RE: SKOS meets Semantic blogging (again)?

Cool! Haven't looked at the stuff in great detail, but one thing struck me:

> Still needs a more polished use case, but I'm interested to 
> see what we find from a few dozen folks sites. Dave B 
> mentioned that Wordpress categories are closer to 'keyword 
> lists' than Thesauri, which is fair comment, but they do 
> allow hierarchy and do reflect the scruffy way people tend to 
> manage their bookmarks, filetrees, documents etc. 

I like the idea of using this as an additional similarity metric. 

So 'find items like X' could be 'find items with textual content similar to
X' or
'find more items which people have co-categorised with X". 

Or both. 

Or more refined versions of each (for the content based approach you could
have LSI techniques or trigrams etc, for the semantic based technique one
could have semantic distance along a taxonomy).

This is sort of what I'm doing in a project I call "semantic
classification".

Steve
 

> -----Original Message-----
> From: public-esw-request@w3.org 
> [mailto:public-esw-request@w3.org] On Behalf Of Dan Brickley
> Sent: 12 August 2004 13:56
> To: public-esw@w3.org
> Subject: SKOS meets Semantic blogging (again)?
> 
> 
> 
> Just a quick pointer to a hack I started at the weekend, 
> to export SKOS (in N3, for now) from the PHP weblogging system, 
> Wordpress.
> 
> src: http://danbri.org/words/skos.phps
> categories from my blog: http://danbri.org/words/skos.php
> 
> Source code has a few more pointers. It's very basic, but 
> should fit well with the FOAF and other RDF addons for Wordpress. 
> 
> More notes/background in: 
> http://rdfig.xmlhack.com/2004/08/09/2004-08-09.html including 
> feeds from others...
> 
>  - http://crschmidt.net/wordpress/skos.php
>  - http://www.brain-stream.org/wordpress/skos.php
> 
> Folks were also, as a spinoff from this discussion, playing 
> with export of categories from del.icio.us (shared 
> bookmarking 'tags'). I've also pinged the Annotea team to see 
> where they're up to w/ SKOS adoption. See rdfig.xmlhack.com 
> link above for more re deli.icio.us feeds.
> 
> 
> This demo was really "one to throw away". RDF/N3 isn't the 
> right format to do exports in (except as a negotiable option 
> amongst concenting parties in the privacy of their own 
> HTTP/1.1 transaction). Morten F has begun re-working this 
> code as an RDF/XML 'categories in RDF' feature of his 
> Wordpress FOAF plugin.
> 
> I'm quite psyched about this, 'cos there are lots of sites 
> using Wordpress, and the FOAF/XFN stuff in there already 
> gives us a basic 
> framework for inter-site navigation, discovery and 
> harvesting. This means we can send a crawler around and scoop 
> up different people's category sets. Which should be an 
> interesting dataset, particularly as Wordpress allows for 
> category-specific RSS feeds.
> 
> Still needs a more polished use case, but I'm interested to 
> see what we find from a few dozen folks sites. Dave B 
> mentioned that Wordpress categories are closer to 'keyword 
> lists' than Thesauri, which is fair comment, but they do 
> allow hierarchy and do reflect the scruffy way people tend to 
> manage their bookmarks, filetrees, documents etc. 
> 
> thoughts welcomed,
> 
> Dan
>

Received on Thursday, 12 August 2004 13:15:59 UTC