Tracing Mistakes and Finding Gaps in Automatic Word Alignments for Latvian-English Translation.
Valdis GirgzdisMaija KaleMartins VaicekauskisIeva ZarinaInguna SkadinaPublished in: Baltic HLT (2014)
Keyphrases
- sentence pairs
- statistical machine translation
- machine translation system
- machine translation
- parallel corpus
- english words
- query translation
- bilingual dictionaries
- cross language information retrieval
- english chinese
- source language
- parallel corpora
- target language
- word level
- translation model
- word alignment
- word pairs
- cross lingual
- labor intensive
- word sense disambiguation
- unknown words
- multiword
- chinese english
- named entity recognizer
- english text
- bilingual lexicon
- pronominal anaphora
- language model
- word sense
- co occurrence
- language specific
- cross language
- target word
- training corpus
- cross language retrieval
- keywords
- natural language
- pairwise
- sense disambiguation
- character n grams
- language learning
- lexical information
- word order
- relation extraction
- linguistic features
- english language
- semantic roles