ACL RD-TEC 1.0 Summarization of E06-1006
Paper Title:
PHRASE-BASED BACKOFF MODELS FOR MACHINE TRANSLATION OF HIGHLY INFLECTED LANGUAGES
PHRASE-BASED BACKOFF MODELS FOR MACHINE TRANSLATION OF HIGHLY INFLECTED LANGUAGES
Authors: Mei Yang and Katrin Kirchhoff
Primarily assigned technology terms:
- backoff translation
- bootstrapping
- compound splitting
- cross-language information retrieval
- decoder
- decoding
- decomposition
- frequency estimation
- giza
- information retrieval
- kneser-ney smoothing
- language modeling
- language processing
- machine translation
- machine translation system
- modeling
- morphological decomposition
- morphological processing
- morphology
- mt system
- natural language processing
- normalization
- part-of-speech tagger
- phrasal translation
- phrase extraction
- phrase prediction
- phrase translation
- phrase-based machine translation
- phrase-based translation
- preprocessing
- processing
- score combination
- segmentation
- segmentation method
- segmentation process
- semantic role labeling
- single reference translation
- smoothing
- smt system
- splitting
- statistical language modeling
- statistical mt
- tagger
- training procedure
- translation system
- word alignment
- word-by-word translation
- word-to-word translation
Other assigned terms:
- agglutinative language
- alignment error rate
- backoff
- backoff model
- backoff system
- baseline model
- bitext
- bleu
- bleu score
- bleu scores
- case
- characters
- compounds
- data sets
- determiners
- development set
- distribution
- english translation
- error rate
- estimation
- europarl corpus
- evaluation method
- foreign language
- hierarchical lexicon
- hypotheses
- inflected forms
- inflectional morphology
- knowledge
- labeling
- language model
- language model score
- lexical translation
- lexicon
- lexicon model
- linguistic
- linguistic knowledge
- machine translation output
- method
- morpho-syntactic knowledge
- morphological information
- morphological knowledge
- n-gram
- natural language
- normalization factor
- nouns
- paraphrase
- part-of-speech
- part-of-speech information
- part-of-speech tags
- phrase
- phrase level
- precision
- probabilities
- probability
- probability distributions
- probability estimates
- procedure
- process
- reference translation
- relative frequency
- reordering
- role labeling
- semantic
- semantic role
- sentence
- sentences
- source language
- statistical significance
- stem
- stems
- syntactic constructions
- tags
- target language
- technique
- test data
- test set
- text
- tokens
- training
- training data
- training set
- translation model
- translation output
- translation probabilities
- translation probability
- translations
- trigram
- trigram language model
- verb
- verb forms
- vocabulary
- vocabulary size
- word
- word alignments
- word error rate
- word form
- word order
- word-to-word translation model
- words
