T-Rex : A Large Scale Alignment of Natural Language with Knowledge Base Triples

With 11 million triple alignments made from 3.09 million DBpedia abstracts (6.2 million sentences), T-REx is two orders of magnitude larger than the largest available alignments to the community.

Download Data Package

Data and Resources

T-REx NIF ttl Sample ttl
Sample annnotation of one document and it's equivalent aligned triples...

More information Go to resource
T-REx JSON Sample JSON
Sample of 1 document and its alignments in T-REx

More information Go to resource

Additional Info

Field	Value
Source	w3id.org/t-rex/
Author	Hady Elsahar
Maintainer	hady elsahar
Version	1.0
Last Updated	May 15, 2017, 16:11 (UTC)
Created	May 15, 2017, 15:52 (UTC)
github	https://github.com/hadyelsahar/t-rex
license	https://creativecommons.org/licenses/by-sa/4.0/