ACL RD-TEC 1.0 Summarization of P04-1039
Paper Title:
RELIEVING THE DATA ACQUISITION BOTTLENECK IN WORD SENSE DISAMBIGUATION
RELIEVING THE DATA ACQUISITION BOTTLENECK IN WORD SENSE DISAMBIGUATION
Primarily assigned technology terms:
- algorithm
- automatic tagging
- bootstrap
- bootstrapping
- bootstrapping approach
- classification
- classifiers
- comparable bootstrapping
- data acquisition
- dependency parser
- disambiguation
- generation algorithm
- grouping
- information retrieval
- iterative algorithm
- learner
- learning
- learning algorithm
- learning approach
- learning framework
- learning methods
- learning system
- likelihood estimate
- machine learning
- machine learning algorithm
- maximum likelihood
- minipar
- parser
- querying
- salaam tagging
- semantic translation
- sense annotation
- sense disambiguation
- sense selection
- sense tagging
- supervised learning
- supervised learning framework
- supervised learning system
- supervised method
- supervised word sense disambiguation
- support vector machines
- tagging
- unsupervised approach
- unsupervised bootstrapping
- unsupervised tagging
- weighting
- word sense disambiguation
- wsd algorithm
Other assigned terms:
- aligned parallel corpus
- ambiguity
- ambiguous words
- annotation
- approach
- brown corpus
- case
- classification problem
- cluster
- clusters
- collocate
- compounds
- context features
- contextual features
- corpora
- correlation
- data corpus
- data set
- dictionary
- distribution
- document
- document frequency
- english corpus
- entropy
- fact
- feature
- french
- generation
- genre
- grammatical features
- hypothesis
- inverse document frequency
- knowledge
- labeling
- learning paradigm
- likelihood
- linguistics
- maximum likelihood estimate
- meaning
- measure
- measures
- method
- nouns
- oracle
- paragraph
- parallel corpora
- parallel corpus
- parallel text
- performance ratio
- perplexity
- polysemous words
- precision
- probability
- process
- seed
- seed words
- semantic
- semcor
- semcor data
- statistical significance
- stress
- support vector
- tagging precision
- tags
- target word
- terms
- test corpora
- test data
- test set
- text
- thesaurus
- tokens
- training
- training corpora
- training corpus
- training data
- training data corpus
- training examples
- training set
- translations
- wall street journal corpus
- window size
- word
- word sense
- word senses
- wordnet
- words
