ACL RD-TEC 1.0 Summarization of W06-2903
Paper Title:
NON-LOCAL MODELING WITH A MIXTURE OF PCFGS
NON-LOCAL MODELING WITH A MIXTURE OF PCFGS
Authors: Slav Petrov and Leon Barrett and Dan Klein
Primarily assigned technology terms:
- add-one smoothing
- algorithm
- categorization
- chart parser
- computational linguistics
- computational natural language learning
- dynamic programming
- dynamic programming algorithm
- em algorithm
- em training
- estimator
- expectation-maximization
- human language
- human language generation
- java
- language generation
- language learning
- learning
- machine learning
- maximum-likelihood
- modeling
- natural language learning
- parser
- parsers
- parsing
- pcfg parser
- pcfg parsing
- programming algorithm
- smoothing
- splitting
- statistical estimation
- statistical parsers
- tagging
- text categorization
- tuning
- validation
- vertical markovization
Other assigned terms:
- annotation
- approach
- association for computational linguistics
- break
- brown corpus
- case
- coefficient
- conditional probabilities
- conditional probability
- conll-x
- context-free grammar
- convergence
- corpora
- correlation
- correlations
- data log likelihood
- derivation
- derivation process
- derivations
- distribution
- estimation
- fact
- formalism
- formalisms
- generation
- generative model
- genre
- grammar
- grammar formalisms
- grammar rules
- grammars
- hierarchical model
- implementation
- index
- joint distribution
- knowledge
- language corpora
- likelihood
- likelihood ratio
- linguistics
- local maximum
- log-likelihood
- measure
- measures
- mixture models
- model parameters
- natural language
- natural language corpora
- nonterminal
- nonterminals
- parse
- parse tree
- parsing accuracy
- pcfg
- pcfg model
- pcfgs
- penn treebank
- phrase
- phrase structure
- posterior
- probabilities
- probability
- process
- quantifier
- relation
- sbar
- sentence
- sentences
- single-grammar model
- statistics
- stem
- symbols
- syntactic structure
- test data
- testing data
- text
- training
- training data
- training set
- tree
- tree node
- treebank
- treebank grammar
- trees
- word
- words