In-context exact (ICE) matching

• US 7,983,896 B2
• Filed: 03/03/2005
• Issued: 07/19/2011
• Est. Priority Date: 03/05/2004
• Status: Active Grant

• Current Assignee: SDL Inc (RWS Holdings Plc)
• Original Assignee: SDL Language Technologies
• Inventors: Ross, Russ, Gillespie, Kevin
• Primary Examiner(s): Dorvil; Richemond
• Assistant Examiner(s): Colucci; Michael C
• Application Number:
  US11/071,706
• Publication Number:
  US 20050197827A1
• Time in Patent Office: 2,329 Days
• Field of Search: 704/4, 704/9, 704/260, 704/251, 704/2, 704/235, 704/270, 704/277, 704/5, 341/107, 715/708, 715/532, 715/268, 707/3
• US Class Current: 704/2
• CPC Class Codes: G06F 16/90344 by using string matching techniques
• CPC Class Codes: G06F 40/47 Machine-assisted translation, e.g. using translation memory

Abstract

Methods, systems and program product are disclosed for determining a matching level of a text lookup segment with a plurality of source texts in a translation memory in terms of context. In particular, the invention determines any exact matches for the lookup segment in the plurality of source texts, and determines, in the case that at least one exact match is determined, that a respective exact match is an in-context exact (ICE) match for the lookup segment in the case that a context of the lookup segment matches that of the respective exact match. The degree of context matching required can be predetermined, and results prioritized. The invention also includes methods, systems and program products for storing a translation pair of source text and target text in a translation memory including context, and the translation memory so formed. The invention ensures that content is translated the same as previously translated content and reduces translator intervention.

Claims

1. A method, performed on a computer system, for determining a matching level of a plurality of source texts stored in a translation memory to a lookup segment to be translated, the method comprising:

using the computer system to perform the following;

determining any exact matches for the lookup segment in the plurality of source texts;

determining, in the case that at least one exact match is determined, that a respective exact match is an in-context exact (ICE) match for the lookup segment in the case that a context of the lookup segment matches that of the respective exact match, wherein the context includes at least two levels, wherein the at least two levels includes a usage context level and an asset context level, the usage context level including a preceding usage context level and a post usage context level, the preceding usage context level having a preceding usage context hash code associated therewith and the post usage context level having a post usage context hash code associated therewith; and

determining, in the case that greater than one ICE match is determined, a prioritization for each ICE match that is based on a degree of context matching, wherein a more appropriate ICE match is prioritized over one or more other ICE matches, wherein the prioritization includes;

first preferring an ICE match having a full usage context match with the lookup segment over an ICE match having only a partial usage context match with the lookup segment, wherein the full usage context match with the lookup segment includes a match with both the preceding usage context hash code and the post usage context hash code and the partial usage context match with the lookup segment includes a match with only one of the preceding usage context hash code and the post usage context hash code,

where if the first preferring step is non-conclusive, second preferring an ICE match from the same asset as the lookup segment over an ICE match from a different asset.

2. The method of claim 1, wherein the ICE match determining indicates that a respective exact match is an ICE match for the lookup segment only in the case that each context level of the lookup segment matches that of the respective exact match.

3. The method of claim 1, wherein the ICE match determining indicates that a respective exact match is an ICE match for the lookup segment only in the case that both the usage context level and the asset context level of the lookup segment matches that of the respective exact match.

4. The method of claim 1, wherein if the second preferring is non-conclusive, third preferring an ICE match with a closest position to a position of the lookup segment within the asset.

5. The method of claim 4, further comprising allowing a user to select the ICE match based on the prioritization.

6. The method of claim 1, wherein the lookup segment includes a plurality of lookup segments that are substantially identical in terms of content, and

wherein the ICE match determining includes determining an ICE match for each lookup segment.

7. The method of claim 5, wherein at least one lookup segment has a different ICE match than at least one other lookup segment.

8. The method of claim 1, further comprising allowing retrieval of at least one source text based on the matching level.

34. The method of claim 4, wherein the preceding usage context hash code is generated from a text stream that precedes the segment to be translated and the post usage context hash code is generated from a text stream that follows the segment to be translated.

