Skip to content
conftrace_
2017 EMNLP EMNLP 2017

Deciphering Related Languages

Abstract

AbstractWe present a method for translating texts between close language pairs. The method does not require parallel data, and it does not require the languages to be written in the same script. We show results for six language pairs: Afrikaans/Dutch, Bosnian/Serbian, Danish/Swedish, Macedonian/Bulgarian, Malaysian/Indonesian, and Polish/Belorussian. We report BLEU scores showing our method to outperform others that do not use parallel data.

🌉 Interdisciplinary Bridge - Artificial Intelligence and Natural Language Processing
🧭 Keyword Pioneer - unsupervised translation

Authors

Nima Pourdamghani, Kevin Knight

Topics

Machine Learning > Core Methods > Classification Natural Language Processing > Applications > Machine Translation Deep Learning > Learning Types > Unsupervised Learning

Keywords

machine translation unsupervised translation parallel data related language text translation close language pair
Download PDF

More from EMNLP 2017

All EMNLP 2017 papers →
A Challenge Set Approach to Evaluating Machine Translation
Knowledge Distillation for Bilingual Dictionary Induction
Machine Translation, it’s a question of style, innit? The case of English tag questions
Identifying Cognate Sets Across Dictionaries of Related Languages
Learning Language Representations for Typology Prediction
Cheap Translation for Cross-Lingual Named Entity Recognition