Optimizing parameters for machine translation

• US 8,285,536 B1
• Filed: 07/31/2009
• Issued: 10/09/2012
• Est. Priority Date: 07/31/2009
• Status: Expired due to Fees

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Kumar, Shankar, Macherey, Wolfgang, Och, Franz Josef, Dyer, Christopher James
• Primary Examiner(s): Godbold, Douglas
• Application Number:
  US12/533,519
• Time in Patent Office: 1,166 Days
• Field of Search: 704 2- 8
• US Class Current: 704/2
• CPC Class Codes: G06F 40/42 Data-driven translation

Abstract

Methods, systems, and apparatus, including computer program products, for language translation are disclosed. In one aspect, a method includes accessing a translation hypergraph that represents a plurality of candidate translations, the translation hypergraph including a plurality of paths including nodes connected by edges; calculating first posterior probabilities for each edge in the translation hypergraph; calculating second posterior probabilities for each n-gram represented in the translation hypergraph based on the first posterior probabilities; and performing decoding on the translation hypergraph using the second posterior probabilities to convert a sample text from a first language to a second language.

Claims

1. A computer-implemented method comprising:

accessing, at a computing device including a processor, a translation hypergraph that represents a plurality of candidate translations, the translation hypergraph including a plurality of paths including nodes connected by edges;

calculating, at the computing device, first posterior probabilities for each edge in the translation hypergraph;

calculating, at the computing device, second posterior probabilities for each n-gram represented in the translation hypergraph based on the first posterior probabilities; and

performing, at the computing device, decoding on the translation hypergraph using the second posterior probabilities to convert a sample text from a first language to a second language,

where calculating the second posterior probabilities includes calculating;

2. A computer-implemented method comprising:

accessing, at a computing device including a processor, a translation lattice that represents a plurality of candidate translations, the translation lattice including a plurality of paths including nodes connected by edges;

calculating, at the computing device, posterior probabilities for each n-gram represented in the translation lattice including;

determining, for each node, a score for each n-gram that is represented by a path from a source node of the translation lattice to the node, where the score is a highest posterior probability for edges on paths that include the n-gram; and

performing, at the computing device, decoding on the translation lattice using the scores to convert a sample text from a first language to a second language,

where calculating the posterior probabilities includes calculating;

3. A computer program product, encoded on a non-transitory program carrier, operable to cause a processor of a computing device to perform operations comprising:

accessing a translation hypergraph that represents a plurality of candidate translations, the translation hypergraph including a plurality of paths including nodes connected by edges;

calculating first posterior probabilities for each edge in the translation hypergraph;

calculating second posterior probabilities for each n-gram represented in the translation hypergraph based on the first posterior probabilities; and

performing decoding on the translation hypergraph using the second posterior probabilities to convert a sample text from a first language to a second language,

where calculating the second posterior probabilities includes calculating;

4. A computer program product, encoded on a non-transitory program carrier, operable to cause a data processing apparatus to perform operations comprising:

accessing a translation lattice that represents a plurality of candidate translations, the translation lattice including a plurality of paths including nodes connected by edges;

calculating posterior probabilities for each n-gram represented in the translation lattice including;

determining, for each node, a score for each n-gram that is represented by a path from a source node of the translation lattice to the node, where the score is a highest posterior probability for edges on paths that include the n-gram; and

performing decoding on the translation lattice using the scores to convert a sample text from a first language to a second language,

where calculating the posterior probabilities includes calculating;