ACL RD-TEC 1.0 Summarization of H05-1050
Paper Title:
BOOTSTRAPPING WITHOUT THE BOOT
BOOTSTRAPPING WITHOUT THE BOOT
Authors: Jason Eisner and Damianos Karakos
Primarily assigned technology terms:
- active learning
- algorithm
- binary classifier
- bootstrap
- bootstrapping
- bootstrapping method
- categorization
- classification
- classifier
- classifier combination
- classifiers
- co-training
- computational linguistics
- computer science
- disambiguation
- estimator
- genetic algorithms
- grammar induction
- human language
- human language technology
- induction
- inside-outside algorithm
- iterative bootstrapping
- kernel
- language processing
- language technology
- learner
- learning
- learning process
- linear kernel
- linear regression
- machine translation
- modeling
- mt system
- naive bayes
- naive bayes classifiers
- natural language processing
- page classification
- parser
- processing
- query expansion
- ranking
- reasoning
- regression
- regularization
- sampling
- searching
- sense disambiguation
- single classifier
- text categorization
- tf\/idf
- tuning
- unsupervised disambiguation
- unsupervised learning
- unsupervised technique
- unsupervised text categorization
- validation
- web page classification
- word sense disambiguation
- word-sense disambiguation
- yarowsky algorithm
Other assigned terms:
- ambiguous words
- annotation
- approach
- association for computational linguistics
- baseline performance
- bias
- bilingual dictionaries
- bilingual text
- case
- coefficient
- content words
- contextual feature
- contextual features
- correlation
- cosine measure
- determiners
- development set
- dictionaries
- disambiguation system
- document
- english sentence
- entropy
- evaluation function
- feature
- french
- french sentence
- function words
- gaussian mixture
- gold standard
- grammar
- grammars
- heuristic
- homonym
- human annotation
- knowledge
- language model
- lexicon
- likelihood
- linguistic
- linguistics
- log-likelihood
- log-likelihood ratio
- measure
- method
- mutual information
- named-entity
- names
- natural language
- nouns
- passage
- perplexity
- phrase
- pointwise mutual information
- polarity
- precision
- probabilistic grammar
- probability
- process
- pronouns
- proper names
- query
- random sample
- regularization parameter
- relation
- seed
- seed words
- semantic
- semantic features
- sense distinction
- sentence
- sentences
- sparse data
- statistical significance
- statistics
- subjective nouns
- support vector
- target word
- technique
- technology
- term
- test data
- text
- tokens
- training
- training corpus
- training data
- training set
- translation lexicon
- translation model
- translations
- user
- web page
- web pages
- word
- word order
- word sense
- word senses
- word type
- word types
- words
