Enterprise Record Matching With Sequential Identifier Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing record matching methods are computationally intensive and require significant data manipulation, which can violate data ownership rights and compromise data integrity, making them inefficient for large-scale data sets.
Innovation Solution
The Enterprise Matching Service (EMS) uses unique identifiers and match codes to identify potential matches through a series of sequential strategies, followed by a 'sanity check' to confirm matches, minimizing CPU-intensive activities and allowing for flexible algorithm adjustments.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional record matching methods are used to match large volumes of data records, then matching accuracy can be maintained, but server efficiency deteriorates and computational time increases significantly
Solution Approach 1:
The patent segments the record matching process into distinct phases: generating unique identifiers for incoming records, creating match codes from record attributes, comparing unique identifiers first (coarse filtering), and then comparing match codes only for records with matching unique identifiers (fine filtering). This segmentation reduces the computational burden by dividing the matching task into hierarchical stages, improving server efficiency while maintaining matching accuracy.
Solution Approach 2:
The patent performs preliminary actions by pre-generating unique identifiers and match codes for all records in the data warehouse before matching operations. This preliminary preparation allows the matching system to quickly compare pre-computed values rather than performing complex attribute comparisons for every record pair, significantly improving processing efficiency while preserving matching accuracy.
2Measurement precision
If data manipulation and cleansing are performed on incoming records before matching, then matching accuracy improves, but data integrity may be compromised and data ownership rights may be violated
Solution Approach 1:
The patent extracts only the essential identifying attributes from records to generate unique identifiers and match codes, rather than manipulating or cleansing the entire record set. This extraction approach maintains matching accuracy by focusing on key identifying features while preserving the integrity and ownership of the original data, as the full records remain unchanged and under their original ownership.
Solution Approach 2:
The patent performs preliminary generation of unique identifiers and match codes from the original records without manipulating the records themselves. This preliminary extraction and coding process enables accurate matching while maintaining data integrity, as the original records are never altered and remain under their rightful ownership.
3Measurement precision
If multiple elements of data are compared against the entire database to identify matches, then matching completeness improves, but computational load and processing time increase significantly
Solution Approach 1:
The patent segments the comparison process into hierarchical stages: first comparing unique identifiers (which are generated from all record attributes to ensure completeness), and then comparing match codes only for records with matching unique identifiers. This segmentation ensures matching completeness by systematically evaluating all attributes through the unique identifier generation process, while reducing processing time by limiting detailed attribute comparisons to only potentially matching record pairs.
Solution Approach 2:
The patent performs preliminary generation of unique identifiers that incorporate information from multiple record elements before the matching process begins. This preliminary action ensures that the essence of all record attributes is captured in the unique identifier, allowing for complete matching evaluation without requiring multiple sequential comparisons against the entire database, thus reducing processing time while maintaining matching completeness.
Data Source
AI summary
Methods and systems for an improved Enterprise Matching Service (“EMS”) that is designed to match incoming data records to a database of records, using less system resources and using those resources more efficiently. The EMS identifies potential matches by generating unique identifiers and match codes for incoming data records, and then, using a series of sequential matching strategies, matching the unique identifiers and match codes to pre-generated unique identifiers and match codes for database records to identify a subset of potential matches. If no matches are identified, the next matching strategy is executed. Once a subset of potential matches are identified, the methods and systems may weigh the subset of potential matches to determine whether there is a match.