ACL RD-TEC 1.0 Summarization of P01-1003
Paper Title:
IMPROVEMENT OF A WHOLE SENTENCE MAXIMUM ENTROPY LANGUAGE MODEL USING GRAMMATICAL FEATURES
IMPROVEMENT OF A WHOLE SENTENCE MAXIMUM ENTROPY LANGUAGE MODEL USING GRAMMATICAL FEATURES
Authors: Fredy A. Amaya and José Miguel BenedÃ
Primarily assigned technology terms:
- algorithm
- approximation
- automatic translation
- bracketing
- character recognition
- computational linguistics
- computing
- coupling
- information retrieval
- iterative scaling
- language modeling
- learning
- learning procedure
- linear interpolation
- maximum entropy
- maximum entropy model
- model training
- modeling
- normalization
- optical character recognition
- parameter estimation
- parameter learning
- parser
- recognition
- sampling
- smoothing
- speech recognition
- statistical language modeling
- statistical sampling
- training procedure
- training process
- tuning
Other assigned terms:
- bias
- case
- chomsky normal form
- concept
- conditional probability
- context-free grammar
- context-free grammars
- convergence
- derivation
- derivation tree
- derivation trees
- derivations
- distribution
- entropy
- estimation
- events
- experimental results
- fact
- feature
- gaussian prior
- grammar
- grammar rules
- grammars
- grammatical features
- grammatical framework
- grammatical information
- grammatical structure
- implementation
- interpolation
- labeling
- language model
- language modeling toolkit
- language models
- linguistic
- linguistics
- log-likelihood
- markov chain
- maximum entropy principle
- method
- model parameters
- modeling toolkit
- mutual information
- n-gram
- n-gram model
- n-gram models
- n-grams
- normal form
- parse
- parse tree
- part-of-speech
- part-of-speech tag
- penn treebank
- penn treebank corpus
- perplexity
- prior distribution
- prior probability
- probabilistic model
- probabilities
- probability
- probability distribution
- probability distributions
- procedure
- process
- production rules
- random sample
- relation
- relative frequency
- sentence
- sentences
- statistics
- stochastic context-free grammar
- stochastic context-free grammars
- structure of the sentence
- symbol
- symbols
- technique
- terminals
- test set
- time complexity
- toolkit
- training
- training corpus
- training data
- training set
- transition probabilities
- tree
- treebank
- treebank corpus
- trees
- trigram
- trigram model
- vocabulary
- word
- word distribution
- word sequences
- words
- wsj corpus
