On Nov 28, 2011, at 09:58 , Rupert Westenthaler wrote:

> Hi Lukas
> 
> On 27.11.2011, at 19:01, Lukas Kahwe Smith wrote:
> 
>> Aloha,
>> 
>> My name is Lukas Smith.
>> I work for Liip in Switzerland. I have already met some of you at the IKS 
>> workshop in Paris. Liip also was fortunate enough to win the IKS UX contest 
>> and we are working hard on VIE and hallo integration with the Symfony2 PHP 
>> framework as well as porting JCR to PHP.
>> 
> 
> I have read the proposal and really liked it. Congratulations!

thx!

>> At any rate, I joined this list because I wanted to keep tabs on Stanbol in 
>> general, but more specifically since I want to add automated tagging of 
>> content to the resolutionfinder.org admin interface. This platform aims to 
>> provide a centralized database of UN resolutions. The goal is to annotate 
>> the resolutions with additional data, as well as providing different ways to 
>> examine the data than just one document at a time.
> 
> I think you presented this initiative during the Semantic / NLP Hackathon of 
> the Berlin Buzzwords.

yes, i did

>> The challenge will be in dealing with our very specific domain language 
>> (legal documents) and mapping them to our closed list of tags. I guess I 
>> will need to look at the following doc page and see how far I get with that:
>> http://incubator.apache.org/stanbol/docs/trunk/customvocabulary.html
>> 
> Yes this is the right place to start from.
> 
> 
> May I askt two things:
> 
> * Is there already some list of controlled terms (e.g. a thesaurus) you plan 
> to use? Plan you use the existing Tags as an starting point?

What I currently have is the result of 2 years of research, where a team of 
students identified a set of relevant tags (or better I should say phrases).
You can see some of these in the left hand column under "tags" with this sample 
search:
http://resolutionfinder.org/?q=UN&dc=&st=document&p=0

> * Do you need to deal with multiple languages or are all UN resources 
> available in English? 


Eventually we will, but right now the database only contains english and our 
list phrases also only contains english. So for the time being english only.

regards,
Lukas Kahwe Smith
[email protected]



Reply via email to