Automated Translator Selection via Profile Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing machine translation programs are inadequate for translating difficult or less-studied languages, particularly for formal and informal communications, and human translators cannot ensure constant availability and quick response for internet applications.
Innovation Solution
A data-driven automated system selects profiles of translation professionals by matching subject area expertise, using machine learning for quality evaluation and workflow planning to optimize translation resources and workflow.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If machine translation programs are used, then translation speed and availability are improved, but translation accuracy for difficult or less-studied languages deteriorates
Solution Approach 1:
The patent introduces a human-in-the-loop intermediary system where machine translation outputs are reviewed and corrected by human translators. The system automatically selects appropriate human translators based on subject matter expertise and uses their corrections to improve future machine translations, thereby maintaining high speed while improving accuracy for difficult languages.
Solution Approach 2:
The translation process is segmented into multiple stages: initial machine translation, quality assessment, selective human review based on subject matter expertise, and correction. This segmentation allows the system to maintain high throughput for straightforward translations while applying human expertise only where needed, resolving the contradiction between speed and accuracy.
2Reliability
If human translators are used, then translation accuracy for difficult languages is improved, but availability and response time deteriorate
Solution Approach 1:
The system applies human translation effort partially - only for segments or documents that require it based on subject matter analysis and quality assessment. Most translations proceed through machine translation alone, while human translators are engaged only when needed, maintaining both accuracy and availability.
Solution Approach 2:
The system performs preliminary analysis of the translation request to assess subject matter expertise requirements and estimated quality needs before engaging human translators. This preliminary action allows the system to prepare appropriate human translator selections in advance, ensuring both accuracy and rapid response time.
3Measurement precision
If manual parameter adjustment is used for translator selection, then matching accuracy is improved, but system complexity and operational burden increase
Solution Approach 1:
The system performs self-service by automatically analyzing translation requests, identifying subject matter requirements, selecting appropriate human translators based on their profiles and expertise, and evaluating translation quality without manual intervention. This automation maintains high matching accuracy while eliminating operational complexity.
Solution Approach 2:
The system implements feedback loops where translation outcomes are automatically evaluated and used to refine future translator selections and machine translation models. This automated feedback mechanism improves matching accuracy over time without requiring manual parameter adjustment, reducing system complexity.
Data Source
AI summary
A method for translating a current electronic document is disclosed that includes storing previous translations of prior electronic documents for profiles of translation professionals, extracting terms from prior electronic documents, and generating glossaries that are each associated with one of the profiles and include a respective subset of terms. The method also includes receiving a request to translate the current electronic document, selecting one or more of the profiles based on proximity of the respective subset of terms to extracted terms of the current electronic document, evaluating qualities of the previous translations for each of the selected profiles, planning a workflow for translation of the current electronic document based on the selected profiles, and causing the current electronic document to be translated according to the planned workflow.