Hi Václav,
Thanks for the introduction. You may want to take a look at existing NER
solutions, such as the Stanford NER or the one from the Mallet toolkit, and
try to train them for Czech. You may also be able to benefit from DBpedia
Spotlight, which performs entity (and concept) extraction and annotation.
Our output is a superset of NER, since it segments the input, assigns types
and further assigns unique identifiers to each entity as well as concept. We
are working on the internationalization, so your help with Czech would be
most welcome.
Cheers
Pablo
On Sep 29, 2011 3:08 PM, "Václav Zeman" <[email protected]> wrote:
> Hello,
>
>
>
> I am a student from the Czech Technical University in Prague and i am
> working on semantic web expansion for Czech language. I need to extract
> semantic data from Wikipedia and use it for development of a "named entity
> recognition" service. I have great interest in cooperation on dbpedia for
> Czech language. Is it possible to add a new namespace for Czech (cs) in
> dbpedia mapping?
>
> My account at mappings.dbpedia.org is "Propan".
>
>
>
> Thank you
>
> Václav Zeman
>
------------------------------------------------------------------------------
All of the data generated in your IT infrastructure is seriously valuable.
Why? It contains a definitive record of application performance, security
threats, fraudulent activity, and more. Splunk takes this data and makes
sense of it. IT sense. And common sense.
http://p.sf.net/sfu/splunk-d2dcopy2
_______________________________________________
Dbpedia-discussion mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/dbpedia-discussion
