ACL RD-TEC 1.0 Summarization of P01-1010
Paper Title:
WHAT IS THE MINIMAL SET OF FRAGMENTS THAT ACHIEVES MAXIMAL PARSE ACCURACY?
WHAT IS THE MINIMAL SET OF FRAGMENTS THAT ACHIEVES MAXIMAL PARSE ACCURACY?
Primarily assigned technology terms:
- algorithm
- atis
- boosting
- capitalization
- chart parsing
- cky algorithm
- computing
- data oriented parsing
- disambiguation
- encoding
- estimator
- frequency estimator
- hyphenation
- language parsing
- lexicalization
- likelihood estimation
- maximum likelihood
- maximum likelihood estimation
- monte carlo disambiguation
- n-best search
- natural language parsing
- parser
- parsers
- parsing
- pruning
- reporting
- reranking
- sampling
- search
- smoothing
- statistical parsing
- stochastic parsing
- subtree-torule conversion
- tagger
- viterbi
Other assigned terms:
- annotated corpus
- approach
- case
- co-reference
- context-free grammar
- context-free rule
- convergence
- conversion method
- corpora
- data sparseness
- dependency relation
- dependency relations
- derivation
- derivation forest
- derivations
- development set
- elementary tree
- estimation
- fact
- forest
- good-turing estimation
- grammar
- grammars
- heuristic
- knowledge
- large corpus
- lattice
- lexical context
- lexical head
- lexical information
- lexical item
- lexical items
- likelihood
- linguistic
- maximum subtree depth
- method
- natural language
- nonterminal
- parse
- parse tree
- parse tree probability
- parsing model
- parsing models
- part-ofspeech
- penn treebank
- precision
- prior probability
- priori
- probabilities
- probability
- procedure
- random sample
- relation
- relative frequency
- root node
- semantic
- semantic tags
- sentence
- sentences
- statistical natural language
- statistics
- stochastic context-free grammar
- subsumption
- subtree
- subtrees
- symbols
- syntactic structures
- tags
- technique
- test set
- training
- training set
- tree
- tree-insertion grammar
- treebank
- treebank grammar
- trees
- unknown word model
- verb
- word
- word model
- word strings
- words
- wsj corpus
