ACL RD-TEC 1.0 Summarization of W99-0901
Paper Title:
HIDING A SEMANTIC HIERARCHY IN A MARKOV MODEL
HIDING A SEMANTIC HIERARCHY IN A MARKOV MODEL
Authors: Steven Abney and Marc Light
Primarily assigned technology terms:
- algorithm
- ambiguity resolution
- approximation
- disambiguation
- em algorithm
- final state
- forward-backward algorithm
- hidden markov
- hidden markov model
- hmms
- illustration
- induction
- language modeling
- markov model
- modeling
- nlp
- normalization
- parameter estimation
- reasoning
- recognition
- sense disambiguation
- smoothing
- smoothing method
- speech recognition
- stochastic generation
- stochastic process
- structure disambiguation
- tile
- training method
- unsupervised parameter estimation
- word sense disambiguation
- word-sense disambiguation
Other assigned terms:
- ambiguity
- ambiguous word
- approach
- bias
- break
- case
- class hierarchy
- co-occurrence
- concept
- concepts
- conditional probability
- contextual information
- corpora
- descriptive adequacy
- distribution
- entropy
- estimation
- evaluation method
- events
- fact
- generation
- generation model
- hyponym
- hyponymy
- hypothesis
- interpretation
- language model
- leaf
- measures
- method
- minimum description length
- nonterminal
- normalization factor
- nouns
- parameter settings
- posterior
- posterior probability
- priori
- probabilities
- probability
- probability distribution
- process
- relation
- root node
- selectional preference
- selectional preferences
- selectional restriction
- semantic
- semantic class
- semantic classes
- semantic hierarchy
- sense ambiguity
- slot
- stochastic model
- symbol
- synonyms
- syntactic structure
- term
- terms
- test data
- tokens
- training
- training and test data
- training corpora
- training corpus
- training data
- transition probabilities
- uniform distribution
- uniform probability
- verb
- word
- word sense
- word sense ambiguity
- word senses
- wordnet
- words