9. A computer system for determining a matching level of a plurality of source texts stored in a translation memory to a lookup segment to be translated, the computer system comprising:

at least one processing unit;

memory operably associated with the at least one processing unit;

an in-context exact (ICE) match translation system storable in memory and executable by the at least one processing unit, the ICE translation system comprising;

an exact match determinator that determines any exact matches for the lookup segment in the plurality of source texts;

an ICE match determinator that determines in the case that at least one exact match is determined, that a respective exact match is an ICE match for the lookup segment in the case that a context of the lookup segment matches that of the respective exact match, wherein the context includes at least two levels, wherein the at least two levels includes a usage context level and an asset context level, the usage context level including a preceding usage context level and a post usage context level, the preceding usage context level having a preceding usage context hash code associated therewith and the post usage context level having a post usage context hash code associated therewith; and

an ICE match prioritizer that in the case that greater than one ICE match is determined by the ICE match determinator, prioritizes each ICE match based on a degree of context matching, wherein a more appropriate ICE match is prioritized over other ICE matches, wherein the prioritizing includes;

first preferring an ICE match having a full usage context match with the lookup segment over an ICE match having only a partial usage context match with the lookup segment, wherein the full usage context match with the lookup segment includes a match with both the preceding usage context hash code and the post usage context hash code and the partial usage context match with the lookup segment includes a match with only one of the preceding usage context hash code and the post usage context hash code,

where if the first preferring step is non-conclusive, second preferring an ICE match from the same asset as the lookup segment over an ICE match from a different asset.

10. The computer system of claim 9, wherein the ICE match determinator indicates that a respective exact match is an ICE match for the lookup segment only in the case that each context level of the lookup segment matches that of the respective exact match.

11. The computer system of claim 9, wherein the ICE match determinator indicates that a respective exact match is an ICE match for the lookup segment only in the case that both the usage context level and the asset context level of the lookup segment matches that of the respective exact match.

12. The computer system of claim 9, wherein if the second preference is non-conclusive, the ICE match prioritizer third prefers an ICE match with a closest position to a position of the lookup segment within the asset.

13. The computer system of claim 12, wherein the ICE match prioritizer allows a user to select the ICE match based on the prioritization.

14. The computer system of claim 9, wherein the lookup segment includes a plurality of lookup segments that are substantially identical in terms of content, and

wherein the ICE match determinator determines an ICE match for each lookup segment.

15. The computer system of claim 14, wherein at least one lookup segment has a different ICE match than at least one other lookup segment.

16. The computer system of claim 9, further comprising a segment retriever that allows retrieval of at least one source text based on the matching level.

35. The computer system of claim 12, wherein the preceding usage context hash code is generated from a text stream that precedes the segment to be translated and the post usage context hash code is generated from a text stream that follows the segment to be translated.

17. A program product stored on a non-transitory computer readable medium, which when executed by a computer system, determines a matching level of a plurality of source texts stored in a translation memory to a lookup segment to be translated, the non-transitory computer readable medium comprising program code for performing the following:

determining any exact matches for the lookup segment in the plurality of source texts;

determining, in the case that at least one exact match is determined, that a respective exact match is an in-context exact (ICE) match for the lookup segment in the case that a context of the lookup segment matches that of the respective exact match, wherein the context includes at least two levels, wherein the at least two levels includes a usage context level and an asset context level, the usage context level including a preceding usage context level and a post usage context level, the preceding usage context level having a preceding usage context hash code associated therewith and the post usage context level having a post usage context hash code associated therewith; and

determining, in the case that greater than one ICE match is determined, a prioritization for each ICE match that is based on a degree of context matching, wherein a more appropriate ICE match is prioritized over one or more other ICE matches, wherein the prioritization includes;

first preferring an ICE match having a full usage context match with the lookup segment over an ICE match having only a partial usage context match with the lookup segment, wherein the full usage context match with the lookup segment includes a match with both the preceding usage context hash code and the post usage context hash code and the partial usage context match with the lookup segment includes a match with only one of the preceding usage context hash code and the post usage context hash code,

where if the first preferring step is non-conclusive, second preferring an ICE match from the same asset as the lookup segment over an ICE match from a different asset.

