ACL RD-TEC 1.0 Summarization of W02-1405
Paper Title:
IMPROVING A GENERAL-PURPOSE STATISTICAL TRANSLATION ENGINE BY TERMINOLOGICAL LEXICONS
IMPROVING A GENERAL-PURPOSE STATISTICAL TRANSLATION ENGINE BY TERMINOLOGICAL LEXICONS
Primarily assigned technology terms:
- adaptative statistical language modeling
- algorithm
- automatic translation
- backtracking
- consistency checking
- databases
- decoder
- decoding
- hidden markov
- language modeling
- language translation
- linking
- machine translation
- maximum entropy
- modeling
- ranking
- search
- search algorithm
- spoken language translation
- statistical decoder
- statistical language modeling
- statistical machine translation
- statistical translation
- symbolic translation
- terminology
- thresholding
- translation engine
- translation memory
- translation process
- translation system
- translator
- translators
- transtype
- tuning
- viterbi
- viterbi alignment
Other assigned terms:
- alignment models
- approach
- bilingual lexicons
- bitext
- cache-based language model
- case
- community
- comparative study
- corpora
- distribution
- english sentence
- entropy
- entropy models
- error rate
- fact
- french
- gold standard
- hansard corpus
- hypotheses
- hypothesis
- ibm models
- implementation
- independence assumption
- language model
- language models
- lexicon
- lexicon entry
- likelihood
- linguistic
- maximum entropy models
- measures
- morphological information
- normal distribution
- oracle
- pairs of words
- perplexity
- posterior
- probabilities
- procedure
- process
- recursion
- relation
- search space
- sentence
- sentence level
- sentences
- sniper corpus
- source sentence
- source text
- spoken language
- standard deviation
- statistical models
- stress
- target language
- target language model
- target string
- target word
- term
- terminological research
- terms
- test corpora
- test corpus
- text
- tokens
- training
- training corpus
- training time
- translation model
- translation output
- translation task
- translations
- trigram
- trigram language model
- trigram model
- unigram
- user
- vocabulary
- word
- word alignments
- word error rate
- word sequences
- words
