ACL RD-TEC 1.0 Summarization of C04-1060
Paper Title:
SYNTAX-BASED ALIGNMENT: SUPERVISED OR UNSUPERVISED?
SYNTAX-BASED ALIGNMENT: SUPERVISED OR UNSUPERVISED?
Authors: Hao Zhang and Daniel Gildea
Primarily assigned technology terms:
- algorithm
- alignment evaluation
- automatic system
- binary branching
- bracketing
- capitalization
- chart parsing
- collapsing
- collins parser
- cross validation
- cross-validation
- decoding
- english parser
- expectation maximization
- expectation maximization algorithm
- giza
- inside-outside expectation maximization
- learning
- machine translation
- maximization algorithm
- measuring
- parallel training
- parameter estimation
- parser
- parsing
- pre-processing
- pruning
- re-estimation
- sampling
- scoring
- search
- searching
- statistical approaches
- statistical machine translation
- statistical parser
- syntax-based alignment
- training procedure
- transduction
- translation process
- tree-based alignment
- tree-to-string alignment
- unification
- validation
- verb phrases
- viterbi
Other assigned terms:
- adverb
- alignment error rate
- alignment model
- alignment models
- alignment task
- annotation
- annotators
- approach
- bilingual sentence
- bilingual text
- binary branching trees
- branching trees
- break
- case
- chinese treebank
- chinese words
- chinese-english newswire text
- community
- computational complexity
- context-free grammar
- corpora
- development set
- distribution
- english sentence
- english translations
- english tree
- english vocabulary
- error rate
- estimation
- evaluations
- fact
- french
- french sentence
- french translation
- french word
- generative model
- gold standard
- grammar
- grammar rule
- grammar rules
- grammars
- hansard corpus
- head-word
- hmm model
- human annotators
- hypothesis
- ibm model
- ibm models
- implementation
- inversion transduction grammar
- language pair
- language pairs
- large corpora
- lexical translation
- likelihood
- main verb
- meaning
- measure
- measures
- modifier
- nonterminal
- nonterminals
- noun phrases
- pairs of words
- parallel corpora
- parallel corpus
- parallel training corpus
- parse
- parse tree
- parser output
- penn chinese treebank
- perplexity
- polynomial time
- precision
- probabilistic model
- probabilities
- probability
- probability model
- procedure
- process
- punctuation
- recursion
- reordering
- sentence
- sentence pair
- sentences
- source language
- stochastic context-free grammar
- style
- subtree
- supervised model
- symbol
- symbols
- syntactic categories
- syntactic category
- syntactic constituents
- syntactic information
- syntactic parse
- syntactic parse tree
- syntactic structure
- syntactic tree
- target language
- target language string
- technique
- term
- terminals
- terms
- test data
- test set
- text
- tokens
- training
- training corpus
- training data
- transduction grammar
- transformation
- translation model
- translation models
- translation probabilities
- translation rules
- translations
- tree
- tree representation
- tree structure
- tree structures
- treebank
- trees
- understanding
- uniform probability
- unsupervised alignment model
- verb
- vocabulary
- word
- word order
- words
- wsj treebank
