Entity Disambiguation via Commerce Graph Association Metrics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Financial transaction histories are hindered by complexities such as spelling errors and differences in account information, leading to redundant entities that interfere with determining statistical associations and providing targeted services, reducing the relevance and usefulness of advertisements and services.
Innovation Solution
A computer system disambiguates entities by calculating similarity and association metrics based on historical financial data represented as a commerce graph, combining entities if they are determined to be the same, and requesting feedback to confirm entity identity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If account information is stored separately for different applications, then data can be managed across multiple platforms, but redundant entities appear due to spelling errors and information changes
Solution Approach 1:
The patent merges multiple account representations into a unified entity by calculating association metrics between financial transaction histories. When the association metric exceeds a threshold, separate account entries are consolidated into a single entity record, eliminating redundancy while preserving multi-platform data connectivity.
Solution Approach 2:
The system implements feedback mechanisms where entity disambiguation results are continuously refined. Association metrics are calculated and compared against thresholds, with feedback loops that adjust entity groupings based on financial interaction patterns, improving entity identity accuracy over time.
2Measurement precision
If entity disambiguation is performed manually, then accuracy can be maintained, but processing time and operational complexity increase significantly
Solution Approach 1:
The system performs automatic entity disambiguation by computing association metrics between financial transaction histories without manual intervention. The algorithm independently determines entity identity by analyzing financial interaction patterns, eliminating the need for time-consuming manual review while maintaining high accuracy through objective metric thresholds.
Solution Approach 2:
The patent replaces manual entity disambiguation processes with automated computational algorithms. Instead of human analysts comparing account information, a computer system calculates association metrics and applies disambiguation rules automatically, dramatically reducing processing time while maintaining consistent accuracy standards.
3Reliability
If all account variations are treated as separate entities, then data integrity is preserved, but statistical association analysis becomes unreliable
Solution Approach 1:
The patent combines account variations that represent the same entity by calculating association metrics between their financial transaction histories. When metrics indicate high association, accounts are merged into a single entity representation, preserving data integrity through systematic criteria while enabling accurate statistical analysis by eliminating false entity distinctions.
Solution Approach 2:
The system changes the parameter of entity identification from static account information to dynamic association metrics based on financial interactions. This parameter transformation allows the system to distinguish between genuine entity differences and variations due to spelling errors or information changes, improving both data integrity and statistical analysis accuracy.
Data Source
AI summary
During an analysis technique, information associated with two entities may be compared to determine a similarity metric. For example, a string distance between the information may be computed. Then, an association metric between financial-transaction histories for the two entities may be calculated. This calculation may involve comparing the nodes and branches in a commerce graph that represents financial interrelationships among a set of entities, including inputs received by the set of entities, outputs provided by the set of entities, and financial transactions among the set of entities. Next, a determination of whether the entities are likely to be a same entity may be based on the similarity metric and/or the association metric. If the entities are likely to be the same entity, the entities may be combined in a data structure. Alternatively, if the entities are not likely to be the same, they may remain separate.