Unsupervised Training for Large Vocabulary Translation Using Sparse Lexicon and Word Classes.
Yunsu KimJulian SchamperHermann NeyPublished in: EACL (2) (2017)
Keyphrases
- supervised learning
- lexical semantics
- translation model
- n gram
- training corpus
- handwritten words
- statistical machine translation
- machine translation
- machine readable dictionaries
- speech recognition systems
- unsupervised learning
- machine translation system
- bilingual lexicon
- supervised training
- bilingual dictionaries
- training set
- speech recognition
- linguistic knowledge
- neural network
- semi supervised
- query translation
- word sense disambiguation
- english words
- deep architectures
- domain specific
- probabilistic model
- high dimensional
- speech recognizer
- word alignment
- parallel corpus
- sentence level
- pointwise mutual information
- supervised methods
- parallel corpora
- sparse representation
- training examples
- training samples
- language model
- co occurrence
- multi class
- feature selection