Are there any tools available for generating RDF from natural language? A list of RDFizers compiled by the SIMILE project only mentions one, the Monrai Cypher. Unfortunately, it seems to have been a proprietary tool developed by Monrai Technologies, which has since disappeared, and I can't find any download links. Has anyone seen anything similar?
+1
A:
you could use the python nltk to parse the text and emit the rdf tripplets
Nikolaus Gradwohl
2010-07-26 17:24:34
+1 to above, also lookup OpenNLP.
Mikos
2010-07-28 13:18:37
+1
A:
You might look into OpenCalias, Zemanta and Hakia which all have nice APIs for extracting semantic data out of internet resources. Not familiar with Monrai Cypher, but possibly these might help.
Michael
2010-07-26 17:28:10
These look interesting, although it's unfortunate they all appear to be proprietary (albeit free) web services, and not downloadable code.
Chris S
2010-07-26 17:47:27
These generally tend to not be open-source because they heavily rely on named entity recognition to determine the subject and the object of the triple and that knowledge base takes a lot of expert man hours if you want it to be good.
Thien
2010-07-26 19:42:28
+1
A:
You want some ontology learning and population tools.
This online article lists 4 different systems:
- Text2Onto,
- Abraxas,
- KnowItAll,
- OntoLearn
You may want to check out the book; it reviews several ontology learning tools as well: Ontology learning from text: methods, evaluation and applications, by Paul Buitelaar, Philipp Cimiano, Bernardo Magnini
William
2010-07-28 06:07:46