Domain specific salient point translation

Patent metadata

Field	Value
Publication number	US-9547641-B2
Application number	US-201314038088-A
Country	US
Kind code	B2
Filing date	Sep 26, 2013
Priority date	Sep 26, 2013
Publication date	Jan 17, 2017
Grant date	Jan 17, 2017

Abstract

A computer generates a target language text phrase from a source language text phrase. The computer receives a text phrase in a source language. The computer then determines one or more salient points of the received source language text phrase. The computer determines one or more salient points in a target language that correspond to the one or more source language salient points. The computer then generates a target language text phrase based on the one or more salient points in the target language.

First claim

What is claimed is:

1. A method for generating a target language narrative, the method comprising:
receiving, by one or more processors, a text phrase in a source language;
determining, by the one or more processors, an industry domain of the received source language text phrase;
performing, by the one or more processors, a syntactic and semantic analysis of the received source language text phrase to determine one or more salient points representing the received source language text phrase, wherein the one or more salient points are context free with respect to the determined industry domain, such that the one or more salient points have unambiguous meanings outside of the context of the determined industry domain;
translating, by the one or more processors, the determined one or more salient points of the source language, without translating the received source language text phrase itself, into one or more salient points of a target language; and
generating, by the one or more processors, a target language narrative that includes the translated one or more salient points in the target language, utilizing a target language text generator configured with respect to the determined industry domain, such that a context of the generated narrative is specific to the determined industry domain.

2. A method in accordance with claim 1 , wherein the one or more target language salient points are determined, by the one or more processors, from a stored list of source language salient points and corresponding target language salient points.

3. A method in accordance with claim 2 , wherein the stored list includes corresponding target language salient points in multiple target languages.

4. A method in accordance with claim 1 , wherein the one or more source language salient points are determined based, at least in part, on an ontology model, wherein the ontology model includes a listing of relationships between words within the context of the determined industry domain.

5. A method in accordance with claim 1 , further comprising:
determining, by the one or more processors, a format of the received source language text phrase.

6. A method in accordance with claim 5 , further comprising:
determining, by the one or more processors, a format of the generated target language narrative based on the determined format of the received source language text phrase.

7. The method of claim 1 , wherein the determined format of the generated target language narrative is the same as the determined format of the received source language text phrase.

8. The method of claim 7 , wherein the determined format of both the received source language text phrase and the generated target language narrative includes at least a text color.

9. The method of claim 7 , wherein the determined format of both the received source language text phrase and the generated target language narrative includes at least a caption.

10. A computer program product for generating a target language narrative, the computer program product comprising:
one or more computer-readable storage devices and program instructions stored on at least one of the one or more storage media, the program instructions comprising:
program instructions to receive a text phrase in a source language;
program instructions to determine an industry domain of the received source language text phrase;
program instructions to perform a syntactic and semantic analysis of the received source language text phrase to determine one or more salient points representing the received source language text phrase, wherein the one or more salient points are context free with respect to the determined industry domain, such that the one or more salient points have unambiguous meanings outside of the context of the determined industry domain;
program instructions to translate the determined one or more salient points of the source language, without translating the received source language text phrase itself, into one or more salient points of a target language; and
program instructions to generate a target language narrative that includes the translated one or more salient points in the target language, utilizing a target language text generator configured with respect to the determined industry domain, such that a context of the generated narrative is specific to the determined industry domain.

11. A computer program product in accordance with claim 10 , wherein the one or more target language salient points are determined from a stored list of source language salient points and corresponding target language salient points.

12. A computer program product in accordance with claim 11 , wherein the stored list includes corresponding target language salient points in multiple target languages.

13. A computer program product in accordance with claim 10 , wherein the one or more source language salient points are determined based, at least in part, on an ontology model, wherein the ontology model includes a listing of relationships between words within the context of the determined industry domain.

14. A computer program product in accordance with claim 10 , further comprising:
program instructions to determine a format of the received source language text phrase.

15. A computer program product in accordance with claim 14 , further comprising:
program instructions to determine a format of the generated target language narrative based on the determined format of the received source language text phrase.

16. A system for generating a target language narrative, the system comprising:
one or more processors, one or more computer-readable memories, one or more computer-readable storage devices, and program instructions stored on at least one of the one or more storage media for execution by at least one of the one or more processors via at least one of the one or more memories, the program instructions comprising:
program instructions to receive a text phrase in a source language;
program instructions to determine an industry domain of the received source language text phrase;
program instructions to perform a syntactic and semantic analysis of the received source language text phrase to determine one or more salient points representing the received source language text phrase, wherein the one or more salient points are context free with respect to the determined industry domain, such that the one or more salient points have unambiguous meanings outside of the context of the determined industry domain;
program instructions to translate the determined one or more salient points of the source language, without translating the received source language text phrase itself, into one or more salient points of a target language; and
program instructions to generate a target language narrative that includes the translated one or more salient points in the target language, utilizing a target language text generator configured with respect to the determined industry domain, such that a context of the generated narrative is specific to the determined industry domain.

17. A system in accordance with claim 16 , wherein the one or more target language salient points are determined from a stored list of source language salient points and corresponding target language salient points.

18. A system in accordance with claim 16 , wherein the one or more source language salient points are determined based, at least in part, on an ontology model, wherein the ontology model includes a listing of relationships between words within the context of the determined industry domain

Assignees

• IBM

Inventors

• Bostick James E
• Ganci Jr John M
• Hua Kent
• Trim Craig M

Classifications

• G06F40/44

  Statistical methods, e.g. probability models · CPC title

• G06F40/30

  Semantic analysis · CPC title

• G06F40/42

  Data-driven translation · CPC title

• G06F40/45 Primary

  Example-based machine translation; Alignment · CPC title

• G06F16/374

  Thesaurus · CPC title