English

Embedding Learning Through Multilingual Concept Induction

Computation and Language 2018-06-28 v3

Abstract

We present a new method for estimating vector space representations of words: embedding learning by concept induction. We test this method on a highly parallel corpus and learn semantic representations of words in 1259 different languages in a single common space. An extensive experimental evaluation on crosslingual word similarity and sentiment analysis indicates that concept-based multilingual embedding learning performs better than previous approaches.

Keywords

Cite

@article{arxiv.1801.06807,
  title  = {Embedding Learning Through Multilingual Concept Induction},
  author = {Philipp Dufter and Mengjie Zhao and Martin Schmitt and Alexander Fraser and Hinrich Schütze},
  journal= {arXiv preprint arXiv:1801.06807},
  year   = {2018}
}

Comments

ACL 2018

R2 v1 2026-06-22T23:51:06.327Z