ACL RD-TEC 1.0 Summarization of P06-1091

Paper Title:
A DISCRIMINATIVE GLOBAL TRAINING ALGORITHM FOR STATISTICAL MT

Authors: Christoph Tillmann and Tong Zhang

Other assigned terms:

  • annotation
  • approach
  • arabic-english translation task
  • arabic-to-english translation task
  • association for computational linguistics
  • bigram
  • binary features
  • bleu
  • bleu metric
  • bleu score
  • bleu scores
  • candidate translation
  • case
  • classification performance
  • classification problem
  • convergence
  • data set
  • decoder parameter
  • dimensionality
  • domain knowledge
  • estimation
  • evaluation metric
  • evaluation test
  • experimental results
  • fact
  • feature
  • feature sets
  • feature types
  • feature vector
  • feature vectors
  • generation
  • generation process
  • heuristic
  • implementation
  • knowledge
  • labeling
  • language annotation
  • language model
  • language model probability
  • learning rate
  • lemma
  • linear combination
  • linguistics
  • method
  • model parameters
  • model probability
  • monolingual corpus
  • multi-class classification problem
  • natural language
  • noisy channel
  • optimization criterion
  • part-of-speech
  • phrase
  • phrase translation accuracy
  • possible translation
  • probabilities
  • probability
  • procedure
  • process
  • punctuation
  • reference translation
  • reference translations
  • regularization parameter
  • search procedure
  • search space
  • sentence
  • sentence pair
  • sentences
  • sequence model
  • source sentence
  • standard arabic-english translation task
  • system performance
  • target language
  • target sentence
  • target word
  • terms
  • test data
  • test set
  • training
  • training data
  • translation accuracy
  • translation model
  • translation probabilities
  • translation score
  • translation task
  • translations
  • weight vector
  • word
  • word level
  • word sequence
  • words

Extracted Section Types:


This page last edited on 10 May 2017.

*** ***