Indicating and correcting errors in machine translation systems

• US 8,959,011 B2
• Filed: 01/26/2012
• Issued: 02/17/2015
• Est. Priority Date: 03/22/2007
• Status: Expired due to Fees

• Current Assignee: ABBYY Production LLC (ABBYY Software)
• Original Assignee: ABBYY InfoPoisk LLC
• Inventors: Anisimovich, Konstantin, Danielyan, Tatiana, Selegey, Vladimir, Zuev, Konstantin
• Primary Examiner(s): Lerner, Martin
• Application Number:
  US13/359,392
• Publication Number:
  US 20120123766A1
• Time in Patent Office: 1,118 Days
• Field of Search: 704/2, 704/3, 704/4, 704/5, 704/9, 704/277
• US Class Current: 704/2
• CPC Class Codes: G06F 40/51 Translation evaluation
• CPC Class Codes: G06F 40/58 Use of machine translation, e.g. for multi-lingual retrieval, for server-side translation for client devices or for real-time translation

Abstract

The preferred embodiments provide an automated machine translation from one language to another. The source language may contain expressions or words that are not readily handled by the translation system. Such problematic words or word combinations may, for example, include the words not found in the dictionary of the translation system, as well as text fragments corresponding to structures with low ratings. To improve translation quality, such potentially erroneous words or questionable word combinations are identified by the translation system and displayed to a user by distinctive display styles in the display of a document in the source language and in its translation to a target language. A user is provided with a capability to correct erroneous or questionable words so as to improve the quality of translation.

Claims

1. A method of providing language translation of text from a source language to a target language in a document, the method comprising:

performing, using a processor, lexical-morphological analysis of a source sentence of the text in the source language for generating a lexical-morphological structure of the source sentence;

if the lexical-morphological analysis identifies words which are not found in dictionaries or which have no corresponding morphological description or lexical description, displaying, using an output device, at least one such word in a visually distinctive manner to a user so as to indicate that at least one such word is potentially erroneous;

performing syntactic analysis of the lexical-morphological structure of the source sentence for generating a language-independent semantic structure of the source sentence;

if, during the syntactic analysis, a hypothesis with a low rating is selected by a processor as a preferred syntactic structure of the source sentence for generating the language-independent semantic structure, displaying in a visually distinctive manner to a user a fragment of the source sentence corresponding to the preferred syntactic structure so as to indicate that the fragment is potentially erroneous; and

performing syntactic synthesis for generating a target sentence in the target language from the source sentence.

2. The method of claim 1, further comprising, if, during syntactic synthesis, a hypothesis with a low rating or a default structure is selected by a processor, displaying in a visually distinctive manner to a user a fragment of the source sentence corresponding to the hypothesis with a low rating or the default structure, for indicating that the fragment is potentially erroneous.

3. The method of claim 2, further comprising receiving an input from a user so as to change in the text in the source language at least one word which has been indicated by a processor as potentially erroneous.

4. The method of claim 3, further comprising modifying translation of the source sentence in a constant manner with the received input of a user.

5. The method of claim 3, wherein the step of receiving input from a user further comprises displaying, by a processor, resembling substitution alternatives for the at least one word in the document indicated as potentially erroneous.

6. The method of claim 5, further comprising

enabling a user to manually select a subject of the source language from a list of source language subjects in a window;

and

adjusting the resembling substitution alternatives selected by a user from the window with the list to terms consistent with the subject as chosen by a user.

7. The method of claim 1, wherein the step of performing syntactic synthesis includes generating a surface syntactic structure of the target sentence in the target language and performing lexical selection.

