ACL RD-TEC 1.0 Summarization of W04-3238
Paper Title:
SPELLING CORRECTION AS AN ITERATIVE PROCESS THAT EXPLOITS THE COLLECTIVE KNOWLEDGE OF WEB USERS
SPELLING CORRECTION AS AN ITERATIVE PROCESS THAT EXPLOITS THE COLLECTIVE KNOWLEDGE OF WEB USERS
Authors: Silviu Cucerzan and Eric Brill
Primarily assigned technology terms:
- computing
- decomposition
- distance function
- error analysis
- internet
- internet search
- iterative correction
- iterative process
- levenshtein
- lexicon-based spelling correction
- listing
- maximum likelihood
- measuring
- noisy channel model
- path search
- processing
- query spelling
- search
- search engine
- spell checker
- spelling
- spelling correction
- spelling-correction
- splitting
- synthesis
- tokenization
- viterbi
- viterbi search
- web search
- word substitution
Other assigned terms:
- alphabet
- annotators
- approach
- bigram
- case
- checker
- compounds
- corpora
- correlation
- correlation matrix
- data sets
- data structure
- document
- document collections
- edit distance
- english lexicon
- estimation
- evaluation data
- evaluation set
- evaluations
- fact
- gold standard
- implementation
- knowledge
- language model
- language model probability
- large corpus
- levenshtein distance
- lexical information
- lexicon
- likelihood
- measure
- measures
- model probability
- noisy channel
- noun phrase
- phrase
- precision
- prepositions
- prior probability
- probabilities
- probability
- probability model
- procedure
- process
- punctuation
- queries
- query
- relative frequency
- search procedure
- search query
- search space
- spelling suggestion
- statistics
- substring
- target language
- technique
- test data
- test set
- text
- tokens
- training
- transformation
- transition probabilities
- transposition
- unigram
- user
- web query
- word
- word boundaries
- word form
- word sequences
- words
