ACL RD-TEC 1.0 Summarization of P05-1034
Paper Title:
DEPENDENCY TREELET TRANSLATION: SYNTACTICALLY INFORMED PHRASAL SMT
DEPENDENCY TREELET TRANSLATION: SYNTACTICALLY INFORMED PHRASAL SMT
Authors: Chris Quirk and Arul Menezes and Colin Cherry
Primarily assigned technology terms:
- algorithm
- alignment process
- decision tree
- decoder
- decoding
- dependency analysis
- dependency parser
- dynamic programming
- dynamic programming approach
- exhaustive search
- giza
- heuristic combination
- induction
- kneser-ney smoothing
- lambda training
- language analysis
- learning
- learning techniques
- left-to-right decoding
- likelihood estimate
- likelihood estimation
- machine learning
- machine learning techniques
- machine translation
- maximum likelihood
- maximum likelihood estimation
- modeling
- mt system
- optimization
- parallel parsing
- parameter tuning
- parser
- parsing
- phrasal translation
- phrase extraction
- pruning
- re-ranking
- scoring
- search
- segmentation
- sentence realization
- single reference translation
- smoothing
- smt system
- statistical machine translation
- supervised training
- terminology
- transduction
- translation process
- treelet phrase extraction
- treelet translation
- tuning
- unsupervised word alignment
- word alignment
- word alignment process
- word segmentation
Other assigned terms:
- adjective
- aligned sentence
- alignment problem
- annotation
- approach
- bigram
- bitext
- bleu
- bleu metric
- bleu score
- bleu scores
- cache
- candidate translation
- case
- characters
- cohesion
- community
- contextual information
- data sparsity
- dependency parse
- dependency structure
- dependency tree
- dependency trees
- distribution
- domain-specific terminology
- estimation
- feature
- feature vector
- feature vectors
- forest
- french
- grammar
- grammars
- heuristic
- heuristics
- hmm model
- ibm models
- input string
- inversion transduction grammars
- lambda
- language model
- language pair
- language pairs
- leaf
- lexical items
- likelihood
- linguistic
- linguistic structure
- logical form
- mapping
- mappings
- maximum likelihood estimate
- method
- mle channel model
- model combination
- modifier
- n-best list
- parallel corpus
- parse
- part-of-speech
- part-of-speech tags
- phrase
- preposition
- prepositional phrases
- probabilities
- probability
- probability distribution
- process
- programming approach
- projection
- reference translation
- reordering
- representations
- search space
- sentence
- sentence pair
- sentence realization component
- sentences
- set size
- slot
- source language
- source text
- statistical models
- subgraph
- subtree
- subtrees
- syntactic constituents
- syntactic information
- syntactic model
- syntax
- tags
- target language
- target language model
- target languages
- target sentence
- target string
- terms
- test set
- text
- training
- training corpus
- training data
- training set
- training set size
- translation candidate
- translation candidates
- translation model
- translation models
- translation pair
- translation pairs
- translation probabilities
- translation quality
- translations
- tree
- tree representation
- tree structure
- tree-like representation
- trees
- trigram
- trigram model
- uniform probability
- vocabulary
- vocabulary size
- wildcard
- word
- word alignments
- word count
- word sequences
- words
