WordNet Conversion

Hi,

I'm Mark van Assem, a PhD student of Guus Schreiber doing work on 
thesauri and cultural heritage.

Guus asked me to see if I could convert WordNet (Prolog version) based 
on the work of the SWBP WN task force, and suggested to mail the 
results to this list for review/comments/further processing.

The work I did is based on a merger of the OWL schema Aldo Gangemi 
prepared and the (public) discussions of the TF (all referred to in 
the documentation). It concerns a straightforward conversion to this 
schema, e.g. it does not identify parts of WN that can be seen as a 
class hierarchy etcetera.

Below I describe the files I prepared. My apologies for any errors you 
may find; I just finished the conversion and did not do a lot in the 
way yet of detecting errors in the result, besides browsing it in 
Triple20.

Kind regards,
Mark.

________________________________________________________________________


Please find the following at http://www.cs.vu.nl/~mark/wn/

- convertwn.pl (conversion program with documentation on usage in the 
comments)

- wordnet_rationale_mark.txt (explanation of my choices & background 
info including refs to where I got the info)

- wordnet_rationale_mark.html (quickly hacked html version of previous 
file)

- wordnet_datamodel_Mark.owl (slightly adapted version of Aldo 
Gangemi's wordnet_datamodel.owl - see the rationale)

- directory 'src' (contains the WN Prolog 2.0 src files)
- directory 'owl' (output of running convertwn.pl)

Instructions on how you can repeat the conversion can be found in 
convertwn.pl.


-- 
  Mark F.J. van Assem - Vrije Universiteit Amsterdam
        mark@cs.vu.nl - http://www.cs.vu.nl/~mark

Received on Monday, 8 August 2005 13:43:44 UTC