8. One or more non-transitory physical computer-accessible media encoded with instructions for a processor, for performing a method, the method comprising:

performing, using a processor, lexical-morphological analysis of a source sentence of the text in the source language for generating a lexical-morphological structure of the source sentence;

if the lexical-morphological analysis identifies words which are not found in dictionaries or which have no corresponding morphological description or lexical description, providing at least one such word for displaying, using an output device, in a visually distinctive manner to a user so as to indicate that the at least one such word is potentially erroneous;

performing syntactic analysis of the lexical-morphological structure of the source sentence for generating a language-independent semantic structure of the source sentence;

if, during the syntactic analysis, a hypothesis with a low rating is selected by a processor as a preferred syntactic structure, providing a fragment of the source sentence corresponding to the preferred syntactic structure for display in a visually distinctive manner so as to indicate that the fragment is potentially erroneous; and

performing syntactic synthesis for generating a target sentence in a target language from the source sentence.

9. The non-transitory media of claim 8, wherein the method further comprises, if, during syntactic synthesis, a hypothesis with a low rating or a default structure is selected by a processor, providing for display in a visually distinctive manner a fragment of the source sentence corresponding to the hypothesis with the low rating or the default structure, for indicating that the fragment is potentially erroneous.

10. The non-transitory media of claim 8, wherein the method further comprises enabling a user to provide input for changing in the text in the source language at least one word or fragment indicated as potentially erroneous.

11. The non-transitory media of claim 10, wherein the method further comprises modifying translation of the source sentence in a constant manner with the input provided by the user.

12. The non-transitory media of claim 10, wherein the method further comprises receiving input from a user for changing in the text in the source language at least one word indicated as potentially erroneous and displaying, by a processor, resembling substitution alternatives for the at least one word in the document indicated as potentially erroneous.

13. The non-transitory media of claim 12, further comprising

enabling the user to manually select a subject of the source language from a list of source language subjects in a window;

and

adjusting the resembling substitution alternatives selected by a user from the window with the list to terms consistent with the subject in the target sentence.

14. The non-transitory media of claim 8, wherein the step of performing syntactic synthesis includes generating a surface syntactic structure of the target sentence in the target language and performing lexical selection.

15. A system comprising:

a processor in communication with a memory, wherein the memory is configured with instructions to cause the processor to perform actions comprising;

performing, using a processor, lexical-morphological analysis of a source sentence of the text in the source language for generating a lexical-morphological structure of the source sentence;

if the lexical-morphological analysis identifies words which are not found in dictionaries or which have no corresponding morphological description or lexical description providing at least one such word for displaying in a visually distinctive manner so as to indicate that the at least one such word is potentially erroneous;

performing syntactic analysis of the lexical-morphological structure of the source sentence for generating a language-independent semantic structure of the source sentence;

if, during the syntactic analysis, a hypothesis with a low rating is selected by a processor as a preferred syntactic structure of the source sentence for generating the language-independent semantic structure, providing a fragment of the source sentence corresponding to the preferred syntactic structure for displaying in a visually distinctive manner so as to indicate that the fragment is potentially erroneous; and

performing syntactic synthesis for generating a target sentence in a target language from the source sentence.

16. The system of claim 15, wherein the processing device further is configured so that, if, during syntactic synthesis, a hypothesis with a low rating or a default structure is selected by a processor, providing for display in a visually distinctive manner a fragment of the source sentence corresponding to the hypothesis with the low rating or the default structure, for indicating that the fragment is potentially erroneous.

17. The system of claim 15, wherein the processing device further is enabling a user to provide input so as to change in the text in the source language at least one word or fragment which has been indicated by a processor as potentially erroneous.

18. The system of claim 17, wherein the processing device further is modifying translation of the source sentence in a constant manner with the input provided by a user.

19. The system of claim 17, wherein the processing device further is

receiving input from a user for changing in the text in the source language at least one word which has been indicated by a processor as potentially erroneous and

displaying resembling substitution alternatives for the at least one word indicated as potentially erroneous.

20. The system of claim 19, wherein the processing device further is

enabling the user to manually select a subject of the source language from a list of source language subjects in a window;

and

adjusting the resembling substitution alternatives selected by a user from the window with the list to terms consistent with the subject as chosen by a user.