Virtual Node Data Linking for Fraud Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Financial institutions face challenges in linking and analyzing data from multiple sources to effectively identify and mitigate various types of bank fraud activities, such as money laundering and credit card fraud, due to the complexity of understanding relationships between different pieces of information.
Innovation Solution
A system and method for linking data from multiple sources by examining data elements for shared characteristics, creating virtual nodes, and forming networks to identify relationships and behavioral patterns, allowing for the identification and analysis of risk factors associated with entities and transactions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data from multiple sources is collected and stored separately, then data security and source integrity are maintained, but the ability to identify relationships and detect fraud is reduced
Solution Approach 1:
The patent introduces virtual nodes as intermediary elements that connect data from multiple sources without requiring direct access to the original data sources. These virtual nodes serve as mediators that store and link data elements through common characteristics, enabling fraud detection across institutions while maintaining data security and source integrity.
Solution Approach 2:
The patent segments data into discrete data elements that can be independently stored in virtual nodes. Each data element is parsed and stored separately with its characteristics, allowing flexible linking and querying without handling complete datasets, thus reducing complexity while improving detection capability.
2Loss of information
If data elements are examined and linked based on common characteristics, then relationships between entities are uncovered, but processing time and computational resources increase
Solution Approach 1:
The patent performs preliminary actions by pre-parsing data elements and storing them in virtual nodes with their characteristics already extracted. This preliminary processing allows for rapid querying and relationship detection later, as the data is already structured and indexed by common characteristics rather than requiring full re-analysis.
Solution Approach 2:
The patent creates copies of data elements in virtual nodes that contain only the necessary characteristics for linking, not the complete original data. These copies enable relationship detection through common characteristics while significantly reducing processing time and computational resources compared to analyzing full datasets.
3Reliability
If comprehensive data analysis is performed across multiple sources, then fraud detection capability is improved, but data privacy and security requirements become more complex
Solution Approach 1:
Virtual nodes act as secure intermediaries that store and link data elements without requiring access to the original data sources. This intermediary layer enables comprehensive fraud analysis across multiple institutions while maintaining data privacy and security, as no single entity needs access to complete sensitive datasets.
Solution Approach 2:
The patent extracts only the necessary data elements and their characteristics from complete datasets, storing them in virtual nodes. This extraction process removes sensitive information while retaining the essential features needed for fraud detection, thereby simplifying security management while maintaining detection effectiveness.
Data Source
AI summary
A plurality of institutions (such as financial institutions) contribute data to a data analysis and linking system. The system analyzes the data to create data nodes (records) associated with an entity, where the entity may be, for example, a person/individual, business, organization, account, address, telephone number, etc. After data is linked, and in order to retrieve linked data, a requester may provide to the system an identifier associated with an entity. The linked data provided by the system in response to the identifier may be in the form of a network of data nodes associated with the entity and for use in assessing risk, such as risk associated with a transaction being conducted by a person. The linked data may also be analyzed at the system to score risk associated with the entity, and the risk score provided in conjunction with or in lieu of the network of data nodes.