ACL RD-TEC 1.0 Summarization of P98-2140
Paper Title:
FEATURE LATTICES FOR MAXIMUM ENTROPY MODELLING
FEATURE LATTICES FOR MAXIMUM ENTROPY MODELLING
Primarily assigned technology terms:
- algorithm
- boundary disambiguation
- capitalization
- conditional maximum entropy
- constraint removal
- cross-validation
- decomposition
- disambiguation
- entropy modelling
- feature combination
- feature induction
- feature ranking
- feature selection
- feature selection algorithm
- feature selection process
- frequency redistribution
- induction
- induction algorithm
- iterative scaling
- language engineering
- language modelling
- lattice construction
- linear interpolation
- maximum entropy
- maximum entropy framework
- maximum entropy model
- model building
- modelling
- neural network
- nlp
- parameter estimation
- parsing
- part-of-speech tagging
- processing
- processor
- ranking
- search
- selection algorithm
- selection process
- sentence boundary disambiguation
- smoothing
- speech generation
- statistical language modelling
- stress prediction
- tagging
- text processing
- tokenisation
Other assigned terms:
- abbreviation
- abbreviations
- approach
- case
- collocate
- collocation
- community
- corpora
- distribution
- entropy
- entropy models
- error rate
- estimation
- events
- fact
- feature
- feature set
- feature sets
- feature space
- feature weights
- frequency counts
- frequency distribution
- generation
- interpolation
- interpretation
- knowledge
- kullback-leibler divergence
- language model
- lattice
- lattices
- lexicon
- likelihood
- log-likelihood
- maximum entropy models
- maximum entropy principle
- method
- model fit
- names
- nlp community
- parallel corpora
- part-of-speech
- parts-of-speech
- probabilities
- probability
- probability distribution
- procedure
- process
- processing tasks
- proper names
- proper noun
- relation
- sentence
- sentence boundary
- sentences
- statistics
- stress
- tagging model
- test data
- text
- training
- training phase
- training samples
- trigram
- trigram model
- uniform distribution
- uniform probability
- word
- words
- wsj corpus