18. The program product of claim17, wherein the ICE match determining indicates that a respective exact match is an ICE match for the lookup segment only in the case that each context level of the lookup segment matches that of the respective exact match.

19. The program product of claim 17, wherein the ICE match determining indicates that a respective exact match is an ICE match for the lookup segment only in the case that both the usage context level and the asset context level of the lookup segment matches that of the respective exact match.

20. The program product of claim 17, wherein if the second preferring is non-conclusive, third preferring an ICE match with a closest position to a position of the lookup segment within the asset.

21. The program product of claim 20, further comprising allowing a user to select the ICE match based on the prioritization.

22. The program product of claim 17, wherein the lookup segment includes a plurality of lookup segments that are substantially identical in terms of content, and

wherein the ICE match determining includes determining an ICE match for each lookup segment.

23. The program product of claim 22, wherein at least one lookup segment has a different ICE match than at least one other lookup segment.

24. The program product of claim 17, further comprising allowing retrieval of at least one source text based on the matching level.

36. The program product of claim 20, wherein the preceding usage context hash code is generated from a text stream that precedes the segment to be translated and the post usage context hash code is generated from a text stream that follows the segment to be translated.

25. A method, performed on a computer system, for determining a matching level of a plurality of translation pairs of source text and target text stored in a translation memory to a lookup segment to be translated, the method comprising:

using the computer system to perform the following;

assigning a context to each of the plurality of translation pairs stored in the translation memory, wherein the context includes a usage context level and an asset context level, the usage context level including a preceding usage context level and a post usage context level, the preceding usage context level having a preceding usage context hash code associated therewith and the post usage context level having a post usage context hash code associated therewith;

storing the context with the translation pairs in the translation memory;

determining any exact matches for the lookup segment with the plurality of translation pairs of source text and target text stored in the translation memory;

determining, in the case that at least one exact match is determined, that a respective exact match is an in-context exact (ICE) match for the lookup segment in the case that a context of the lookup segment matches that of the respective exact match; and

determining, in the case that greater than one ICE match is determined, a prioritization for each ICE match that is based on a degree of context matching, wherein a more appropriate ICE match is prioritized over one or more other ICE matches, wherein the prioritization includes;

first preferring an ICE match having a full usage context match with the lookup segment over an ICE match having only a partial usage context match with the lookup segment, wherein the full usage context match with the lookup segment includes a match with both the preceding usage context hash code and the post usage context hash code and the partial usage context match with the lookup segment includes a match with only one of the preceding usage context hash code and the post usage context hash code,

where if the first preferring step is non-conclusive, second preferring an ICE match from the same asset as the lookup segment over an ICE match from a different asset.

31. The method of claim 25, wherein the preceding usage context hash code is generated from a hash algorithm that operates on a text stream that precedes the source text and the post usage context (UC) hash code is generated from the hash algorithm that operates on a text stream that follows the source text, and wherein the asset context level comprises an asset code that identifies an asset for the source text.

26. A computer system for determining a matching level of a plurality of translation pairs of source text and target text stored in a translation memory to a lookup segment to be translated, the computer system comprising:

at least one processing unit;

memory operably associated with the at least one processing unit;

an in-context exact (ICE) match translation system storable in memory and executable by the at least one processing unit, the ICE match translation system comprising;

a segment identifier assigner that assigns a context to each of the plurality of translation pairs stored in the translation memory, wherein the context includes a usage context level and an asset context level, the usage context level including a preceding usage context level and a post usage context level, the preceding usage context level having a preceding usage context hash code associated therewith and the post usage context level having a post usage context hash code associated therewith;

a translation memory generator that stores the context with each of the plurality of translation pairs in the translation memory;

