ACL RD-TEC 1.0 Summarization of W04-0846
Paper Title:
CONTEXT CLUSTERING FOR WORD SENSE DISAMBIGUATION BASED ON MODELING PAIRWISE CONTEXT SIMILARITIES
CONTEXT CLUSTERING FOR WORD SENSE DISAMBIGUATION BASED ON MODELING PAIRWISE CONTEXT SIMILARITIES
Authors: Cheng Niu and Wei Li and Rohini K. Srihari and Huifeng Li and L. Crist
Primarily assigned technology terms:
- algorithm
- bootstrapping
- bootstrapping approach
- clustering
- clustering algorithm
- computational linguistics
- conditional maximum entropy
- context clustering
- context modeling
- corpus construction
- disambiguation
- generative modeling
- hard clustering
- iterative scaling
- language processing
- latent semantic analysis
- learning
- learning approaches
- many-to-one mapping
- maxent
- maximum entropy
- maximum entropy model
- modeling
- monte carlo simulation
- naive bayes
- natural language processing
- normalization
- optimization
- parser
- parsing
- processing
- sampling
- search
- searching
- semantic analysis
- sense disambiguation
- smoothing
- supervised learning
- training procedure
- weakly supervised learning
- word sense disambiguation
Other assigned terms:
- ambiguous word
- annotated corpus
- annotated training corpus
- approach
- association for computational linguistics
- bayesian framework
- case
- category level
- cluster
- clusters
- co-occurrence
- conditional probabilities
- conditional probability
- context cluster
- context features
- context model
- context similarity
- corpora
- correlation
- distribution
- entropy
- fact
- feature
- generative model
- generative probability
- independence assumption
- information independence
- joint probability
- knowledge
- latent semantic
- linguistics
- mapping
- measure
- measures
- method
- natural language
- normalization factor
- prior probability
- probabilities
- probability
- probability distribution
- procedure
- process
- search time
- semantic
- semantic space
- sense distinction
- senses of a word
- similarity measure
- similarity measures
- similarity model
- statistical models
- symbols
- technique
- testing corpus
- text
- tokens
- training
- training corpus
- training data
- vocabulary
- window size
- word
- word sense
- word senses
- words
