ACL RD-TEC 1.0 Summarization of P96-1010
Paper Title:
COMBINING TRIGRAM-BASED AND FEATURE-BASED METHODS FOR CONTEXT-SENSITIVE SPELLING CORRECTION
COMBINING TRIGRAM-BASED AND FEATURE-BASED METHODS FOR CONTEXT-SENSITIVE SPELLING CORRECTION
Authors: Andrew Golding and Yves Schabes
Primarily assigned technology terms:
- algorithm
- approximation
- bayesian approach
- classifiers
- context-sensitive spelling
- context-sensitive spelling correction
- error checking
- grammar checker
- hybrid method
- maximum-likelihood
- naive bayesian
- noisy channel model
- part-of-speech tagging
- postprocessing
- preprocessing
- pruning
- smoothing
- smoothing technique
- spelling
- spelling correction
- spelling-correction
- tagging
- tribayes
- word-processing
Other assigned terms:
- approach
- brown corpus
- case
- checker
- collocation
- conditional probability
- context words
- contextual information
- determiners
- fact
- feature
- function words
- geometric mean
- grammar
- heuristics
- independence assumption
- knowledge
- measure
- measures
- method
- methodology
- noisy channel
- nouns
- part of speech
- part-of-speech
- part-of-speech tag
- part-of-speech tags
- part-of-speech trigram
- parts of speech
- prediction accuracy
- probabilities
- probability
- right-hand side
- sentence
- sentences
- syntactic constraints
- system performance
- tag sequence
- tags
- target sentence
- target word
- technique
- term
- terms
- test phase
- test set
- text
- training
- training corpus
- training examples
- training phase
- training set
- trigram
- typographical errors
- unigram
- unigram probability
- user
- word
- word usage
- words
