Automated Entity Resolution with Confidence Score Feedback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Automated entity resolution processes can be inaccurate when dealing with insufficient, inaccurate, or low-quality data, leading to potential security breaches and other downstream issues, as they may not cover all possible data circumstances and do not effectively account for user feedback.
Innovation Solution
A computer system dynamically updates confidence scores for data records assigned to entities using automated entity resolution, incorporating user feedback and new data records to quantify misassignment likelihood, with visual indicators in a graphical user interface to help users identify and mitigate potential errors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If automated entity resolution is used to process data records, then processing speed and efficiency are improved, but accuracy and reliability deteriorate due to insufficient data quality and inability to account for all data circumstances
Solution Approach 1:
The system implements feedback by displaying confidence scores to users for each entity-record assignment. Users can review these scores and provide corrections or confirmations, which feed back into the system to improve future automated assignments. This closed-loop feedback mechanism allows the system to learn from user corrections and progressively improve accuracy while maintaining high processing speeds.
Solution Approach 2:
The confidence score acts as an intermediary between the automated entity resolution system and the user. Instead of directly trusting or rejecting automated assignments, users review the confidence scores as an intermediate step. This intermediary mechanism helps users quickly identify high-risk assignments that need manual verification while accepting low-risk assignments automatically, thus maintaining productivity while improving reliability.
2Reliability
If manual entity resolution is performed to verify data accuracy, then reliability and accuracy are improved, but processing time and operational complexity increase significantly
Solution Approach 1:
Instead of manually verifying all entity-record assignments, the system applies partial action by only requiring manual review for assignments with low confidence scores. High-confidence assignments are accepted automatically without manual verification. This selective approach maintains data accuracy for critical assignments while minimizing the time loss associated with manual processing.
Solution Approach 2:
The system applies different levels of verification scrutiny to different assignments based on their local characteristics (confidence scores). High-confidence assignments receive minimal scrutiny (automatic acceptance), while low-confidence assignments receive intensive scrutiny (manual review). This local quality approach optimizes the balance between accuracy and processing time by concentrating manual verification efforts where they are most needed.
3Measurement precision
If confidence scores are calculated and displayed for all entities, then identification of high-risk misassignments is improved, but system complexity and computational resources increase
Solution Approach 1:
The system extracts only the essential confidence score information needed for decision-making and displays it in a simplified format to users. Instead of presenting all possible data attributes and complex calculation details, the system extracts and presents the key metric (confidence score) that users need to quickly assess assignment quality. This extraction approach improves misassignment identification while minimizing the added system complexity.
Data Source
AI summary
One example described herein involves a system that can receive a set of data records and execute an automated entity resolution (AER) process configured to assign the set of data records to a set of entities. For each entity in the set of entities, the system can generate a respective consistency score for the entity, generate a respective confidence score for the entity based on the respective consistency score for the entity, and determine a respective visual indicator based on the respective confidence score for the entity. The respective visual indicator can indicate a risk of record misassignment to a user. The system can then generate a graphical user interface that includes the respective visual indicator for each of the entities.