Skip to Main content Skip to Navigation
Conference papers

The KBGen Challenge

Eva Banik 1 Claire Gardent 2 Eric Kow 1 
2 SYNALP - Natural Language Processing : representations, inference and semantics
LORIA - NLPKD - Department of Natural Language Processing & Knowledge Discovery
Abstract : Given a preselected set of relations extracted from the AURA knowledge base on biology, the KBGEN Task consisted in generating a sentence verbalising these relations. Three team submitted the results of their systems. The systems were compared using both automatic metrics (BLEU, NIST) and subjective ratings by 12 human users for three dimensions namely, fluency, grammaticality and meaning similarity. In this report, we summarise the KBGen Task, the evaluation methods and the results obtained.
Document type :
Conference papers
Complete list of metadata

Cited literature [3 references]  Display  Hide  Download
Contributor : Claire Gardent Connect in order to contact the contributor
Submitted on : Wednesday, December 18, 2013 - 7:04:11 PM
Last modification on : Saturday, October 16, 2021 - 11:26:06 AM
Long-term archiving on: : Thursday, March 20, 2014 - 11:26:32 AM


Files produced by the author(s)


  • HAL Id : hal-00920605, version 1



Eva Banik, Claire Gardent, Eric Kow. The KBGen Challenge. the 14th European Workshop on Natural Language Generation (ENLG), Aug 2013, Sofia, Bulgaria. pp.94-97. ⟨hal-00920605⟩



Record views


Files downloads