Document Attribute Extraction via Dynamic Template Scoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for extracting attributes from paper-based documents using OCR technology face challenges when dealing with documents having similar attributes but different contents, leading to low extraction accuracy due to the need for pre-set numeric identifiers and templates, which can be inconsistent or change format.
Innovation Solution
A computer system that uses a processor and storage device to execute character recognition, calculate scores based on template and dictionary information, and select the template with the highest extraction accuracy for attribute extraction, allowing for high-accuracy attribute extraction from paper-based documents.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If pre-set numeric identifiers and templates are used for attribute extraction, then work efficiency is improved, but extraction accuracy deteriorates when documents have similar attributes but different contents
Solution Approach 1:
The patent applies dynamics by making the template selection process adaptive rather than static. The system dynamically selects templates based on calculated scores that evaluate how well each template matches the actual document characteristics. This allows the system to adapt to different document formats and content variations while maintaining high extraction accuracy, resolving the contradiction between using pre-set templates for efficiency and achieving accuracy for varying document types.
Solution Approach 2:
The patent changes the parameter of template selection from fixed pre-assignment to dynamic scoring-based selection. By introducing a score calculation mechanism that evaluates template-document compatibility, the system can adjust which template is applied based on the specific document being processed. This parameter change enables both efficient automated processing and accurate extraction even when documents have similar attributes but different contents.
2Speed
If a fixed template is used for all documents, then processing speed is improved, but adaptability to different document formats deteriorates
Solution Approach 1:
The patent implements universality by creating a template selection system that can handle multiple document formats and types through a unified scoring mechanism. Rather than requiring separate fixed templates for each document type, the system uses a universal score calculation approach that evaluates any template against any document, enabling the same processing framework to adapt to various document formats while maintaining processing speed.
Solution Approach 2:
The system introduces dynamic template selection based on calculated scores, allowing the template application process to adapt to different document formats. The scoring mechanism dynamically determines the best match between templates and documents, providing versatility across different formats while maintaining the efficiency of automated template-based extraction.
3Measurement precision
If manual verification of paper-based documents is performed, then extraction accuracy is improved, but work efficiency deteriorates
Solution Approach 1:
The patent applies self-service by enabling the system to automatically evaluate and select the most appropriate template for each document through score calculation. This self-service template selection mechanism eliminates the need for manual verification while maintaining high extraction accuracy, as the system autonomously determines the best template match based on document characteristics, thereby improving work efficiency without sacrificing accuracy.
Solution Approach 2:
The system implements feedback through the score calculation mechanism that evaluates template-document compatibility. This feedback loop allows the system to automatically adjust template selection based on how well each template matches the document's actual characteristics, achieving accurate extraction without manual verification and thus improving productivity while maintaining precision.
Data Source
AI summary
A computer, which is configured to extract an attribute being a character string indicating a feature of a paper-based document, the computer stores template information dictionary information. The computer is configured to: execute character recognition processing on image data on the paper-based document; extract an attribute corresponding to each of the at least one type of attribute, which is defined in each of the plurality of templates, through use of a result of the character recognition processing and the plurality of templates; calculate a score regarding the extracted attribute for each of the plurality of templates; select one of the plurality of templates that has the highest extraction accuracy of the attribute based on the score; and generate output information through use of the selected template.