Hi Alessandro and Rupert,
thanks for the welcome and the information on how contribute to the project.
As you suggested I posted the code of our engines as a new Feature issue
(https://issues.apache.org/jira/browse/STANBOL-583?focusedCommentId=13251505#comment-13251505)
A demo installation of Stanbol with our engines is now accessible at
http://research.celi.it:8082/
I have worked with revision 1306971 (dated 3/29/12). From the OSGI
console it seems possible to define custom chains.
However I am surely available to keep on working on the contribution and
add the support for the language specific chains if is actually missing
Let me know if you have any trouble/feedback on the engines
Bests,
Alessio
On 04/03/2012 06:11 PM, Alessandro Adamou wrote:
Hi Alessio, and welcome aboard!
It is great to see you're jogging the idea of contributing with so
many engines.
I would like to contribute to the project by submitting this module
(obviously as open source).
Is this possible? How can I obtain a user/password in order to submit
the code to STANBOL svn repository?
You might want to run for committership. If so you are very welcome to
register an account on the Apache issue tracker, open "Improvement" or
"New feature" tickets and post the code of your engines as patches:
https://issues.apache.org/jira/browse/STANBOL
I downloaded the latest version of STANBOL
(0.9.0-incubating-SNAPSHOT) and successfully tested the module
integration within the framework.
What svn revision have you worked with? All pre-release modules are
versioned as 0.9.0-incubating-SNAPSHOT.
I'm asking you because recent revisions added support for enhancement
chains, and given their take on multilinguality, your engines look
pretty much like they could be articulated into pre-configured, or
selectable chains. I actually don't know if there is branching
support, but it would be super if a language-specific chain could be
invoked upon detection by your Language Identifier engine.
The services included so far in the module as Enhancement Engines are:
- a Named Entity Recognition service for French
- a Lemmatizer for Italian, German, Romanian, Russian, Danish (it
creates an annotation on the document whose content is the lemmatized
form of the document)
- a Language Identifier for Italian, French,German,Spanish,
Portuguese, Polish, Hungarian, Dutch, Swedish,Arabic,
Russian,Turkish, Romanian, Greek, Norwegian
- a Document Classification services for Italian, French, German,
English, Spanish, Portuguese that associates a document to DBPedia
classes
The services are free to use for research purposes, while require a
license fee for commercial use.
The Enhancement Engines are by default provided with a demo license
that allows a certain number of daily requests in order to test them.
Proper license keys (both for research purposes and commercial use)
can be obtained on http://linguagrid.org/ web page (or sending me an
email request) and you are encouraged to request them in order to
override the limits of the demo license key.
So, you wrapped calls to your linguistic services into engines for the
Stanbol enhancer, am I right? I guess that for a Stanbol contribution
it's pretty fine to do so as the Zemanta engine does, so long as it is
okay to license the engine's client code under the Apache License 2.
Do you have a demo installation of Stanbol online with your engines
set up?
Best,
Alessandro
--
*************************************
Alessio Bosca, Ph.D.
CELI s.r.l.
Via San Quintino 31
10121 Torino
Tel. +39 011.562.71.15
Fax +39 011.506.40.86
http://www.celi.it
*************************************
