ACL RD-TEC 1.0 Summarization of N04-1039
Paper Title:
EXPONENTIAL PRIORS FOR MAXIMUM ENTROPY MODELS
EXPONENTIAL PRIORS FOR MAXIMUM ENTROPY MODELS
Primarily assigned technology terms:
- algorithm
- ambiguity resolution
- approximation
- audio processing
- bayesian classification
- classification
- computational linguistics
- computer speech
- conditional maximum entropy
- disambiguation
- disambiguation problem
- entropy estimation
- entropy training
- feature selection
- fuzzy maxent
- generalized iterative scaling
- good-turing smoothing
- grammar checking
- interpolated kneser-ney smoothing
- iterative scaling
- kneser-ney smoothing
- language ambiguity resolution
- language model smoothing
- language modeling
- language modelling
- language processing
- learning
- learning algorithm
- learning algorithms
- logistic regression
- m-gram language modeling
- machine learning
- machine translation
- matching
- maxent
- maximum entropy
- maximum entropy approach
- maximum entropy method
- maximum entropy model
- maximum likelihood
- model smoothing
- modeling
- modeling technique
- modelling
- natural language processing
- neural networks
- nlp
- parameter setting
- parameter-free smoothing
- parsing
- part-of-speech tagging
- pattern analysis
- processing
- pruning
- reading
- regression
- regularization
- search
- searching
- sense disambiguation
- smoothing
- smoothing technique
- smoothing techniques
- statistical language modeling
- statistical language modelling
- statistical modeling
- stochastic language modeling
- structuring
- supervised learning
- tagging
- word selection
- word sense disambiguation
Other assigned terms:
- ambiguity
- approach
- average error rate
- backoff
- bigram
- binary features
- case
- constraint satisfaction
- convergence
- data set
- data sets
- derivation
- distribution
- entropy
- entropy models
- error rate
- estimation
- events
- experimental results
- exponential distribution
- exponential model
- fact
- feature
- gaussian prior
- grammar
- heuristic
- heuristics
- histogram
- information theory
- intelligence
- kullback-leibler divergence
- language ambiguity
- language model
- likelihood
- linguistics
- log-linear models
- maximum entropy models
- measure
- measures
- method
- model parameters
- natural language
- natural language ambiguity
- noise
- parameter values
- part-of-speech
- perplexity
- phrase
- phrase attachment
- posterior
- posterior probability
- prepositional phrase
- prepositional phrase attachment
- prior distribution
- probabilities
- probability
- process
- selection operator
- sense disambiguation problem
- sentence
- sentence boundaries
- sigmoid function
- standard deviation
- statistics
- style
- svms
- technique
- television
- terms
- test data
- test set
- text
- theory
- trained model
- training
- training data
- training examples
- trigram
- trigram model
- uniform distribution
- unigram
- word
- word sense
- words
- wsj corpus
