The data originally released on my data release page has now been made available as an API. It is accessed by making REST requests to the service and the result is returned in a SOAP envelope. Two requests are currently supported: A bounding box and a Wikipedia Article.

Request format:

http://www.doc.ic.ac.uk/~seo01/numenore/api.php?operation=getwikipages&minlat=51.25&maxlat=51.75&minlon=-0.25&maxlon=0.25&language=en

The Reply is the number of references made to locations within that bounding box, and a list of Wikipedia articles describing those locations.

Reply Format:

<soap:Envelope soap:encodingStyle="http://www.w3.org/2001/12/soap-encoding">
  <soap:Body>
    <sum frequency="4424"/>
    <pages>
      <page name="Greater_London"/>
      <page name="London"/>
      <page name="Wilmington"/>
      <page name="City_of_Westminster"/>
      <page name="Bayswater"/>
      <page name="Bermondsey"/>
      ...
    </pages>
  </soap:Body>
</soap:Envelope>

The request is the title of a Wikipedia Article.

Request Format:

http://www.doc.ic.ac.uk/~seo01/numenore/api.php?operation=getsynonyms&title=France&language=en

If the page does not refer to a location the reply contains following message:

<soap:Envelope soap:encodingStyle="http://www.w3.org/2001/12/soap-encoding">
  <soap:Body>
    <notlocation/>
  </soap:Body>
</soap:Envelope>

Otherwise the reply is the list synonyms used to refer to that article and how often each synonym is used.

Reply Format:

<soap:Envelope soap:encodingStyle="http://www.w3.org/2001/12/soap-encoding">
  <soap:Body>
    <syns>
      <syn text="France" freq="6045"/>
      <syn text="French" freq="1257"/>
      <syn text="(France)" freq="114"/>
      <syn text="F" freq="46"/>
      <syn text="Frenchman" freq="5"/>
    </syns>
  </soap:Body>
</soap:Envelope>

Google Map overlays are available of the maps produced as part of my PhD showing the frequency of references to locations in Wikipedia. A sample of the nessessary Javascript is shown below and assumes familiarity with the Google Maps API:

tileUrl = 'http://www.doc.ic.ac.uk/~seo01/numenore/global/mapper.php?map=en&x={X}&y={Y}&scale={Z}';
var tileLayerOverlay = new GTileLayerOverlay(
  new GTileLayer(null, null, null, {
    tileUrlTemplate: tileUrl, 
    isPng:true,
    opacity:0.7
  })
);
map.addOverlay(tileLayerOverlay);

The map argument in the Title url refers to the language version of Wikipedia being used. Currently 10 languages are supported: English (en), German (de), French (fr), Spanish (es), Portuguese (pt), Esperanto (esp), Welsh (cy), Chinese (cn), Arabic (ar), and Hebrew (he).

PhotoSoup is a photo word puzzle developed while I was on internship at Yahoo Research Barcelona. PhotoSoup uses creative commons licenced photos shared on Flickr to generate a word puzzle. The photos are provided as clues and the tags of the photos have to be found in the puzzle. Photo soup began as an entry for the Yahoo EU hack day where it won “Coolest Hack”.

PhotoSoup was developed by Lluis Garcia Pueyo, Borkur Sigurbjornsson, Vanessa Murdock, Carlos “Chato” Castillo, Simon Overell and Roelof van Zwol. Personalised photo soup puzzles can be generated and embeded into web pages or facebook pages and shared. Read more on the Photo Soup project page and Yahoo Next Blog.

Image Navigator is a JavaScript Application used to browse high definition pictures over the internet. Image Navigator allows users to zoom, pan, drag and click to navigate images in an intuitive manner. Interactive overlays and labels can be displayed to provide extra information to the user. Image Navigator is highly configurable and easily integrated into any website.
Click here for a demo