ACL RD-TEC 1.0 Summarization of I05-5003
Paper Title:
USING MACHINE TRANSLATION EVALUATION TECHNIQUES TO DETERMINE SENTENCE-LEVEL SEMANTIC EQUIVALENCE
USING MACHINE TRANSLATION EVALUATION TECHNIQUES TO DETERMINE SENTENCE-LEVEL SEMANTIC EQUIVALENCE
Authors: Andrew Finch and Young-Sook Hwang and Eiichiro Sumita
Primarily assigned technology terms:
- algorithm
- automatic machine translation
- automatic mt
- automatic mt evaluation
- classification
- classification method
- classifier
- classifiers
- confidence estimation
- database
- dialogue translation
- dialogue translation technology
- distance calculation
- edit distance algorithm
- encoding
- entailment recognition
- equivalence determination
- feature encoding
- information extraction
- internet
- kernels
- machine translation
- machine translation evaluation
- machine translation systems
- matching
- mt system
- paraphrase classification
- pos tagger
- pre-processing
- radial basis function
- recognition
- scoring
- semantic equivalence determination
- speech dialogue translation
- stemmer
- support vector machine
- tagger
- tokenization
- tokenizer
- translation evaluation
- translation systems
- translation technology
- tree edit distance
- word ordering
- word\/n-gram matching
Other assigned terms:
- ambiguity
- annotators
- approach
- arithmetic mean
- bag of words
- bleu
- bleu score
- brevity penalty
- case
- classification accuracy
- classification performance
- classification task
- comparable document
- corpora
- correlation
- distance measure
- distance score
- distribution
- document
- edit distance
- entailment
- error rate
- estimation
- fact
- feature
- feature vector
- feature vectors
- function words
- geometric mean
- grammatical structure
- grammaticality
- human annotators
- interannotator agreement
- labeling
- large corpus
- length similarity
- meaning
- measure
- measures
- method
- methodology
- mt evaluation
- n-gram
- n-grams
- ngram
- nist
- noise
- nouns
- paraphrase
- paraphrase corpus
- paraphrases
- part of speech
- pos tag
- precision
- process
- punctuation
- recognition corpus
- reference translation
- reference translations
- relation
- segments
- semantic
- semantic content
- semantic distance
- semantic information
- sentence
- sentence level
- sentence pair
- sentence-by-sentence basis
- sentences
- speech information
- statistics
- support vector
- surface form
- synonym
- synonyms
- tag set
- technique
- technology
- terms
- test data
- training
- training and test data
- training examples
- translation candidates
- translations
- tree
- unigram
- upenn tag set
- word
- word error rate
- word order
- word sequences
- wordnet
- wordnet database
- words