an exact match determinator that determines any exact matches for the lookup segment in the plurality of translation pairs of source text and target stored in the translation memory;

an ICE match determinator that determines in the case that at least one exact match is determined, that a respective exact match is an ICE match for the lookup segment in the case that a context of the lookup segment matches that of the respective exact match; and

an ICE match prioritizer that in the case that greater than one ICE match is determined by the ICE match determinator, prioritizes each ICE match based on a degree of context matching, wherein the prioritization includes;

first preferring an ICE match having a full usage context match with the lookup segment over an ICE match having only a partial usage context match with the lookup segment, wherein the full usage context match with the lookup segment includes a match with both the preceding usage context hash code and the post usage context hash code and the partial usage context match with the lookup segment includes a match with only one of the preceding usage context hash code and the post usage context hash code,

where if the first preferring step is non-conclusive, second preferring an ICE match from the same asset as the lookup segment over an ICE match from a different asset.

32. The computer system of claim 26, wherein the preceding usage context hash code is generated from a hash algorithm that operates on a text stream that precedes the source text and the post usage context hash code is generated from the hash algorithm that operates on a text stream that follows the source text, and wherein the asset context level comprises an asset code that identifies an asset for the source text.

27. A program product stored on a non-transitory computer readable medium, which when executed by a computer system, determines a matching level of a plurality of translated text segments stored in a translation memory to a lookup segment to be translated, the non-transitory computer readable medium comprising program code for performing the following:

assigning a context to each of the plurality of translated text segments stored in the translation memory, wherein the context includes a usage context level and an asset context level, wherein the usage context level comprises a preceding usage context hash code generated from a text stream that precedes a segment of translated text and a post usage context hash code generated from a text stream that follows the translated text segment, and wherein the asset context level comprises an asset code that identifies an asset for the translated text segment;

storing the context with the translated text segments in the translation memory;

determining any exact matches for the lookup segment with the plurality of translated text segments stored in the translation memory;

determining, in the case that at least one exact match is determined, that a respective exact match is an in-context exact (ICE) match for the lookup segment in the case that a context of the lookup segment matches that of the respective exact match; and

determining, in the case that greater than one ICE match is determined, a prioritization for each ICE match that is based on a degree of context matching, wherein a more appropriate ICE match is prioritized over one or more other ICE matches, wherein the prioritization includes;

first preferring an ICE match having a full usage context match with the lookup segment over an ICE match having only a partial usage context match with the lookup segment, wherein the full usage context match with the lookup segment includes a match with both the preceding usage context hash code and the post usage context hash code and the partial usage context match with the lookup segment includes a match with only one of the preceding usage context hash code and the post usage context hash code,

where if the first preferring step is non-conclusive, second preferring an ICE match from the same asset as the lookup segment over an ICE match from a different asset.

28. A computer system, comprising:

a translation memory comprising a plurality of source texts for comparison to a lookup segment; and

a context identifier for each of the plurality of source texts, wherein the context identifier identifies a usage context level and an asset context level, the usage context level including a preceding usage context level and a post usage context level, the preceding usage context level having a preceding usage context hash code associated therewith and the post usage context level having a post usage context hash code associated therewith;

at least one processing unit;

memory operably associated with the at least one processing unit;

an in-context exact (ICE) match translation system storable in memory and executable by the at least one processing unit, the ICE match translation system determining a matching level of the plurality of source texts stored in the translation memory to a lookup segment to be translated, the ICE match translation system comprising an exact match determinator that determines any exact matches for the lookup segment in the plurality of source texts;

an ICE match determinator that determines in the case that at least one exact match is determined, that a respective exact match is an ICE match for the lookup segment in the case that a context of the lookup segment matches that of the respective exact match; and

an ICE match prioritizer that in the case that greater than one ICE match is determined by the ICE match determinator, prioritizes each ICE match based on a degree of context matching, wherein a more appropriate ICE match is prioritized over one or more other ICE matches, wherein the prioritizing includes;

first preferring an ICE match having a full usage context match with the lookup segment over an ICE match having only a partial usage context match with the lookup segment, wherein the full usage context match with the lookup segment includes a match with both the preceding usage context hash code and the post usage context hash code and the partial usage context match with the lookup segment includes a match with only one of the preceding usage context hash code and the post usage context hash code,

where if the first preferring step is non-conclusive, second preferring an ICE match from the same asset as the lookup segment over an ICE match from a different asset.

