Machine Translation/Statistics

Statistical machine translationEdit

Language modelsEdit

Language models are used in MT for a) scoring arbitrary sequences of words (tokens) and b) given a sequence of tokens, they predict what token will likely to follow the sequence. Formally, language models are probability distributions over sequences of tokens in a given language.

N-gram modelsEdit

Character-based modelsEdit

Recently, it was shown that it is possible to use sub-words, characters or even bytes as basic units for language modelling[citation needed]. There are a few events focused particularly on such models and in general, processing language data on sub-word units, e.g. SCLem 2017.

Translation modelsEdit

IBM models 1-5Edit

Phrase-based modelsEdit

Factored translation modelsEdit

Syntax- and tree-based modelsEdit

Synchronous phrase grammarEdit

Parallel tree-banksEdit

Syntactic rules extractionEdit

DecodingEdit

Beam searchEdit

Hybrid systemsEdit

Computer-aided translationEdit

Translation memoryEdit