Ivan’s private site

March 24, 2010

Twitter search results in RDF

Filed under: Semantic Web,Work Related — Ivan Herman @ 23:58
Tags: , , ,

This is cool: with the Shredded Tweet site one can search a twitter tag, get back the result in RDFa. Ie, the result can be ran through an RDFa tool to get the data in RDF, ready to be mashed up with other stuffs. For example, one can search the #w3c tag to get to the RDFa page; run it to the RDFa distiller, to yield an RDF data:

<http://twitter.com/Eyalsela/statuses/10981280232> a <sioc:Post> ;
     dcterms:created "2010-03-24T14:23:06+00:00" ;
     sioc:content "At our second meetup: Mobile applications and websites development
        #w3cdf #W3C [English:http://j.mp/bS8qLU Hebrew: http://j.mp/a3i8Tv"@en ;
     sioc:has_creator <http://semantictweet.com/Eyalsela#me> ;
     sioc:has_topic <http://search.twitter.com/search?q=%23W3C>,  ;
     sioc:links_to <http://j.mp/a3i8Tv> . 

<http://twitter.com/JimThijssen/statuses/10979370533> a <sioc:Post> ;
     dcterms:created "2010-03-24T13:40:04+00:00" ;
     sioc:content "Yeahh #www.stagematcher.nl is volledig #html #w3c #certified B-)"@en ;
     sioc:has_creator <http://semantictweet.com/JimThijssen#me> ;
     sioc:has_topic &t;http://search.twitter.com/search?q=%23certified>,
       <http://search.twitter.com/search?q=%23html>,
       <http://search.twitter.com/search?q=%23w3c> .
...

The RDF content also be directed used, eg, in a SPARQL call, but the combined URI-s of the distiller service and the original data; ugly URI, but can be minted programatically fairly easily. This is cool…

Reblog this post [with Zemanta]
Advertisements

8 Comments

  1. Great service, thanks for drawing my attention to it (via Twitter).

    You can just GET the RDF directly (without using the Distiller) though via URIs like:
    http://pegasus.chem.soton.ac.uk/search.rdf?q=%23cash-gordon

    Comment by Barry Norton — March 25, 2010 @ 8:07

    • Ah! That makes it even quicker, I did not realize that is also possible. That is even cooler:-)

      Comment by Ivan Herman — March 25, 2010 @ 10:36

  2. It’d be cool to have an SPARQL endpoint of such data

    Comment by Sergio — March 25, 2010 @ 8:55

  3. I tried Shredded Tweet, and it seems cool.. I downloaded the firefox extension Fuzz(http://rdfa.digitalbazaar.com/fuzz/trac/) to detect triples on search results..

    Comment by Nadin — March 25, 2010 @ 9:20

  4. […] I was reading my rss feeds, I saw a blog post of Ivan Herman. In this post, he’s talking about this new engine: Shredded Tweet. As the […]

    Pingback by nadinima » Blog Archive » Shredded Tweet — March 25, 2010 @ 10:24

  5. Another option is to use rtsw – it delivers live streamed RDF data from the Twitter stream API.
    You can restrict by tag to get only posts matching a particular #hashtag

    http://code.google.com/p/rtsw/

    Comment by Alex — March 25, 2010 @ 18:33

  6. pegasus data has issue:
    any23: Could not parse input: org.openrdf.rio.RDFParseException: unexpected literal [line 2, column 39]
    raptor: librdf error 1 column 0 – XML parser error: Extra content at the end of the document

    element http://gitorious.org/element and triplr http://triplr.org transform to RDF from twitter atom results:

    Comment by carmen — April 1, 2010 @ 20:55


RSS feed for comments on this post.

Create a free website or blog at WordPress.com.

%d bloggers like this: