Translation and dictionary selection by context

Patent metadata

Field	Value
Publication number	US-9817821-B2
Application number	US-201214653223-A
Country	US
Kind code	B2
Filing date	Dec 19, 2012
Priority date	Dec 19, 2012
Publication date	Nov 14, 2017
Grant date	Nov 14, 2017

Abstract

Methods are described for translation of one or more words in a source language into a target language based on context, history and meaning of portions of the source text. Translation may involve selection of electronic dictionaries when translating from a source language to one or more target languages. Various aspects of history, context and structures of words that reflect lexical, morphological, syntactic, and semantic properties facilitate selection or presentation of translations and options to a user. The methods are applicable to genre classification, topic detection, news analysis, authorship analysis, internet searches, and creating corpora for other tasks, etc.

First claim

We claim:

1. A method comprising:
generating, by at least one processor, a first translation of a first portion of a source text into a target language, wherein generating comprises extracting at least one first lexical meaning of the first portion;
selecting, by the processor, a first semantic class for the first lexical meaning from one or more first semantic classes in a semantic hierarchy, wherein the first semantic class corresponds to the first lexical meaning in the semantic hierarchy;
extracting, by the processor, a plurality of lexical meanings of a second portion of the source text, wherein the semantic hierarchy comprises a plurality of second semantic classes corresponding to the lexical meanings;
selecting, by the processor, at least one second lexical meaning for the second portion from the lexical meanings based on distances between the second semantic classes and the first semantic class within the semantic hierarchy;
generating, by the processor, a second translation of the second portion into the target language based on the selected second lexical meaning; and
outputting the generated second translation.

2. The method of claim 1 , further comprising selecting a translation dictionary from a plurality of translation dictionaries, wherein the selected translation dictionary is used in the generation of the second translation.

3. The method of claim 2 , wherein the selection of the translation dictionary is based on a selected subject matter of the first portion.

4. The method of claim 1 , further comprising identifying an instance of a plurality of parallel texts based on a selected subject matter of the first portion.

5. The method of claim 4 , wherein each of the plurality of parallel texts comprises a sequence of words.

6. The method of claim 4 , wherein each of the plurality of parallel texts comprises one or more sentences.

7. The method of claim 1 , further comprising:
identifying a plurality of translation variants for a word or expression in the second translation; and
outputting the plurality of translation variants in an order of relevancy.

8. The method of claim 1 , further comprising:
identifying a plurality of translation variants for a word or expression in the second translation;
identifying a respective relevancy value for each of the plurality of translation variants; and
outputting one or more of the plurality of translation variants whose respective relevancy value exceeds a threshold value.

9. The method of claim 1 , wherein the extraction of the first lexical meaning and the second lexical meanings is based on one or more lexical descriptions of a source language of the source text.

10. The method of claim 1 , wherein the extraction of the first lexical meaning and the second lexical meanings is based on one or more lexical selections.

11. The method of claim 1 , wherein the extraction of the first semantic meaning and the second lexical meanings is based on a deep linguistic analysis of the source text.

12. A system comprising:
at least one memory to store a source text; and
at least one processor, operatively coupled to the memory, configured to:
generate a first translation of a first portion of the stored source text into a target language, wherein, to generate the first translation, the processor is to extract at least one first lexical meaning of the first portion;
select a first semantic class for the first lexical meaning from one or more first semantic classes in a semantic hierarchy, wherein the first semantic class corresponds to the first lexical meaning in the semantic hierarchy;
extract a plurality of lexical meanings of a second portion of the source text, wherein the semantic hierarchy comprises a plurality of second semantic classes corresponding to the lexical meanings;
select at least one second lexical meaning for the second portion from the lexical meanings based on distances between the second semantic classes and the first semantic class within the semantic hierarchy;
generate a second translation of the second portion into the target language based on the selected second lexical meaning; and
output the generated second translation.

13. The system of claim 12 , wherein the processor is further configured to select a translation dictionary from a plurality of translation dictionaries, and wherein the selected translation dictionary is used in the generation of the second translation.

14. The system of claim 13 , wherein the selection of the translation dictionary is based on a selected subject matter of the first portion.

15. The system of claim 12 , wherein the processor is further configured to identify an instance of a plurality of parallel texts based on a selected subject matter of the first portion.

16. The system of claim 15 , wherein each of the plurality of parallel texts comprises a sequence of words.

17. The system of claim 15 , wherein each of the plurality of parallel texts comprises one or more sentences.

18. The system of claim 12 , wherein the processor is further configured to:
identify a plurality of translation variants for a word or expression in the second translation; and
output the plurality of translation variants in an order of relevancy.

19. The system of claim 12 , wherein the processor is further configured to:
identify a plurality of translation variants for a word or expression in the second translation;
identify a respective relevancy value for each of the plurality of translation variants; and
output one or more of the plurality of translation variants whose respective relevancy value exceeds a threshold value.

20. A non-transitory computer-readable medium having instructions stored therein that, when executed by at least one processor, cause the processor to:
generate, by the processor, a first translation of a first portion of a source text into a target language, wherein, to generate the first translation, the instructions cause the processor to extract at least one first lexical meaning of the first portion;
select, by the processor, a first semantic class for the first lexical meaning from one or more first semantic classes in a semantic hierarchy, wherein the first semantic class corresponds to the first lexical meaning in the semantic hierarchy;
extract, by the processor, a plurality of lexical meanings of a second portion of the source text, wherein the semantic hierarchy comprises a plurality of second semantic classes corresponding to the lexical meanings;
select, by the processor, at least one second lexical meaning for the second portion from the lexical meanings based on distances between the second semantic classes and the first semantic class within the semantic hierarchy;
generate, by the processor, a second translation of the second portion into the target language based on the selected second lexical meaning; and
output the generated second translation.

Assignees

• Abbyy Infopoisk Llc
• Abbyy Dev Llc

Inventors

• Rylov Alexander Gennadievich
• Arkhipov Ivan Sergeevich

Classifications

• G06F40/242

  Dictionaries · CPC title

• G06F40/40 Primary

  Processing or translation of natural language (natural language analysis G06F40/20; semantic analysis G06F40/30) · CPC title

• G06F40/58 Primary

  Use of machine translation, e.g. for multi-lingual retrieval, for server-side translation for client devices or for real-time translation · CPC title

• G06F17/289 Primary

  Physics · mapped topic

• G06F17/28

  Physics · mapped topic