Parsing Miss South Carolina's Statement
It’s not like it’s easy to parse Wikipedia, but at least most of the its text is (usually) written with correct spelling, capitalized proper names, meaningful paragraph structure and so on. A natural question is: how will our system perform on the rest of the Web with all of its slang, non-standard syntax, and so on? To put Powerset to the test, two of our engineers, Lukas Biewald and Brendan O'Connor, ran our entire parsing and indexing system on the hardest corpus we could find: Miss South Carlolina's response to the question, "Recent polls have shown that a fifth of Americans can't locate the US on a map. Why do you think this is?"
They fed this transcription into the XLE verbatim, disfluencies and all:
I personally believe that U.S. Americans are unable to do so because uh some uh people out there in our nation don't have maps and uh I believe that our ed- education like such as in South Africa and uh the- the Iraq everywhere like such as and I believe that they should uh our education over here in the U.S. should help the U.S. or- or- should help South Africa and should help the Iraq and the Asian countries so we will be able to build up our futureOne might think that such a convoluted mess of words (I hesitate to call it "English") would be impossible to parse, but here is the C-structures that our parser generates:
Unsurprisingly, the sentence is fragmented quite a bit, but the parser clearly managed to extract useful structure throughout the sentence. The last large verb phrase “should help South Africa and should help the Iraq and the Asian countries so we will be able to build up our future” seems very close to correct, which is pretty impressive (Language Log has more discussion on the weird “the Iraq” construction). The output of the semantics system is too long to put here, but in some ways, it’s amazing that we were able to extract any semantics at all.
And, believe it or not, we can actually run some queries against the Carolina Index (as it's known at Powerset). It's hard to think of a reasonable question, but we asked, "Who does education help?" and returned and highlighted the right answer: "Americans". Or should we have returned "U.S. Americans"?
Comments (29)
I am a little surprised that the system does not return the U.S. from the phrase “our education over here in the U.S. should help the U.S.” instead of Americans, assuming that semantics at the phrase level / sentence level is being used along with additional lexico-syntactic features etc.. The word/entity “Americans” highlighted seems to be somewhat dislodged with respect to the query context unless some sort of discourse analysis is als taken into account.
Posted by Rajen Subba | September 01, 2007 03:40 AM
Posted on September 01, 2007 03:40 AM
I am a little surprised that the system does not return the U.S. from the phrase “our education over here in the U.S. should help the U.S.” instead of Americans, assuming that semantics at the phrase level / sentence level is being used along with additional lexico-syntactic features etc.. The word/entity “Americans” highlighted seems to be somewhat dislodged with respect to the query context unless some sort of discourse analysis is als taken into account. But if it doing that, and is not merely due to chance based on some statistical model, then thats pretty impressive.
Posted by Rajen Subba | September 01, 2007 03:41 AM
Posted on September 01, 2007 03:41 AM
So if this is what the parse tree for something as horrible as Miss South Carlolina’s response, what does the tree look like for a properly structured paragraph?
Posted by Jeff Smick | September 01, 2007 08:08 PM
Posted on September 01, 2007 08:08 PM
Hilarious. Absolutely amazing. I bet Ms. SC wishes that everyone in the audience was as fluent as Powerset.
Posted by Neal | September 03, 2007 10:54 AM
Posted on September 03, 2007 10:54 AM
Keep feeding this crapola into Powerset and it’ll start thinking “uh” is an important word…
Posted by Jeremy Toeman | September 03, 2007 01:44 PM
Posted on September 03, 2007 01:44 PM
This was a very smart idea by powerset. Kudos on the original thinking behind parsing Miss South Carolina’s statement.
Posted by Avi | September 03, 2007 02:18 PM
Posted on September 03, 2007 02:18 PM
http://www.youtube.com/watch?v=ccYoVnBc_fk&mode=related&search= YouTube - George Carlin: education and the owners of America
Posted by John Smith | September 05, 2007 06:15 PM
Posted on September 05, 2007 06:15 PM
Two of our students ran the answer through UMBC’s Ontosem text understanding system. See the results here…
http://ebiquity.umbc.edu/blogger/2007/09/05/parsing-blond-speak/
Posted by tim finin | September 06, 2007 03:00 PM
Posted on September 06, 2007 03:00 PM
Powerset technology is far from being the state of the art, but their P.R are defiantly do.
Posted by Gili Sabo | September 09, 2007 03:39 PM
Posted on September 09, 2007 03:39 PM
Powerset technology is far from being the state of the art, but their P.R are definitely do.
Posted by Gili Sabo | September 09, 2007 03:40 PM
Posted on September 09, 2007 03:40 PM
By the when are you guys launching the private beta… I signed up for it and waiting to see…
Posted by Anonymous | September 10, 2007 01:47 AM
Posted on September 10, 2007 01:47 AM
I wonder what a search request from the beauty contestant would do to your search engine.. One thing that would be indespensable is to provide a phrase that shows what the search engine understood.. like the google’s “did you mean this?”. Often they are right.
Interested to know what kind of processing power your robots will have? If it takes 10 years to get thru the net, might be out of date before it is done.
