In-Context Exact Matching for Translation Memory Accuracy

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing translation memory systems lack effective validation of exact matches for context appropriateness, leading to potential incorrect translations when reusing previous translations without considering the new context.

Innovation Solution

A method and system that determine in-context exact (ICE) matches by evaluating context levels such as source usage context, target usage context, and structural context to prioritize and select the most appropriate exact matches for a given lookup segment, ensuring relevance and accuracy in translation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If translation memory systems reuse exact matches without context validation, then translation efficiency is improved, but translation accuracy deteriorates

Engineering Contradiction:
Improvetranslation efficiencyVSAvoidtranslation accuracy
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The system implements feedback by validating context appropriateness of exact matches before reuse. The context validation mechanism checks whether the surrounding text and semantic environment of the current lookup segment match those of the stored translation memory entry, providing feedback to determine if the exact match is appropriate for reuse, thereby maintaining both efficiency and accuracy

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The patent introduces context validation as an intermediary layer between exact match retrieval and translation reuse. This intermediary mechanism assesses contextual compatibility without preventing the efficiency benefits of exact matches, acting as a mediator that ensures accuracy while preserving productivity

Inventive Principle:
Principle #24Intermediary (Mediator)

2Adaptability or versatility

If multiple exact matches are found in translation memory, then translation options increase, but selection difficulty increases

Engineering Contradiction:
Improvetranslation optionsVSAvoidmatch selection difficulty
Core Design Contradiction:
Adaptability or versatilityVSDifficulty of detecting and measuring

Solution Approach 1:

The system changes the parameter of match evaluation by introducing context similarity as an additional criterion beyond exact text matching. By evaluating context appropriateness, semantic compatibility, and contextual relevance, the system transforms the selection process from simple exact matching to multi-parameter evaluation, enabling automated prioritization of the most appropriate match

Inventive Principle:
Principle #35Parameter changes

3Manufacturing precision

If context validation is implemented for exact matches, then translation quality is improved, but processing complexity increases

Engineering Contradiction:
Improvetranslation qualityVSAvoidsystem complexity
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The context validation process is segmented into distinct analytical components: exact match identification, context extraction, context similarity evaluation, and appropriateness determination. This segmentation allows the complex validation task to be broken down into manageable steps that can be processed systematically without overwhelming system complexity

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP2317447B1In-context exact (ICE) matching
Publication Date: 2014.01.22 SDL LTD
  • EP2317447B1 patent drawingFigure 1
  • EP2317447B1 patent drawingFigure 2A
  • EP2317447B1 patent drawingFigure 2B

AI summary

A system, a method and computer software adapted to perform the method 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.