Enhanced Name Key Matching for Reducing False Positives
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing name matching systems face challenges with false positives and false negatives due to information loss during key generation, and they fail to account for name variants and typographical errors, leading to inefficient comparisons and reduced accuracy.
Innovation Solution
The method generates enhanced name keys using digraph bitmap signatures and variant codes, which consider cultural provenance and pseudo-phonetic versions, allowing for similarity scoring and reducing false positives and negatives by comparing names based on spelling and pronunciation similarities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional name key generation is used to reduce comparison time, then processing speed improves, but information loss occurs leading to false positives and false negatives
Solution Approach 1:
The name key is segmented into multiple independent components: digraph bitmap signature (capturing character pairs), variant code (capturing cultural spelling variations), and pseudo-phonetic signature (capturing pronunciation). Each component addresses specific aspects of name similarity without requiring full name comparison, enabling efficient filtering while preserving matching information.
Solution Approach 2:
The name key uses a composite structure combining multiple types of signatures (digraph bitmap, variant code, pseudo-phonetic) that together provide comprehensive name representation. This composite approach captures both spelling and pronunciation characteristics, reducing information loss while maintaining fast comparison capability.
2Measurement precision
If strict exact matching is used for name comparison, then false positives are reduced, but false negatives increase due to inability to handle variants and typos
Solution Approach 1:
The system changes the matching parameters by introducing fuzzy matching capabilities through variant codes (for cultural spelling variations) and pseudo-phonetic signatures (for pronunciation-based matching). This allows the system to tolerate typos and variants while maintaining high matching accuracy through multi-parameter comparison.
Solution Approach 2:
The patent introduces intermediary components (variant codes and pseudo-phonetic signatures) that mediate between exact spelling matching and fuzzy matching. These intermediaries enable the system to handle cultural variants and typos by providing alternative representation forms that bridge the gap between different name spellings.
3Measurement precision
If comprehensive name comparison is performed to improve accuracy, then matching precision improves, but processing time increases significantly
Solution Approach 1:
The comparison process is segmented into multiple stages: first comparing digraph bitmap signatures for quick filtering, then variant codes for cultural matching, and finally pseudo-phonetic signatures for pronunciation-based matching. This segmented approach enables comprehensive comparison while reducing overall processing time through early elimination of non-matching candidates.
Solution Approach 2:
The system performs preliminary comparison using digraph bitmap signatures and variant codes before conducting full name comparison. This preliminary action filters out clearly non-matching names early in the process, reducing the number of names requiring comprehensive comparison and thus reducing overall processing time.
Data Source
AI summary
Name matching using enhanced name keys is provided by receiving and parsing a queried name into name phrase(s), building a name key for the queried name, the name key for identifying matches between the queried name and candidate names in a database, and the name key including name phrase digraph bitmap signature(s) for the queried name, variant code(s) for the queried name, and pseudo-phonetic name phrase digraph bitmap signature(s) for the queried name, and performing a name matching comparison that includes comparing the queried name to each candidate name of the candidate names in the database, in which the built name key for the queried name is compared to a name key for the candidate name.