Identity Resolution System with Load Order Independence
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data management systems face challenges in resolving identities due to resolution ambiguity, where the order of account resolution affects the outcome, and they fail to address cases where accounts may equally resolve to multiple entities or not at all, leading to inaccurate entity associations.
Innovation Solution
A system and method that retrieves an account, identifies ambiguity by matching account data with existing entities, unresolves entities when ambiguity arises, and creates new entities to resolve the ambiguity, ensuring a load order independent resolution outcome by relating the new entities to each other.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If existing art resolves accounts into entities based on highest score or first located, then resolution speed is improved, but resolution accuracy deteriorates due to load order dependency and inability to handle ambiguity
Solution Approach 1:
The system performs preliminary actions by collecting all candidate entities that match an account before making a resolution decision. Instead of resolving immediately upon first match, the system gathers all potential matches, evaluates them collectively, and then determines the appropriate resolution outcome. This preliminary collection and evaluation phase eliminates load order dependency while maintaining efficient resolution.
2Quantity of substance
If existing art resolves accounts with partial data matches, then resolution completeness is improved, but resolution reliability deteriorates due to ambiguous matching
Solution Approach 1:
The system implements feedback by evaluating all candidate entities against the account data and using this collective evaluation to determine resolution. The feedback mechanism considers the presence of multiple candidates and their respective match qualities, allowing the system to identify ambiguous cases where multiple entities have comparable match scores. This feedback loop enables reliable handling of partial matches by distinguishing between true ambiguities and clear winners.
3Quantity of substance
If existing art creates new entities for ambiguous accounts, then entity coverage is improved, but data management complexity increases due to unresolved ambiguity
Solution Approach 1:
The system inverts the traditional approach by not immediately creating new entities for ambiguous accounts. Instead of assuming new entities are needed for ambiguous matches, the system first evaluates whether existing entities can appropriately represent the account. The inversion occurs in the decision logic: rather than creating new entities as the default for ambiguous cases, the system explores whether existing entities with comparable match scores represent valid alternative interpretations, thereby reducing unnecessary entity proliferation.
Data Source
AI summary
A system and method for resolving identities that are indefinitely resolvable is presented. A resolution manager retrieves an account and resolves the account to existing entities. When the account is resolved into an existing entity, the resolution manager proceeds through a sequence of steps to ensure a load order independent resolution outcome by unresolving the entity, and possibly related entities, if resolution ambiguity arises. When the account does not resolve to an existing entity, the resolution manager creates a new entity with the account, and checks whether the new entity has relationships with other entities. If so, the resolution manager proceeds through the sequence of steps to ensure a load order independent resolution outcome as discussed above.