33. The computer system of claim 28, wherein the preceding usage context hash code is generated from a hash algorithm that operates on a text stream that precedes the source text and the post usage context hash code is generated from the hash algorithm that operates on a text stream that follows the source text, and wherein the asset context level comprises an asset code that identifies an asset for the source text.

29. A client-side computer system for interacting with a translation system including a translation memory having a plurality of source texts, the computer system comprising:

at least one processing unit;

memory operably associated with the at least one processing unit;

an in-context exact (ICE) match translation system storable in memory and executable by the at least one processing unit, the ICE match translation system comprising;

a segment identifier assigner that assigns a segment identifier to a segment to be translated by the translation system, the segment identifier indicating a usage context of the segment, wherein the usage context comprises a preceding usage context hash code generated from a text stream that precedes the segment to be translated and a post usage context hash code generated from a text stream that follows the segment to be translated;

a communicator that communicates the segment identifier assignment for storage as part of the translation memory;

an exact match determinator that determines any exact matches in the plurality of source texts with the segment to be translated;

an ICE match determinator that determines in the case that at least one exact match is determined, that a respective exact match is an ICE match for the segment in the case that a context of the segment matches that of the respective exact match, wherein the context includes at least two levels, wherein the at least two levels includes a usage context level and an asset context level; and

an ICE match prioritizer that in the case that greater than one ICE match is determined by the ICE match determinator, prioritizes each ICE match based on a degree of context matching, wherein a more appropriate ICE match is prioritized over one or more other ICE matches, wherein the prioritization includes;

first preferring an ICE match having a full usage context match with the lookup segment over an ICE match having only a partial usage context match with the lookup segment, wherein the full usage context match with the lookup segment includes a match with both the preceding usage context hash code and the post usage context hash code and the partial usage context match with the lookup segment includes a match with only one of the preceding usage context hash code and the post usage context hash code,

where if the first preferring step is non-conclusive, second preferring an ICE match from the same asset as the lookup segment over an ICE match from a different asset.

30. A program product stored on a non-transitory computer readable medium, which when executed by a computer system, interacts with a translation system including a translation memory having a plurality of source texts, the non-transitory computer readable medium comprising program code for performing the following:

assigning a segment identifier to a segment to be translated by the translation system, the segment identifier indicating a usage context of the segment, wherein the usage context comprises a preceding usage context hash code generated from a text stream that precedes the segment to be translated and a post usage context hash code generated from a text stream that follows the segment to be translated;

communicating the segment identifier assignment for storage as part of the translation memory;

determining any exact matches for the segment with the plurality of source texts stored in the translation memory;

determining, in the case that at least one exact match is determined, that a respective exact match is an in-context exact (ICE) match for the segment in the case that a context of the segment matches that of the respective exact match, wherein the context includes at least two levels, wherein the at least two levels includes a usage context level and an asset context level, the usage context level including a preceding usage context level and a post usage context level, the preceding usage context level having the preceding usage context hash code associated therewith and the post usage context level having the post usage context hash code associated therewith; and

determining, in the case that greater than one ICE match is determined, a prioritization for each ICE match that is based on a degree of context matching, wherein a more appropriate ICE match is prioritized over one or more other ICE matches, wherein the prioritization includes;

first preferring an ICE match having a full usage context match with the lookup segment over an ICE match having only a partial usage context match with the lookup segment, wherein the full usage context match with the lookup segment includes a match with both the preceding usage context hash code and the post usage context hash code and the partial usage context match with the lookup segment includes a match with only one of the preceding usage context hash code and the post usage context hash code,

where if the first preferring step is non-conclusive, second preferring an ICE match from the same asset as the lookup segment over an ICE match from a different asset.