Value Extraction Guided Review Document Comparison
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The comparison of documents is a time-consuming and computer-intensive process, often requiring human assistance to increase accuracy and speed, especially in industries like legal, publishing, and banking, where identifying changes is crucial.
Innovation Solution
The Value Extraction and Guided Review (VEGR) system uses a search enhancing metadata imbued list to compare documents by linking adjacent words, determining x and y coordinates, and employing algorithms like Jaro Winkler Distance and Dice Coefficient to efficiently extract relevant values and guide users to content of interest, reducing processing time and improving efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If document comparison is performed using traditional methods, then accuracy can be maintained through human review, but processing time and computational resources are excessive
Solution Approach 1:
The patent segments the document comparison process into distinct phases: OCR processing of individual pages, extraction of specific data fields (policy numbers, effective dates, limits), and structured comparison of extracted values. This segmentation allows parallel processing of multiple documents and reduces the computational burden of comparing entire documents as single units, thereby improving processing speed while maintaining accuracy through focused field-by-field verification
Solution Approach 2:
The system performs preliminary OCR processing and data extraction on all documents before the comparison phase. By pre-processing documents to extract and structure key information (policyholder names, coverage limits, dates) in advance, the system eliminates the need for real-time processing during comparison, significantly reducing processing time while ensuring accurate comparison of standardized fields
2Productivity
If automated extraction methods are used, then processing efficiency increases, but accuracy may decrease without human verification
Solution Approach 1:
The patent implements a feedback mechanism where extracted values are automatically verified against expected formats and ranges (e.g., validating that policy numbers match patterns, dates are in correct format, numerical values are positive). The system provides confidence scores for each extraction and automatically flags low-confidence results for manual review, creating a feedback loop that maintains high accuracy while preserving automated processing efficiency for high-confidence extractions
Solution Approach 2:
The system performs self-verification of extracted data by cross-checking consistency across multiple documents (e.g., verifying that policy numbers appear consistently across policy pages, comparing effective dates against renewal dates). This self-service validation mechanism catches obvious errors automatically without requiring human intervention for every extraction, maintaining reliability while maximizing automated processing
3Loss of information
If all words in documents are processed for comparison, then comprehensive analysis is achieved, but computational complexity increases significantly
Solution Approach 1:
The patent extracts only the specific information fields relevant to policy comparison (policy numbers, effective dates, coverage limits, premium amounts) from the full document text, discarding unnecessary content. This selective extraction reduces the data volume for comparison from entire document texts to structured key-value pairs, dramatically lowering computational complexity while preserving all information necessary for policy verification and change detection
Solution Approach 2:
The system applies different processing quality levels to different parts of the document: high-precision OCR and extraction for critical fields (policy numbers, dates, monetary values) and lower-precision processing for less critical content. This local quality approach optimizes computational resources by focusing detailed analysis only where needed for accurate policy comparison, reducing overall complexity while maintaining information completeness for decision-critical data
Data Source
AI summary
A method of automatic extraction of values includes scanning a document and linking a first word of a plurality of words in the document, to each word of the plurality of words in the document positionally adjacent to the right, left, top, bottom of the first word. The method further includes repeating said scanning and linking for every word of the plurality of words in the document. The method further includes determining the x and y coordinate in the document for each of the plurality of words. The method further includes providing a plurality of checklist words corresponding to information of interest. The method further includes searching the plurality of words in the document for each of the plurality of checklist words. The method further includes for each checklist word found in the plurality of words of the documents determining if a linked positionally adjacent words to the right, left, top, bottom is a value that matches typical values for the plurality of checklist words and if so, extract said value. The method further includes directing a user interface to position in the document related to content of interest related to at least one of the plurality of checklist words.