W3C home > Mailing lists > Public > semantic-web@w3.org > December 2010

Re: csv to rdf

From: Boris Villazón Terrazas <bvillazon@fi.upm.es>
Date: Fri, 10 Dec 2010 02:46:57 +0100
Message-ID: <4D018691.6030206@fi.upm.es>
To: adasal <adam.saltiel@gmail.com>
CC: semantic-web at W3C <semantic-web@w3c.org>
Hi Adam

Thanks, I'm glad that the information is useful for you.

On 09/12/2010 16:13, adasal wrote:
> Boris,
> Those are a couple of incredibly useful links for me. Thank you!
> Especially your own link!
> I haven't had time to go through it yet, but I think I will find your 
> paper very useful, basically essential to what I am trying to understand.
>
> However, if you have a moment, can you say in one or two sentences the 
> relationship between your approach and google refine, which I belief 
> can lift XML, and TopBraid, which does lift XML, but, from my 
> experiments, requires post processing probably with SPIN, if I 
> understand correctly my own results. I should just add that SPIN 
> seemed like a solution too bound to the tool to be of interest to me, 
> and also represented further knowledge absorption on my part.
I'll try to do my best ...
AFAIU Google Refine is very good to make some data cleansing. The RDF 
extension of Google Refine allows to export, in a easiest way, the data 
coming from a non-ontological resource (CSV) to RDF instances. We use 
the term Population for this.

Our approach tries to cover the following transformation approaches:
- TBox transformation, i.e, transforming the resource content into an 
ontology schema
- ABox transformation, i.e., transforming the resource schema into an 
ontology schema, and the resource content, into ontology instances
- Population, i.e., transforming the resource content into instances of 
an existing ontology
This image can be useful: http://mccarthy.dia.fi.upm.es/img/TApproaches.png

Also, we want to cover databases, xml, spreadsheets and flat files, and 
to this end we are implementing a prototype, NOR2O. However, we still 
need to improve several things.

If you want we can discuss about your data and the post-processing stuff 
you are doing with SPIN, and this can be a new feature to take into 
account for us.

I didn't want to write a long email, so if you have doubts/suggestions 
do not hesitate to contact us.

Best and thanks again

Boris



>
> As I say comments welcome.
>
>
> Adam
>
> On 8 December 2010 22:11, Boris Villazón Terrazas <bvillazon@fi.upm.es 
> <mailto:bvillazon@fi.upm.es>> wrote:
>
>     Hi Paola
>
>     You can also try
>     XLWrap:http://xlwrap.sourceforge.net/
>     NOR2O:http://mccarthy.dia.fi.upm.es/nor2o/
>
>     Your suggestion about online conversion services is interesting,
>     maybe something like http://any23.org/
>
>     Best
>
>     Boris
>
>
>
>     On 08/12/2010 23:02, Paola Di Maio wrote:
>>     Bernard
>>
>>     thanks, looks like what I was hoping for
>>     :-)
>>
>>     I must have missed it , and no, not yet played with refine (cant
>>     keep up with all the developments it seems)
>>
>>     Zach - thanks for reply
>>
>>     It would be nice to have something (more than one service even)
>>     web based that does not require download/install to carry out
>>     conversion
>>
>>     online conversion services exist (pdf to doc etc) but not sure if
>>     the functionality comparable
>>
>>     I think
>>
>>     1.allowing users to specify minimal input as to what
>>     properties/values to model, rather than a blind conversion of all
>>     the fields
>>
>>     and
>>
>>      2. outputting the converted rdf file to URI
>>
>>
>>      would be  desirable features
>>
>>     Look forward to developments
>>
>>     cheers
>>
>>
>>     cheers
>>
>>     PDM
>>
>>
>>     On Wed, Dec 8, 2010 at 9:47 PM, Bernard Vatant
>>     <bernard.vatant@mondeca.com <mailto:bernard.vatant@mondeca.com>>
>>     wrote:
>>
>>         Hello Paola
>>
>>         Did you try the RDF extension for Google Refine?
>>         http://lab.linkeddata.deri.ie/2010/grefine-rdf-extension/
>>
>>         Cheers
>>
>>         Bernard
>>
>>         2010/12/8 Paola Di Maio <paola.dimaio@gmail.com
>>         <mailto:paola.dimaio@gmail.com>>
>>
>>             Greets folks
>>
>>             I am glad to read that there is more than one way to
>>             convert csv to RDF
>>             http://www.snee.com/bobdc.blog/2010/08/converting-csv-to-rdf.html
>>
>>             wonder if there is any public web service that allows me
>>             to upload a csv file
>>             (or point to a url of a csv file) , specify what
>>             values/properties I want to model in RDF,
>>             and outputs some valid RDF, ideally directly to a URI
>>             (the system would automagically create the resource)
>>
>>             If not, would there be scope for building such a tool?
>>
>>             (note, I dont write code myself otherwise I would do it,
>>             but surely there must exist coding wizards out there who
>>             could
>>             whip such a functionality up, provided its feasible and
>>             useful to others as well)
>>
>>             cheers
>>
>>             PDM
>>
>>
>>
>>
>>
>>         -- 
>>         Bernard Vatant
>>         Senior Consultant
>>         Vocabulary & Data Engineering
>>         Tel:       +33 (0) 971 488 459
>>         Mail: bernard.vatant@mondeca.com
>>         <mailto:bernard.vatant@mondeca.com>
>>         ----------------------------------------------------
>>         Mondeca
>>         3, cité Nollez 75018 Paris France
>>         Web: http://www.mondeca.com
>>         Blog: http://mondeca.wordpress.com
>>         ----------------------------------------------------
>>
>>
>
>
Received on Friday, 10 December 2010 01:47:28 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 21:45:40 GMT