Address Database Reconciliation Using Progressive Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Reconciling large databases to update address records is a time-consuming and difficult process due to the complexity of matching and comparing address data between related databases.
Innovation Solution
A method involving a 'tight to loose' architecture in the address matching module, which compares address records in an update database to a production database, using multiple matching algorithms to progressively relax match criteria, allowing for the identification of matching records that initial comparisons might miss, and employing ZIP-level and Postal Finance Number-level matching processes to efficiently manage updates and deletions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a single comparison pass is used to match addresses between databases, then the processing speed is faster, but the matching accuracy decreases and records are missed
Solution Approach 1:
The address matching process is divided into multiple sequential passes with progressively relaxed matching criteria. The first pass uses strict matching rules to identify obvious matches, while subsequent passes use progressively looser rules to capture additional matches that were missed by earlier passes, thereby improving overall matching accuracy without requiring a single overly complex comparison algorithm
Solution Approach 2:
The system performs preliminary filtering and indexing of address records before the actual matching process. Addresses are pre-processed to extract key components and create lookup structures, which enables the multi-pass comparison to proceed efficiently by quickly eliminating non-matching records in each subsequent pass
2Reliability
If multiple comparison passes with relaxed criteria are used, then more addresses are matched accurately, but the processing time increases
Solution Approach 1:
The matching criteria dynamically evolve across comparison passes, transitioning from strict to progressively relaxed rules. Each pass adapts its matching threshold based on the results of previous passes, allowing the system to capture increasingly subtle matches while maintaining control over the overall processing workload and time requirements
Solution Approach 2:
The system performs more comparison passes than a single-pass system would, applying partially relaxed criteria in intermediate passes. This excessive action ensures that even difficult-to-match addresses are captured, with the understanding that the additional passes are necessary to achieve the desired level of reliability in the address update process
3Measurement precision
If strict matching criteria are applied, then false positives are reduced, but legitimate matches are missed
Solution Approach 1:
The system changes the matching parameters across different passes, transitioning from strict parameter requirements in early passes to more flexible parameter acceptance in later passes. This allows the system to maintain high precision for obvious matches while becoming increasingly adaptable to capture legitimate but subtle matches in subsequent passes
Data Source
AI summary
Methods for reconciling an address database include comparing addresses in an update database to addresses in a production database. A second comparison of unique addresses from the first comparison is executed, the second comparison allowing matching of addresses that the first comparison did not match.