ACL RD-TEC 1.0 Summarization of C04-1041
Paper Title:
THE IMPORTANCE OF SUPERTAGGING FOR WIDE-COVERAGE CCG PARSING
THE IMPORTANCE OF SUPERTAGGING FOR WIDE-COVERAGE CCG PARSING
Authors: Stephen Clark and James R. Curran
Primarily assigned technology terms:
- algorithm
- beam search
- ccg parser
- ccg parsing
- ccg supertagging
- chart parsing
- chart parsing algorithm
- cky chart parsing
- collins parser
- computing
- cutoff
- deep parsing
- dependency recovery
- discriminative training
- estimation algorithm
- estimation method
- estimation process
- forward-backward algorithm
- generalised iterative scaling
- iterative algorithm
- iterative scaling
- knowledge acquisition
- lexical knowledge acquisition
- markov model
- maximum entropy
- maximum entropy approach
- maximum entropy model
- model estimation
- nlp
- numerical optimisation
- optimisation
- parser
- parsing
- parsing algorithm
- parsing system
- processing
- pruning
- reading
- sampling
- scoring
- search
- smoothing
- statistical tagging
- supertagging
- tagger
- tagging
- viterbi
- viterbi algorithm
Other assigned terms:
- approach
- beam
- ccgbank
- cluster
- composition
- conditional probabilities
- dependency relations
- dependency structure
- dependency structures
- derivation
- derivations
- dictionary
- disk
- dutch
- entropy
- entropy models
- estimation
- extraction process
- f-score
- feature
- feature set
- formalisms
- gaussian prior
- gold standard
- grammar
- grammar formalisms
- grammars
- implementation
- knowledge
- lexical categories
- lexical category
- lexical head
- lexical knowledge
- lexicon
- log-likelihood
- log-linear model
- ltags
- maximum entropy models
- method
- nlp tasks
- oracle
- parse
- parsing accuracy
- parsing model
- parsing models
- parsing strategy
- penn treebank
- pos tag
- probabilities
- probability
- probability category
- process
- sentence
- sentences
- statistical model
- statistics
- supertag
- tagging model
- tags
- target word
- term
- terms
- test set
- text
- training
- training data
- training set
- tree
- treebank
- trees
- word
- words
