Automatic Target Language Extraction Rule Generation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing natural language processing systems require excessive time and labor to create extraction rules for multiple languages due to the need for language experts to summarize rules for each language, which is inefficient and labor-intensive.
Innovation Solution
A method that uses pairs of translation sentences to automatically generate extraction rules for a target language based on an existing source language rule, reducing the need for manual expert summarization by leveraging word alignment relationships and domain-specific word combinations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If extraction rules are manually summarized by language experts for each natural language, then the accuracy and reliability of key information extraction is improved, but the time consumption and labor costs increase excessively
Solution Approach 1:
The patent copies the extraction rule from the source language and applies it to the target language through machine translation. Instead of manually creating new extraction rules for each target language, the system translates the source language extraction rule into the target language, thereby reducing time and labor consumption while maintaining extraction accuracy across multiple languages
Solution Approach 2:
The patent creates a universal extraction rule generation mechanism that can handle multiple natural languages through a single source language rule. The machine translation component enables the extraction rule to be universally applied across different language pairs, eliminating the need for separate manual rule creation for each language
2Adaptability or versatility
If multiple language experts are engaged to create extraction rules for different natural languages, then the quality and adaptability of language processing is improved, but the device complexity and operational difficulty increase
Solution Approach 1:
The patent introduces machine translation as an intermediary between the source language extraction rule and the target language processing. This intermediary automatically translates the extraction rule into the target language, eliminating the need for multiple language experts and reducing system complexity while maintaining multi-language adaptability
3Manufacturing precision
If extraction rules are manually created for each natural language to ensure precision, then the manufacturing precision of language processing is improved, but the productivity decreases due to excessive time and labor requirements
Solution Approach 1:
The system copies the validated extraction rule from the source language and translates it to target languages, ensuring that the precision of the original rule is maintained while dramatically improving productivity by eliminating repetitive manual rule creation for each language
Solution Approach 2:
The patent changes the parameter of extraction rule creation from manual expert summarization to automated machine translation. This parameter change maintains the precision of extraction rules while significantly improving productivity by reducing the time and labor required for rule creation across multiple languages
Data Source
AI summary
A method includes obtaining n pairs of translation sentences of a source language and a target language, where each of the n pairs of translation sentences includes a source language sentence and a target language sentence that are translations of each other, extracting a source language segment from each source language sentence in the n pairs of translation sentences using an extraction rule of the source language, extracting a target language segment from each target language sentence in the n pairs of translation sentences, and generating an extraction rule of the target language based on n target language segments extracted from n target language sentences.