ACL RD-TEC 1.0 Summarization of W06-1607
Paper Title:
PHRASETABLE SMOOTHING FOR STATISTICAL MACHINE TRANSLATION
PHRASETABLE SMOOTHING FOR STATISTICAL MACHINE TRANSLATION
Authors: George Foster and Roland Kuhn and Howard Johnson
Primarily assigned technology terms:
- algorithm
- approximation
- black-box smoothing
- bootstrap
- bootstrap resampling
- chinese-english translation
- computational linguistics
- decoder
- decoding
- decomposition
- glass-box
- glass-box smoothing
- good-turing smoothing
- induction
- induction algorithm
- kneser-ney smoothing
- language modeling
- language processing
- learning
- lexical decomposition
- lexical smoothing
- linear interpolation
- loglinear
- loglinear interpolation
- machine translation
- modeling
- natural language processing
- nlp
- normalization
- phrase induction
- phrase table smoothing
- phrase translation
- phrase-based translation
- phrasetable smoothing
- processing
- resampling
- search
- smoothing
- smoothing distribution
- smoothing method
- smoothing problem
- smoothing technique
- smoothing techniques
- smt system
- statistical machine translation
- statistical mt
- statistical nlp
- translation systems
- tuning
- unigram smoothing
- viterbi
- witten-bell smoothing
- word-alignment
Other assigned terms:
- approach
- association for computational linguistics
- backoff
- bias
- bigram
- bleu
- bleu metric
- bleu score
- break
- case
- coefficient
- conditional probabilities
- conditional probability
- corpora
- corpus size
- data sparseness
- data sparsity
- distribution
- english translation
- estimation
- evaluations
- events
- experimental results
- fact
- feature
- french
- higher-order distribution
- ibm model
- ibm models
- implementation
- interpolation
- interpolation scheme
- language model
- language models
- language pairs
- large corpora
- large corpus
- large training
- large training corpora
- linguistics
- loglinear model
- measure
- method
- natural language
- ngram
- nist
- parallel corpora
- parallel corpus
- phrase
- phrase attachment
- phrase translation model
- prepositional phrase
- prepositional phrase attachment
- priori
- probabilities
- probability
- probability distribution
- recipe
- relative frequency
- reordering
- sentence
- sentences
- small training corpora
- source sentence
- target languages
- target sentence
- technique
- terms
- toolkit
- training
- training corpora
- training corpus
- training data
- training material
- translation model
- translation probabilities
- translation quality
- translations
- trigram
- trigram language model
- unigram
- word
- word alignments
- word sequences
- word-aligned corpus
- words
