Re: How do you explore a SPARQL Endpoint?

On 2015-01-22 15:09, Juan Sequeda wrote:
> Assume you are given a URL for a SPARQL endpoint. You have no idea what
> data is being exposed.
>
> What do you do to explore that endpoint? What queries do you write?
>
> Juan Sequeda
> +1-575-SEQ-UEDA
> www.juansequeda.com
>

I suspect that the obligatory query that everyone is dying to know is to 
get a distinct count of subjects. /me mumbles..

More realistically:

* I would say that getting a sense of which vocabularies/ontologies are 
used is a good way to dive ib, and to come up with more 
specific/useful/interesting queries thereafter.

* Look up and see which VoID information is available. Related to above 
point, e.g., void:vocabulary.

* Check whether there is sufficient human-readable labels for the 
significant portion of the instances.

* Check triples pertaining to provenance.

* Check if there are sufficient interlinks to resources that presumably 
external to the domain in which the endpoint is at.


Sorry, I'm not going to write out SPARQL queries here. Need to preserve 
brain-cells for the remainder of the day.

-Sarven
http://csarven.ca/#i

Received on Thursday, 22 January 2015 14:27:02 UTC