ACL RD-TEC 1.0 Summarization of W02-1005
Paper Title:
AUGMENTED MIXTURE MODELS FOR LEXICAL DISAMBIGUATION
AUGMENTED MIXTURE MODELS FOR LEXICAL DISAMBIGUATION
Authors: Silviu Cucerzan and David Yarowsky
Primarily assigned technology terms:
- 5-fold cross validation
- 5-fold cross-validation
- adaboost
- algorithm
- ambiguity resolution
- boosting
- boosting algorithm
- classification
- classification correction
- classifier
- classifier combination
- classifiers
- computational linguistics
- context-sensitive spelling
- context-sensitive spelling correction
- correction method
- cross validation
- cross-validation
- data representation
- disambiguation
- error-correction
- estimation method
- iterative boosting
- language classification
- language processing
- learner
- learning
- learning algorithm
- lexical ambiguity resolution
- lexical classification
- lexical disambiguation
- machine learning
- machine translation
- maximum likelihood
- modeling
- naive bayes
- named-entity classification
- natural language classification
- parameter estimation
- pos-tagging
- predictor
- probability estimation
- processing
- search
- sense disambiguation
- sense-disambiguation
- smoothing
- spelling
- spelling correction
- statistical methods
- topic classification
- transformation-based learning
- tribayes
- tuning
- validation
- voting
- voting system
- weak learner
- weighting
- word sense disambiguation
Other assigned terms:
- adjective
- ambiguity
- analogy
- approach
- association for computational linguistics
- basque
- bayes decision rule
- bayes model
- bayesian model
- binary features
- binomial distribution
- brown corpus
- case
- classification task
- classification tasks
- co-occurrences
- coefficient
- comparative study
- conditional probabilities
- conditional probability
- context features
- contextual features
- correlation
- data sets
- decision rule
- dictionary
- dimensionality
- disambiguation task
- distribution
- document
- estimation
- events
- experimental results
- extraction process
- fact
- feature
- feature space
- feature type
- feature types
- heuristic
- implementation
- keyword
- knowledge
- lemma
- lexical ambiguity
- lexical choice
- likelihood
- linguistics
- measure
- method
- mixture models
- model parameters
- morphological variant
- naive bayes model
- named-entity
- natural language
- nouns
- part-of-speech
- part-ofspeech
- pos-class
- posterior
- posterior distribution
- posterior probability
- predicate-argument
- prior probability
- probabilities
- probability
- probability distribution
- probability distributions
- probability estimates
- procedure
- process
- regular expressions
- sense inventory
- senses of a word
- sentence
- sparse feature space
- statistical model
- statistics
- syntactic features
- syntactic functions
- syntactic patterns
- system evaluation
- system performance
- target word
- technique
- test data
- test set
- training
- training and test data
- training data
- training set
- transposition
- typographical errors
- verb
- word
- word sense
- words
