ACL RD-TEC 1.0 Summarization of W06-3108

Other assigned terms:

• approach
• bilingual corpus
• bilingual training corpus
• binary features
• bleu
• bleu score
• case
• case information
• classification error
• classification error rate
• classification performance
• classification task
• context information
• convergence
• corpora
• data sparseness
• data sparseness problem
• decision rule
• distribution
• english corpus
• entropy
• error rate
• evaluation set
• evaluations
• events
• experimental results
• feature
• feature set
• feature types
• feature weights
• gaussian prior
• grammars
• heuristic
• ibm model
• lambda
• language model
• language pair
• language pairs
• lexicon
• lexicon model
• likelihood
• local context
• log-linear combination
• machine translation output
• maximum entropy principle
• measures
• method
• model parameters
• multilingual speech
• n-gram
• n-gram language model
• nist
• normalization factor
• optimization criterion
• part-of-speech
• part-of-speech information
• phrase
• phrase boundary
• phrase level
• phrase translation model
• posterior
• posterior probability
• prior distribution
• probabilities
• probability
• procedure
• process
• punctuation
• punctuation marks
• reference translations
• reordering
• sentence
• sentence structure
• sentences
• source language
• source language sentence
• source sentence
• sparseness problem
• speech corpus
• statistics
• target language
• target language sentence
• target sentence
• target word
• technology
• test corpus
• test data
• toolkit
• training
• training corpora
• training corpus
• training criterion
• training data
• translation model
• translation output
• translation quality
• translations
• travel expression corpus
• word
• word classes
• word error rate
• word level
• word order
• word-aligned corpus
• words