You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@jena.apache.org by Howard Burrows <ho...@gmail.com> on 2013/08/07 16:53:41 UTC

Text search indexing of inference models

Is it possible to configure (assemble) the text search module such that the textindexer runs against an inference model and can index the content of inferred statements?

I use the owl:equivalentProperty and owl:equivilantClass to normalize my data into a standard and consistent vocabulary that I would then like to index using the text search indexer.  However, the textindexer appears to only see asserted statements and not inferred statements.  Am I missing something or maybe have it configured incorrectly or does the textindexer only see asserted statements.

Any advice on how to accomplish this would be welcomed and appreciated.

Thanks,

Howard Burrows 

Re: Text search indexing of inference models

Posted by Andy Seaborne <an...@apache.org>.
On 07/08/13 15:53, Howard Burrows wrote:
> Is it possible to configure (assemble) the text search module such
> that the textindexer runs against an inference model and can index
> the content of inferred statements?
>
> I use the owl:equivalentProperty and owl:equivilantClass to normalize
> my data into a standard and consistent vocabulary that I would then
> like to index using the text search indexer.  However, the
> textindexer appears to only see asserted statements and not inferred
> statements.  Am I missing something or maybe have it configured
> incorrectly or does the textindexer only see asserted statements.
>
> Any advice on how to accomplish this would be welcomed and
> appreciated.
>
> Thanks,
>
> Howard Burrows
>

The text indexer simply runs through all the properties for the index so 
if the source is an inference model in a dataset (the target of 
text:dataset in the assembler) it will work.

Look at the code of textindexer:

https://svn.apache.org/repos/asf/jena/trunk/jena-text/src/main/java/jena/textindexer.java

the heart is:

loop on
dataset.find(Node.ANY, Node.ANY, property, Node.ANY) ;

for the indexed propert(y|ies).

	Andy