Fuzzy Entity Matching with User Feedback and ML Weighting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The increasing volume of electronic data makes it difficult to identify matching entities across various data sets, especially when dealing with spelling mistakes, different spellings, and multiple languages, which affects the accuracy of risk estimation and data integration.
Innovation Solution
A fuzzy matching system that applies multiple algorithms to identify potential matches, allowing user customization and feedback to improve accuracy, utilizing machine learning to update weightings and parameters, and includes interactive user interfaces for efficient human-computer interaction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple fuzzy matching algorithms are applied to identify potential matches across data sets, then the accuracy of entity matching is improved, but the complexity of the system increases
Solution Approach 1:
The system segments the entity matching problem into multiple independent fuzzy matching algorithms (e.g., string similarity, phonetic matching, token-based comparison). Each algorithm handles specific aspects of data variation independently, allowing the system to achieve high accuracy through combination while maintaining manageable complexity through modular design.
Solution Approach 2:
The patent creates a universal matching system that integrates multiple fuzzy matching algorithms into a single platform capable of handling various data formats, languages, and spelling variations. This multi-functional system resolves the contradiction by providing comprehensive matching capabilities through a unified interface rather than separate specialized tools.
2Measurement precision
If user feedback is incorporated to update weightings and parameters of matching algorithms, then the accuracy of risk estimation is improved, but the time required for processing increases
Solution Approach 1:
The system performs preliminary matching using default algorithm weightings and parameters before incorporating user feedback. This allows initial rapid processing while setting the stage for subsequent refinement. The preliminary action reduces time loss by establishing baseline matches quickly, with feedback-based optimization occurring selectively for borderline cases.
Solution Approach 2:
The patent implements a feedback mechanism where user corrections and confirmations of matches are used to iteratively update algorithm weightings and parameters. This feedback loop improves accuracy over time by learning from actual user decisions, while the system manages processing time by applying updates progressively rather than requiring complete re-processing of all data.
3Measurement precision
If manual data review is performed to verify matches, then the accuracy of entity identification is improved, but the productivity decreases
Solution Approach 1:
The system applies partial manual review by automatically filtering and prioritizing matches that require human verification. High-confidence matches are processed automatically without manual review, while only borderline or low-confidence cases are submitted for user verification. This partial action maintains high productivity while ensuring accuracy for critical cases.
Solution Approach 2:
The fuzzy matching algorithms perform self-service by automatically identifying and resolving clear matches without requiring manual intervention. The system serves itself by using confidence scores to autonomously determine which matches need human review, thereby maximizing automated processing capacity while minimizing the burden of manual verification.
Data Source
AI summary
A fuzzy matching system matching data records in one or more data sets based on user-customized selection of multiple fuzzy matching algorithms. Possible matches may be displayed to a user, who provides feedback on the accuracy of the matches, which may then be used by a machine learning algorithm to update weightings and parameters of the multiple fuzzy matching algorithms, such as based on machine learning analysis of the matching results and the user feedback.