Inverted indexing for cross-lingual NLP

TY - GEN

T1 - Inverted indexing for cross-lingual NLP

AU - Søgaard, Anders

AU - Agic, Zeljko

AU - Martinez Alonso, Hector

AU - Plank, Barbara

AU - Bohnet, Bernd

PY - 2015

Y1 - 2015

N2 - We present a novel, count-based approach to obtaining inter-lingual word representations based on inverted indexing of Wikipedia. We present experiments applying these representations to 17 datasets in document classification, POS tagging, dependency parsing, and word alignment. Our approach has the advantage that it is simple, computationally efficient and almost parameter-free, and, more importantly, it enables multi-source crosslingual learning. In 14/17 cases, we improve over using state-of-The-Art bilingual embeddings.

AB - We present a novel, count-based approach to obtaining inter-lingual word representations based on inverted indexing of Wikipedia. We present experiments applying these representations to 17 datasets in document classification, POS tagging, dependency parsing, and word alignment. Our approach has the advantage that it is simple, computationally efficient and almost parameter-free, and, more importantly, it enables multi-source crosslingual learning. In 14/17 cases, we improve over using state-of-The-Art bilingual embeddings.

M3 - Article in proceedings

SN - 978-1-941643-72-3

VL - 1

SP - 1713

EP - 1722

BT - The 53rd Annual Meeting of the Association for Computational Linguistics (ACL)

PB - Association for Computational Linguistics

ER -