Hi,
Thanks for the articles :)...much helpful than the technical documentation.
Cristi
On Sun, Mar 25, 2012 at 5:16 PM, Jimmy O'Regan <[email protected]> wrote:
> [Changing CC to the developers' list]
>
> On 25 March 2012 14:46, Cristi Barca <[email protected]> wrote:
> > Hello,
> >
> > My name is Cristian Barca and I'm a fourth year (final year) student at
> > University Politehnica of Bucharest.
> > My study topics for the final year are now Machine learning and
> Distributed
> > Computing.
> > I am currently conducting my bachelor thesis in Artificial Network as I
> am
> > supposed to train an artificial neural network in a distributed way
> (using
> > Hadoop, HDFS and Cassandra...and mostly NN algorithms that work in batch
> > mode). I have been experimenting for a while few neural network
> algorithms
> > such as Backpropagation, Rprop, LMA, etc on large datasets, millions of
> > data, from Physiobank[1].
> >
> > Since I am both passionate about Machine Learning and Distributed
> Computing
> > my attention was drawn by the Hadoop Based Index project and also
> (because
> > the previous one is a starter task) by the Spotting project. I also think
> > that we could connect both, in terms of Grid Computing (Hadoop)...but to
> > extend this idea I would like to ask you some short questions:
> >
> > 1. What is the flow of DBPedia design when it comes to spotting? (it
> would
> > be useful to share some links where this process is described)
>
> Have you read the technical documentation?
> (http://wiki.dbpedia.org/spotlight/technicaldocumentation?v=3qy)
> The paper mentioned under 'Citations'?
> (
> http://www.wiwiss.fu-berlin.de/en/institute/pwo/bizer/research/publications/Mendes-Jakob-GarciaSilva-Bizer-DBpediaSpotlight-ISEM2011.pdf
> )
>
> Not linked from the website (an oversight?):
>
> http://www.wiwiss.fu-berlin.de/en/institute/pwo/bizer/research/publications/Mendes-Daiber-Rajapakse-Sasaki-Bizer-DBpediaSpotlight-LREC2012.pdf
>
> > 2. You say that you have already implemented Named Entity Recognition,
> > Keyphrase Extraction, N-Grams, Lexicon-based, etc. Did you implemented
> from
> > scratch or do you use a framework?
>
> Weka, OpenNLP, and LingPipe are used.
>
> --
> <Sefam> Are any of the mentors around?
> <jimregan> yes, they're the ones trolling you
>
--
-Cristi Bârcă-
------------------------------------------------------------------------------
This SF email is sponsosred by:
Try Windows Azure free for 90 days Click Here
http://p.sf.net/sfu/sfd2d-msazure
_______________________________________________
Dbp-spotlight-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/dbp-spotlight-users
