Financial Data Error Correction via Levenshtein Distance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Character-recognition techniques, such as OCR and ICR, introduce errors in financial information extraction, complicating and increasing the expense of subsequent processing due to their imperfect conversion of data formats.
Innovation Solution
A computer system that identifies and corrects errors in financial information by comparing received data with expected information using error metrics like Levenshtein edit distance, assigning weights to different error types, and adapting based on feedback to improve accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If character-recognition software (OCR/ICR) is used to extract financial information, then data conversion from bitmap to ASCII format is achieved, but errors are introduced in the extracted information
Solution Approach 1:
The system uses feedback from comparing extracted financial information against expected values (from financial histories, vendor lists, customer databases) to identify and correct errors. Error metrics are calculated and fed back into the correction process, allowing the system to learn from mistakes and improve accuracy over time while maintaining high extraction efficiency
Solution Approach 2:
The patent introduces an intermediary error-correction module that sits between the character-recognition software and subsequent processing systems. This intermediary compares extracted data against known reference data (vendor lists, customer databases, financial histories) and corrects errors before data is passed to accounting or financial software, thus protecting downstream systems from erroneous data while maintaining the efficiency of OCR/ICR extraction
2Reliability
If error correction processes are implemented to improve accuracy, then reliability of financial information increases, but processing complexity and expense increase
Solution Approach 1:
The system performs preliminary actions by pre-loading and storing expected financial information in databases (vendor lists, customer databases, financial histories) before the actual extraction process. This allows the error-correction module to quickly compare extracted data against known values without adding complex real-time processing requirements, thus improving reliability without significantly increasing system complexity
Solution Approach 2:
The patent changes parameters by using configurable error metrics and correction thresholds that can be adjusted based on the specific application. Instead of implementing a single complex correction algorithm, the system uses multiple simple comparison operations with adjustable parameters (error tolerance levels, data source priorities), making the system adaptable and easier to implement while maintaining high accuracy
3Speed
If character-recognition software is used for financial information extraction, then information extraction speed is improved, but subsequent processing becomes more expensive and complicated due to errors
Solution Approach 1:
The error-correction module operates autonomously using self-service principles. It automatically compares extracted financial information against expected values from stored databases, identifies discrepancies, and corrects errors without requiring manual intervention. This self-service approach maintains the high extraction speed of OCR/ICR while eliminating the need for expensive manual verification processes
Solution Approach 2:
The patent extracts only the essential comparison operations needed for error detection, separating the error-correction function from the main extraction process. By extracting and comparing only critical fields (vendor names, customer IDs, transaction amounts) against reference databases, the system maintains fast extraction speeds while minimizing the processing overhead required for error correction, thus reducing overall processing expenses
Data Source
AI summary
Embodiments of a computer system, a method, and a computer-program product (e.g., software) for use with the computer system are described. These embodiments may be used to identify and correct errors in financial information that was extracted using character-recognition software, such as optical character recognition software and/or intelligent character recognition software. In particular, potential errors may be identified by comparing the financial information for a current financial transaction of a user with expected financial information from one or more previous financial transactions of the user. Error metrics for these potential errors may be determined and used to correct at least some of the potential errors. For example, values of the Levenshtein edit distance may be determined based on the comparison, and one or more potential errors associated with one or more minimum values of the Levenshtein edit distance may be corrected.