Cypher 1.9 Demo

Hi All,

I'm pleased to announce the Cypher 1.9 demo available now
This announcement proceeds the official release announcement to come in the
next few days containing feature details. The demo runs the Monrai Standard
dataset, and can be pointed against several SPARQL Endpoints, including

Cypher is a natural language transcoder which generates the RDF triple or
SPARQL query representation of phrases and sentences.

Some example outputs are:

-  the person who played in a film (in

-  the writers who played in a film (in

-  the actors wrote what (in

-  the actors who are married wrote what (in

-  the presidents who were married to someone wrote what (in
-  the people who wrote some books (in

-  Bruce Willis stars in what (in

-  the people who were influenced by actors (in

Click the links in the (typically first) result set to view the raw
generated SPARQL, and the second icon to see the result set (having problems
with the vsual browser, should be fixed tommorrow). The biggest remaining
bottle-neck is the lexicon. For each parsed input, there must exists a
lexical entry which is used to construct the triples. For example, the
lexemes actor, writer, and star all specify RDF templates for dbp:starring
property. A more robust lexicon for Dbpedia (via UMBEL
<>and Cyc) is underway. In addition, there is a
collaborative-database of
lexical data that is in the works to help aleviate this shortcoming and
cover more ontologies and public datasets.

Please join the Cypher Mailing <> list to
keep up with release updates, etc.


Received on Thursday, 11 September 2008 19:46:10 UTC