ACL RD-TEC 1.0 Summarization of C04-1103
Paper Title:
DIRECT ORTHOGRAPHICAL MAPPING FOR MACHINE TRANSLITERATION
DIRECT ORTHOGRAPHICAL MAPPING FOR MACHINE TRANSLITERATION
Authors: Min Zhang and Haizhou Li and Jian Su
Primarily assigned technology terms:
- algorithm
- corpus alignment
- coupling
- cross validation
- database
- databases
- decoder
- decoding
- dictionary compilation
- dictionary lookup
- dom transliteration
- em algorithm
- english\/japanese transliteration
- entity translation
- error rate reduction
- error reduction
- grapheme-to-phoneme conversion
- information retrieval
- kneser-ney smoothing
- language processing
- learning
- likelihood approach
- listing
- machine learning
- machine translation
- machine transliteration
- machine transliteration and back-transliteration
- maximum likelihood
- maximum likelihood approach
- model training
- modeling
- name translation
- name transliteration
- named entity translation
- natural language processing
- noisy-channel model
- optimization
- orthographic mapping
- orthographical mapping
- phonetic mapping
- processing
- rate reduction
- reporting
- romanization
- segmentation
- smoothing
- stack decoder
- term processing
- text-to-speech
- training process
- transliteration
- transliteration process
- unsupervised training
- validation
- validation process
- viterbi
- viterbi algorithm
Other assigned terms:
- approach
- back-transliteration
- bigram
- bilingual dictionary
- case
- character error rate
- characters
- chinese characters
- computational complexity
- context information
- contextual information
- data sets
- dictionary
- distribution
- dom framework
- edict dictionary
- english translation
- error rate
- exact match
- foreign language
- french
- generation
- generative probability
- implementation
- interpretation
- joint probability
- katakana
- knowledge
- knowledge base
- language pair
- language pairs
- language processing tasks
- likelihood
- mapping
- meanings
- mechanisms
- method
- multilingual speech
- n-gram
- n-gram transliteration model
- named entities
- named entity
- names
- natural language
- natural language processing tasks
- ngram
- open test
- organization names
- orthography
- personal names
- phonemes
- phonemic representation
- phonetic association
- pinyin
- probability
- probability distribution
- process
- processing tasks
- proper names
- relation
- source language
- source language word
- source-channel model
- statistics
- substring
- system development
- system performance
- target language
- technical terms
- term
- terms
- test set
- tokens
- training
- training corpus
- training data
- training set
- transformation
- transformation rules
- transliteration model
- word
- word error rate
- word error rates
- word pair
- words
