Background for example-based MT

The basic steps

  1. Matching: identify sentence(s) in database that match the input on the basis of fragments of the input.
  2. Alignment/adaptation: identify corresponding fragments in target-language sentences.
  3. Recombination: recombine target-language fragments to form output.

An example

I haven't read the book that you lent me.


Matching: some problems

Learning generalizations from the database (Brown, 2001)

Learning generalizations from the database (Brown, 2001)



Open-source resource for EBMT

CMU-EBMT (Brown, 2011)

Statistical Machine Translation: the noisy channel model

SMT training: the target language model

SMT: IBM Model 3 parameters

SMT training: alignment

SMT training: IBM Model 1

SMT: estimating translation probabilities

Estimating Model 1 probabilities: an example

21st century SMT

Log-linear (maximum entropy) models

Phrase-based and syntactic SMT


SMT: (phrase-based) decoding

Decoding: a partial example

SMT: evaluation

Open-source software toolkits for SMT

Munteanu and Marcu (2005)