Account Duplication Identification via Transaction Graph Clustering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Account aggregation data management systems face challenges in accurately identifying and reconciling user account and transaction data from multiple sources, leading to duplicate account listings due to software bugs, scripting errors, and changes in account provider websites, resulting in incorrect data representation and user dissatisfaction.
Innovation Solution
A dual-phase approach involving a projection phase to group user account and transaction data into transactions entities and project them into n-dimensional space, followed by a clustering phase to identify matching accounts and generate a bi-directional transactions graph, which indicates account duplication with minimal latency and resource usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If traditional methods are used to obtain account data from multiple providers, then comprehensive account coverage is achieved, but duplicate account listings increase
Solution Approach 1:
The system implements a feedback mechanism where account data obtained from multiple providers is continuously monitored for duplicates. When duplicates are detected through comparison of account identifiers and transaction patterns, the system adjusts its data collection and reconciliation processes to eliminate the duplicates, thereby maintaining both comprehensive coverage and high accuracy.
Solution Approach 2:
The patent introduces an intermediary reconciliation process that acts as a mediator between multiple account providers. This intermediary layer standardizes and compares data from different sources, identifying and resolving duplicate accounts before presenting the final consolidated view to the user, thus preventing duplicate listings while maintaining comprehensive coverage.
2Quantity of substance
If account data is obtained from multiple providers using various methods, then data completeness improves, but processing complexity increases
Solution Approach 1:
The system transforms account data into a standardized format with consistent parameters across all providers. By normalizing account identifiers, transaction structures, and data formats, the system enables efficient comparison and duplicate detection without requiring complex provider-specific processing logic, thus reducing overall processing complexity while maintaining data completeness.
Solution Approach 2:
The reconciliation process is segmented into distinct phases: data collection from providers, standardized formatting, duplicate identification through comparison, and final consolidation. This segmentation allows each phase to be optimized independently, reducing overall complexity while ensuring complete and accurate data processing.
3Reliability
If comprehensive account reconciliation is performed, then data accuracy improves, but processing time increases
Solution Approach 1:
The system performs preliminary actions by establishing account identification rules and comparison criteria before the actual reconciliation process. Account data is pre-formatted and indexed with key identifiers, allowing for rapid duplicate detection during reconciliation. This preliminary preparation significantly reduces the time required for comprehensive accuracy checking.
Solution Approach 2:
The patent replaces manual or brute-force comparison methods with automated algorithms that use account identifiers, transaction patterns, and provider metadata to quickly identify duplicates. This substitution of mechanical comparison with intelligent automated systems maintains high data accuracy while dramatically reducing processing time.
Data Source
AI summary
A method and system of identifying account duplication in data management systems utilizes a dual-phase approach, consisting of a projection phase and a clustering phase. In the projection phase, user account and transaction data is obtained and grouped into transactions entities. These transactions entities are then processed to create transactions entity projections. In the clustering phase, the transactions entity projections are grouped and aggregated according to specified parameters, which results in the generation of transactions graphs. The data from the transactions graphs is processed and analyzed to generate reliable indicators representing the presence or absence of account duplication in a data management system.