Improving Neural Knowledge Base Completion with Cross-Lingual Projections

Patrick Klein, Simone Paolo Ponzetto, Goran Glavaš


Abstract
In this paper we present a cross-lingual extension of a neural tensor network model for knowledge base completion. We exploit multilingual synsets from BabelNet to translate English triples to other languages and then augment the reference knowledge base with cross-lingual triples. We project monolingual embeddings of different languages to a shared multilingual space and use them for network initialization (i.e., as initial concept embeddings). We then train the network with triples from the cross-lingually augmented knowledge base. Results on WordNet link prediction show that leveraging cross-lingual information yields significant gains over exploiting only monolingual triples.
Anthology ID:
E17-2083
Volume:
Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics: Volume 2, Short Papers
Month:
April
Year:
2017
Address:
Valencia, Spain
Editors:
Mirella Lapata, Phil Blunsom, Alexander Koller
Venue:
EACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
516–522
Language:
URL:
https://aclanthology.org/E17-2083
DOI:
Bibkey:
Cite (ACL):
Patrick Klein, Simone Paolo Ponzetto, and Goran Glavaš. 2017. Improving Neural Knowledge Base Completion with Cross-Lingual Projections. In Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics: Volume 2, Short Papers, pages 516–522, Valencia, Spain. Association for Computational Linguistics.
Cite (Informal):
Improving Neural Knowledge Base Completion with Cross-Lingual Projections (Klein et al., EACL 2017)
Copy Citation:
PDF:
https://aclanthology.org/E17-2083.pdf