A Multi-Pairwise Extension of Procrustes Analysis for Multilingual Word Translation.
Hagai TaitelbaumGal ChechikJacob GoldbergerPublished in: EMNLP/IJCNLP (1) (2019)
Keyphrases
- pairwise
- procrustes analysis
- machine translation system
- parallel corpus
- cross language information retrieval
- bilingual dictionaries
- machine translation
- translation model
- cross lingual
- statistical machine translation
- language specific
- chinese english
- language resources
- out of vocabulary
- cross language ir
- english chinese
- query translation
- n gram
- cross language
- english words
- cross lingual information retrieval
- comparable corpora
- language independent
- multi class
- semi supervised
- co occurrence
- bilingual lexicon
- lexical knowledge
- word alignment
- markov random field
- loss function
- keywords
- higher order
- wordnet
- word level
- source language
- digital libraries
- character n grams
- semantic similarity
- natural language generation
- similarity measure