You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@stanbol.apache.org by gm41lu53r <gm...@gmail.com> on 2014/04/17 17:40:36 UTC
IPTC indexing for morons
Hi,
Can you provide me with more information on how to do IPTC news indexing
with Stanbol?
Regards,
Scott
Re: IPTC indexing for morons
Posted by Rupert Westenthaler <ru...@gmail.com>.
Hi Scott,
You can find an index for IPTC at [1]. This index is about 2 years
old. Some Years ago I also sent a mail about IPTC indexing to the
stanbol mailing list [2].
Configuring an EntityLinking Chain with IPTC will definitely work.
However typically it is not so useful. The reason for that is that
news do mention Entities (politicians, soccer players, places, cities,
organizations, ...) and the IPTC thesaurus consists categories such as
politic, soccer ... So for Entity linking you will need to use a
controlled Vocabulary containing such entities (e.g. dbpedia or a
custom vocabulary).
IPTC is much better suited for Document Classification. For that you
can use the topic classification engine as described in [3]. You will
need to have pre-classified news articles to train the engine.
best
Rupert
[1] http://dev.iks-project.eu/downloads/stanbol-indices/iptc/
[2] http://markmail.org/message/rgwug74s3u6olrby
[3] http://www.iks-project.eu/sites/default/files/Topic-Classification.pdf
On Thu, Apr 17, 2014 at 5:40 PM, gm41lu53r <gm...@gmail.com> wrote:
> Hi,
>
> Can you provide me with more information on how to do IPTC news indexing
> with Stanbol?
>
> Regards,
> Scott
--
| Rupert Westenthaler rupert.westenthaler@gmail.com
| Bodenlehenstraße 11 ++43-699-11108907
| A-5500 Bischofshofen
| REDLINK.CO ..........................................................................
| http://redlink.co/