ACL RD-TEC 1.0 Summarization of J98-4003
Paper Title:
MACHINE TRANSLITERATION
MACHINE TRANSLITERATION
Authors: Kevin Knight and Jonathan Graehl
Primarily assigned technology terms:
- algorithm
- approximation
- character recognition
- computational linguistics
- corpus analysis
- cryptography
- data collection
- decision trees
- em algorithm
- em training
- finite-state acceptor
- finite-state transducers
- forward transliteration
- graph algorithm
- hybrid neural-net\/expert-system
- japanese\/english machine translation
- language processing
- learning
- learning algorithm
- learning approach
- machine translation
- machine transliteration
- neural-net\/expert-system
- optical character recognition
- pattern-matching
- processing
- pronouncer
- pruning
- recognition
- romanization
- scoring
- scoring method
- search
- shortest-path
- speech recognition
- spelling
- transducers
- translator
- translators
- transliteration
- transliteration process
- transliterator
- unigram scoring
- viterbi
- weighted finite-state transducers
- word processing
- word selection
Other assigned terms:
- accent
- alphabet
- approach
- auxiliary verbs
- back-transliteration
- bigram
- bigram model
- bilingual dictionaries
- bilingual dictionary
- bilingual glossary
- case
- characters
- composition
- conditional probabilities
- corpora
- device
- dictionaries
- dictionary
- distribution
- english translations
- fact
- feature
- frequency counts
- frequency list
- gazetteer
- generative model
- generative models
- glossary
- human transliterator
- hypotheses
- katakana
- knowledge
- language model
- language pairs
- lattices
- linguistics
- mapping
- mappings
- maps
- method
- names
- noise
- pause
- phoneme
- phonetic alphabet
- phrase
- probabilistic models
- probabilities
- probability
- probability distribution
- probability distributions
- process
- pronunciation
- proper names
- recognition errors
- russian
- sentence
- sentential context
- sequence model
- stress
- symbol
- symbols
- technical terms
- terms
- text
- textbook
- theorem
- theory
- tokens
- training
- training corpora
- training set
- translation accuracy
- translations
- transliteration english translations
- trees
- unigram
- vocabulary
- vowel
- wall street journal corpus
- word
- word sequence
- word sequences
- words
- world knowledge
