Entity Name Translation Using Publication Date and Co-occurrence
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current machine translation systems often fail to accurately translate entity names such as persons, locations, and organizations across languages with different scripts, leading to ambiguity and reduced understanding of translated documents.
Innovation Solution
The implementation of a method that involves receiving data records with entity names in a source language, generating candidate translations using a trained sequence alignment model, and querying a data repository to select the most accurate translation based on properties and context, employing techniques like character-level transliteration and normalization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional machine translation systems translate entity names using standard translation methods, then the translation process is simple and fast, but the translation accuracy is low and entity names become ambiguous
Solution Approach 1:
The system segments the translation process into distinct stages: entity name identification, candidate translation generation through multiple methods (transliteration, dictionary lookup, context-based translation), and selection based on property comparison. This segmentation allows each stage to be optimized independently, improving overall translation accuracy while maintaining manageable system complexity
Solution Approach 2:
The system introduces an intermediary data repository that stores properties of entity names across different languages. This repository acts as a mediator between the source text and target translation, enabling accurate entity name translation by comparing properties rather than relying on traditional direct translation methods
2Measurement precision
If multiple candidate translations are generated and compared using property matching, then translation accuracy improves, but the processing time and computational resources increase
Solution Approach 1:
The system performs preliminary actions by pre-processing source text to identify entity names before full translation, and pre-storing properties of entity names in the data repository. This allows the main translation process to focus only on entity names rather than processing entire texts, significantly reducing processing time while maintaining high accuracy
Solution Approach 2:
The system applies different translation methods and levels of processing to different parts of the text: entity names receive detailed property-based matching and multiple candidate generation, while non-entity text uses standard translation methods. This local quality approach ensures high accuracy for entity names without unnecessarily increasing processing time for the entire document
Data Source
AI summary
Methods and apparatus, including computer program products, related to entity name translation. In general, a set of data records including one or more names of one or more entities in a source language is received and candidate translations are generated for the names of the data records. The candidate translations are strings of characters in a target language that has a different orthography than an orthography of the source language. A data repository is queried for records matching the candidate translations. A translation of at least one of the candidate translations is selected based on a comparison of a combination of candidate translations and properties of the source record comprising of the publication date, entity name co-occurrence and their frequency with the result of the querying the data repository.