without it?

        Sure... I can assume everything is spam. Then I have a new problem: How can I detect messages that have no spam?

  • Small drawback (Score:3)

    by AdamWill ( 604569 ) on Friday July 05, 2013 @04:39PM (#44199031) Homepage

    Unfortunately, there's been a setback in the schedule. They tested it on Slashdot and it exploded.

  • to make sure I get my snark tuned exactly right.

  • Yeah, right... (Score:3)

    by hyades1 ( 1149581 ) <hyades1@hotmail.com> on Friday July 05, 2013 @04:43PM (#44199061)

    ...They can't tell that "Give her joy by split her halfways with yoou massif kawk" is spam, but they're going to identify sarcasm with a big analytics package.

    Right.

    • Detect for "right." alone in a sentence, add bonus point when it separate the main text with a cariage return. "Yeah, right." make the detector explode.

      • LMFAO. Sounds like a plan. Sort of like that pathetic computer in the original Star Trek that always seemed to be getting hosed in by being ordered to compute Pi to the last decimal place.

        "Computer: identify and grade snark. Execute!"

  • Often the detection of sarcasm relies on understanding of popular opinion on a topic. I don't think we'll have any magic bullet algorithm to detect sarcasm until we have hard AI with a far-reaching corpus of current knowledge. Take these two sentences: "DRM is the best. It makes everything so much easier!" and "The iPhone is the best! It makes everything so much easier!" Ok, algorithm. Pick the one containing sarcasm...

  • sarcasm = (company.attributes.include?([:big_and_evil]) && comment.classification == "complimentary")

  • False positives in both sides (Score:3)

    by gmuslera ( 3436 ) on Friday July 05, 2013 @04:52PM (#44199153) Homepage Journal
    Is not like you won't end in jail for a sarcastic comment [cnn.com], or get expelled over a joke [theblaze.com], it will work in the other way, seeing sarcams where they aren't and getting you anyway. And getting this mess in your private mail, where you usually joke and don't care a lot about potential readings of what you say, because, well, you don't have anything to hide, will make life interesting in the next years.

    • Re: (Score:3)

      by lemur3 ( 997863 )

      When I first saw this story I wondered if they were to apply it in a legal sense, and not for marketing..

      http://www.publications.parliament.uk/pa/cm200405/cmselect/cmhaff/80/80we20.htm [parliament.uk]

      they have things called "anti social behavior orders" in england to curb.. well, anti social behavior, one such example was this (source above):

      The oldest recipient of an order to date is an 87-year-old who among other things is forbidden from being sarcastic to his neighbours (July 2003). He was subsequently found guilty of b

  • 'Elysium' Trailer [youtube.com], 33 seconds in.

  • " In theory, sarcasm detection can help with customer service, and judging how well products are doing on the open market... "

    Or, just perhaps, marketing could read (listen?) for themselves to see how things turned out...

    Next up....

    Sarcasm in 3D !!!

  • And brought to justice! How dare they making legal and ethical NSA interception and interpretation of all communication harder! That amounts to terrorism! Time to find all these thought-criminals and lock them away for good. All clear speaking and thinking citizens will live in a better world for that.

  • You have to have a clear notion of what's expected to identify irony, and that's a function of the topic, the venue, and the history of the writer.

    Fortunately, the utter brilliance the designers have shown by thinking of the idea in the first place will carry them beyond such minor details and bring them complete success.

  • -1 Woosh (Score:3)

    by doug141 ( 863552 ) on Friday July 05, 2013 @05:18PM (#44199337)
    I've often thought slashdot would benefit from a -1 Woosh mod option.
    • In general it's not a good idea to mod such things down, it's better to reply to them. Because if one person wooshed, other people will too; which means that someone else with mod points will come by later to 'fix' your downmod.

      Modding is primarily to get rid of spam and GNAA, and secondarily to bring attention to really interesting posts. It's not there to act as an adjudicator between 'right' and 'wrong.' That's why there's no -1 Wrong mod option.

