ACL RD-TEC 1.0 Summarization of W03-0317
Paper Title:
ACQUISITION OF ENGLISH-CHINESE TRANSLITERATED WORD PAIRS FROM PARALLEL-ALIGNED TEXTS USING A STATISTICAL MACHINE TRANSLITERATION MODEL
ACQUISITION OF ENGLISH-CHINESE TRANSLITERATED WORD PAIRS FROM PARALLEL-ALIGNED TEXTS USING A STATISTICAL MACHINE TRANSLITERATION MODEL
Authors: Chun-Jen Lee and Jason S. Chang
Primarily assigned technology terms:
- algorithm
- backtracking
- bilingual lexicon construction
- bilingual text alignment
- crosslanguage information retrieval
- decoding
- decomposition
- dictionary lookup
- dynamic programming
- em training
- error analysis
- estimation procedure
- expectation maximization
- information retrieval
- iterative procedure
- language processing
- learning
- learning approach
- lexicon construction
- likelihood estimation
- linguistic processing
- machine translation
- machine transliteration
- matching
- maximum likelihood
- maximum likelihood estimation
- measuring
- model training
- natural language processing
- nlp
- noisy channel model
- parameter estimation
- post-processing
- preprocessing
- processing
- pruning
- re-estimation
- romanization
- scoring
- sentence alignment
- source-to-target word transliteration
- tagger
- text alignment
- training algorithm
- training procedure
- transliteration
- unsupervised learning
- validation
- viterbi
- viterbi decoding
- word extraction
- word transliteration
Other assigned terms:
- alignment procedure
- approach
- back-transliteration
- baseline model
- bilingual corpora
- bilingual dictionaries
- bilingual lexicon
- bilingual text
- case
- characters
- chinese characters
- chinese word
- computational complexity
- convergence
- corpora
- dictionaries
- dictionary
- distribution
- estimation
- evaluation metric
- experimental results
- formal description
- heuristic
- heuristic rules
- knowledge
- language model
- language pairs
- language-dependent knowledge
- ldoce
- lexicon
- likelihood
- linguistic
- linguistic constraints
- mapping
- mapping rules
- measure
- method
- model parameters
- names
- natural language
- nlp applications
- noisy channel
- nouns
- paragraph
- parallel corpus
- parallel text
- parallel texts
- performance evaluation
- phonemes
- pinyin
- precision
- prior probability
- probabilities
- probability
- procedure
- process
- pronunciation
- pronunciation dictionary
- proper name
- proper names
- proper noun
- recursion
- sentence
- sentence level
- similarity score
- similarity scores
- source language
- source sentence
- source text
- statistical information
- symbols
- target language
- target sentence
- target text
- target word
- term
- testing data
- testing set
- text
- training
- training set
- translation model
- translation probability
- translations
- transliteration model
- uniform distribution
- viterbi path
- vowel
- word
- word pair
- words