Posted by Benoit | September 18, 2007 02:29 PM
Posted on September 18, 2007 02:29 PM
Ruthless QA. Very good.
Posted by Vadim Berman | September 20, 2007 02:32 AM
Posted on September 20, 2007 02:32 AM
Hello to all, its my new pages about cloth cloth diaper You can buy here 24\7.
Posted by Cloth | September 21, 2007 10:17 PM
Posted on September 21, 2007 10:17 PM
c482t
Posted by t613t | October 01, 2007 01:38 PM
Posted on October 01, 2007 01:38 PM
c482t
Posted by t613t | October 01, 2007 01:39 PM
Posted on October 01, 2007 01:39 PM
Well, it looks like Miss south carolina’s word of wisdom have started to spread over the WWW. Even great philosophers now recognize the rightness of her views and the accuracy of her thoughts:
http://www.kaizou.org/selectpost.php?id=38
Posted by Kaizouman | November 01, 2007 11:08 AM
Posted on November 01, 2007 11:08 AM
早泄 健康 情感 肝癌 食道癌 脑瘤 子宫颈癌 结肠癌
Posted by srysey | November 09, 2007 09:07 AM
Posted on November 09, 2007 09:07 AM
getrel
Posted by sitgetouboc | November 16, 2007 10:50 PM
Posted on November 16, 2007 10:50 PM
slugs slugs [link=http://alexmax88.ifrance.com/slugs.html]slugs[/link] slugs slugs [link=http://alexmax88.ifrance.com/slugs.html]slugs[/link]
Posted by slugs | November 28, 2007 05:32 AM
Posted on November 28, 2007 05:32 AM
diablo diablo [link=http://declin.servemp3.com/pages/diablo.html]diablo[/link] diablo diablo [link=http://declin.servemp3.com/pages/diablo.html]diablo[/link]
Posted by diablo | December 02, 2007 12:00 PM
Posted on December 02, 2007 12:00 PM
diablo diablo [link=http://declin.servemp3.com/pages/diablo.html]diablo[/link] diablo diablo [link=http://declin.servemp3.com/pages/diablo.html]diablo[/link]
Posted by diablo | December 02, 2007 12:00 PM
Posted on December 02, 2007 12:00 PM
diablo diablo [link=http://declin.servemp3.com/pages/diablo.html]diablo[/link] diablo diablo [link=http://declin.servemp3.com/pages/diablo.html]diablo[/link]
Posted by diablo | December 02, 2007 12:01 PM
Posted on December 02, 2007 12:01 PM
who is hotter starfire or blackfire who is hotter starfire or blackfire [link=http://alexmax88.ifrance.com/who-is-hotter.html]who is hotter starfire or blackfire[/link] who is hotter starfire or blackfire who is hotter starfire or blackfire [link=http://alexmax88.ifrance.com/who-is-hotter.html]who is hotter starfire or blackfire[/link]
Posted by who is hotter starfire or blackfire | December 02, 2007 11:54 PM
Posted on December 02, 2007 11:54 PM
who is hotter starfire or blackfire who is hotter starfire or blackfire [link=http://alexmax88.ifrance.com/who-is-hotter.html]who is hotter starfire or blackfire[/link] who is hotter starfire or blackfire who is hotter starfire or blackfire [link=http://alexmax88.ifrance.com/who-is-hotter.html]who is hotter starfire or blackfire[/link]
Posted by who is hotter starfire or blackfire | December 02, 2007 11:54 PM
Posted on December 02, 2007 11:54 PM
google desktop google desktop [link=http://luciemariano.ayudauni.net/styles/google-desktop.html]google desktop[/link] google desktop google desktop [link=http://luciemariano.ayudauni.net/styles/google-desktop.html]google desktop[/link]
Posted by google desktop | December 19, 2007 03:19 AM
Posted on December 19, 2007 03:19 AM
research information about international adoption http://lesleechacon.ayudauni.net/blog/research-information.html research information about international adoption israel in egypt music http://israel-in-egypt.kotick.in/ israel in egypt music problems of outsourcing american jobs http://lesleechacon.ayudauni.net/blog/problems-of-outsourcing.html problems of outsourcing american jobs maytag performa washer repair http://maytag-performa.favosite.in/ maytag performa washer repair retirement planning services http://retirement-planning.favosite.in/ retirement planning services epirubicin http://lesleechacon.ayudauni.net/blog/epirubicin.html epirubicin joss stone mp3 http://joss-stone-mp3.favosite.in/ joss stone mp3 police memorial week 2005 http://police-memorial.favosite.in/ police memorial week 2005 institutional bed http://lesleechacon.ayudauni.net/blog/institutional.html institutional bed mosquito control http://mosquito-control.kotick.in/ mosquito control
Posted by mosquito control | December 21, 2007 01:36 AM
Posted on December 21, 2007 01:36 AM
Too bad for all this spam on the board. But this is quite funny!
Posted by Dan | February 04, 2008 06:15 PM
Posted on February 04, 2008 06:15 PM
Order natural health care products, herbal, vitamin and nutritional supplements for a longer healthier life http://fqhoodia.info
Posted by asgorden | February 25, 2008 02:10 PM
Posted on February 25, 2008 02:10 PM