Machine translation using global lexical selection and sentence reconstruction
AssigneeAT&T Intellectual Property II, L.P.
InventorsSrinivas Bangalore, Patrick Haffner, Stephan Kanthak
Life of this patent
Estimating the term…
Abstract
Disclosed are systems, methods, and computer-readable media for performing translations from a source language to a target language. The method comprises receiving a source phrase, generating a target bag of words based on a global lexical selection of words that loosely couples the source words/phrases and target words/phrases, and reconstructing a target phrase or sentence by considering all permutations of words with a conditional probability greater than a threshold.
Claims21
A method comprising:
- receiving a source phrase in a source language;
- classifying the source phrase into a phrase meaning;
- matching the phrase meaning to a target phrase automaton in a target language, the target phrase automaton comprising a plurality of states and a plurality of arcs interconnecting the plurality of states to define a sentence path, each of the plurality of states defining a word position slot for inserting a word from a bag of words, and each of the plurality of arcs associated with a pre-defined insertion cost for inserting the word from the bag of words based on a previous state in the sentence path, wherein the predefined insertion cost has one of a positive and negative value which is associated with one of penalizing and rewarding the sentence path based on how many words are in the sentence path in order to produce more words in a target sentence relative to the source phrase;
- determining, for each word position slot in each target phrase automaton, a target word possibility from the bag of words associated with the word position slot based on at least one of a lexical translation of words in the source phrase and a phrase-to-phrase mapping;
- calculating a target sentence probability for the sentence path based at least on the pre-defined insertion cost associated with each of the plurality of arcs for the sentence path;
- and
- upon determining that the sentence path has a probability above a threshold, constructing the target sentence using the sentence path.
A system comprising:
- a processor;
- and
- a computer-readable storage medium storing instructions which, when executed by the processor, cause the processor to perform operations comprising:
- receiving a source phrase in a source language;
- classifying the source phrase into a phrase meaning;
- matching the phrase meaning to a target phrase automaton in a target language, the target phrase automaton comprising a plurality of states and a plurality of arcs interconnecting the plurality of states to define a sentence path, each of the plurality of states defining a word position slot for inserting a word from a bag of words, and each of the plurality of arcs associated with a pre-defined insertion cost for inserting the word from the bag of words based on a previous state in the sentence path, wherein the predefined insertion cost has one of a positive and negative value which is associated with one of penalizing and rewarding the sentence path based on how many words are in the sentence path in order to produce more words in a target sentence relative to the source phrase;
- determining, for each word position slot in each target phrase automaton, a target word possibility from the bag of words associated with the word position slot based on at least one of a lexical translation of words in the source phrase and a phrase-to-phrase mapping;
- calculating a target sentence probability for the sentence path based at least on the pre-defined insertion cost associated with each of the plurality of arcs for the sentence path;
- and
- upon determining that the sentence path has a probability above a threshold, constructing the target sentence using the sentence path.
A computer-readable storage device having instructions stored which, when executed by a computing device, cause the computing device to perform operations comprising:
- receiving a source phrase in a source language;
- classifying the source phrase into a phrase meaning;
- matching the phrase meaning to a target phrase automaton in a target language, the target phrase automaton comprising a plurality of states and a plurality of arcs interconnecting the plurality of states to define a sentence path, each of the plurality of states defining a word position slot for inserting a word from a bag of words, and each of the plurality of arcs associated with a pre-defined insertion cost for inserting the word from the bag of words based on a previous state in the sentence path, wherein the predefined insertion cost has one of a positive and negative value which is associated with one of penalizing and rewarding the sentence path based on how many words are in the sentence path in order to produce more words in a target sentence relative to the source phrase;
- determining, for each word position slot in each target phrase automaton, a target word possibility from the bag of words associated with the word position slot based on at least one of a lexical translation of words in the source phrase and a phrase-to-phrase mapping;
- calculating a target sentence probability for the sentence path based at least on the pre-defined insertion cost associated with each of the plurality of arcs for the sentence path;
- and
- upon determining that the sentence path has a probability above a threshold, constructing the target sentence using the sentence path.
Dependencies are inferred from the claim text.