      • I have visioned that there should be only the possibility to mod posts up. For GNAA junk a "report spam" link.

        And more mod points to people! There's always someone saying "I wish I had mod points for you".

        • I agree absolutely with the second line, and agree tentatively with the first line.

          I would further suggest that limited mod points are a serious factor limiting the growth of Slashdot. People don't want to comment if they have no chance of being modded up. More mod points would help that.
  • Of course, the illegible drivel that sits atop most /.-pages defies classification even by humans, so some margin for error is reasonable.

  • ...sarcasm detection...

    What could go wrong?

  • It's a function of intent and inflection.

    Trying to divine it from raw text is going to fail. Simply because such systems will be deprived of the necessary information to make such a call properly.

    Sure, old chestnuts like "Nothing could POSSIBLY go wrong!" might trip it. But sarcasm extends beyond the basics and into some fairly obscure, arcane and downright subtle usage.

    It's going to be like handing a blind person a ball and asking them to divine the color.

  • "French tech firm Spotter has apparently devised an analytics platform capable of identifying sarcastic comments" Hence forth referred to as Sheldon Cooper.

  • Sarcasm detector pointed at Slashdot, promptly exploded.

  • ...a future episode of The Big Bang Theory. I'd say "notify Lorre" but I'm certain he already knows and is computing the comic possibilities.

  • EU Commission could save some money here. The algorithm to detect sarcasm when speaking of EU Commission is simple, as nobody ever tells anything good about the EU Commission : If a sentence has a positive word, then it is sarcasm.

  • Missing the point (Score:4, Interesting)

    by Required Snark ( 1702878 ) on Friday July 05, 2013 @11:30PM (#44201209)
    Everyone is missing the point. The real question is why do they want a sarcasm detector?

    Implicitly we all realize that they want to filter sarcastic remarks out of online posting. Sarcasm is a very effective way to combine criticism and humor, and the result can be a very effective critique. This makes it very troublesome to those with power and money. They don't want anyone rocking the boat or getting uppity.

    So instead of addressing potentially meaningful critical responses, or accepting the reality that people enjoy making bad jokes, they seek to automate the process of self serving censorship.

    The intent is bad. I'm sure that organizations considering using this technology don't care about false positives. What they want is for you to STFU, unless you say what they want you to say.

    So while Slashdot posters make the truly obvious jokes, or argue about technology and false positive/negative rates, this reveals the ugly truth about the intent of big online organizations. They want to enforce a one way channel where users are censored. Considering that Slashdot considers itself to be an elite corner of the internet, I find it pathetic that no one has a clue about what this means.

    • They are just trying to measure public opinion by sorting comments into "positive" and "negative". They can already sort out most of the negative comments, their shortfalling is false positives due to sarcsm. If their purpose was to censor negative opinion, then they would already be censoring all those non-sarcastic posts they don't like.

      In other words, I have no clue how you got modded up because your argument makes no sense.

  • I have the algorithm here (Score:3)

    by SuperKendall ( 25149 ) on Saturday July 06, 2013 @12:47AM (#44201487)

    1) Is the internet connected?

    2) Well there you go.

  • And can easily be achieved without any sarcasm detection.
    Social media mentions have a strong positive bias, so simply guessing positive all the time will get you
    pretty close to the 80% mark. and with a simple list of negative phrases you can pass the 80% mark in identifying if a social media mention(twitter, facebook, etc.) is positive or negative.

  • This whole idea that sarcasm doesn't come through in text needs to be revisited.

    I have a reputation in my work environment for being perceptive, thoughtful, and lucid. I also have a reputation for having near perfect recall of anything previously discussed that could possibly go wrong, and for sometimes becoming extremely intense and hard to deter from constantly injecting these unhappy reminiscences into self-satisfied negotiations until everyone else glasses over. Others might characterize this as a gee

Slashdot Top Deals

GREAT MOMENTS IN HISTORY (#7): April 2, 1751 Issac Newton becomes discouraged when he falls up a flight of stairs.

Close