ACL RD-TEC 1.0 Summarization of P06-2014
Paper Title:
SOFT SYNTACTIC CONSTRAINTS FOR WORD ALIGNMENT THROUGH DISCRIMINATIVE TRAINING
SOFT SYNTACTIC CONSTRAINTS FOR WORD ALIGNMENT THROUGH DISCRIMINATIVE TRAINING
Authors: Colin Cherry and Dekang Lin
Primarily assigned technology terms:
- algorithm
- alignment evaluation
- alignment method
- alignment system
- approximation
- beam search
- beam search solution
- bilingual word alignment
- bracketing
- chart parser
- classification
- computational linguistics
- constraint generation
- decomposition
- discriminative learning
- discriminative training
- discriminative word alignment
- dynamic programming
- dynamic programming algorithm
- error reduction
- feature representation
- giza
- greedy search
- iterative training
- itg alignment
- itg parse
- itg parser
- learner
- learning
- learning framework
- learning mechanism
- learning method
- learning task
- linking
- loss function
- matching
- matching algorithm
- max-margin alignment
- maximum matching
- minipar
- modeling
- monitoring
- morphology
- numerical optimization
- optimization
- parser
- parsing
- parsing algorithm
- perceptron
- programming algorithm
- scoring
- search
- search method
- statistical translation
- structure search
- supervised learning
- support vector machine
- svm learning
- synchronous parsing
- syntactic word alignment
- training process
- transduction
- translation systems
- tree construction
- tree-to-string alignment
- word alignment
Other assigned terms:
- alignment error rate
- alignment models
- approach
- association for computational linguistics
- beam
- bias
- bitext
- case
- classification problem
- co-occurrence
- coefficient
- cohesion
- correlation
- data set
- dependency tree
- dependency trees
- development set
- dice
- dice coefficient
- discriminative model
- english dependency
- english sentence
- error rate
- evaluation metrics
- feature
- feature vector
- feature vectors
- formalism
- frequency counts
- generation
- generative model
- generative models
- gold standard
- grammar
- hard constraint
- ibm model
- ibm models
- index
- inversion transduction grammar
- knowledge
- language pair
- leaf
- linguistic
- linguistics
- log-linear model
- measure
- method
- optimization problem
- parallel sentence
- paraphrases
- parse
- parse tree
- phrase
- polynomial time
- precision
- probabilities
- probability
- process
- production rule
- production rules
- projection
- relative error reduction
- search space
- seed
- sentence
- sentence pair
- sentences
- statistical models
- statistics
- subtree
- subtrees
- support vector
- svms
- symbol
- symbols
- syntactic constraint
- syntactic constraints
- syntactic features
- syntax
- target word
- technology
- term
- terms
- tokens
- training
- training example
- training set
- training time
- transduction grammar
- translation models
- tree
- tree structure
- trees
- weight vector
- word
- word frequency
- words
