Graph Transformation for Suspicious Transaction Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Complex transaction flows involving multiple user accounts and cyclical transactions make it challenging for online service providers to detect suspicious activities, such as money laundering, using conventional analysis methods, which require significant computing resources and are difficult to visualize.
Innovation Solution
A multi-dimensional random walk traversal technique is employed to analyze transaction patterns, where the system selects a seed account and traverses the transaction graph, terminating at nodes based on a probability factor and edge weights, ranking nodes for further analysis, and presenting the results in a hierarchical structure to simplify detection of malicious activities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If conventional transaction analysis methods are used to detect suspicious activities, then detection capability is maintained, but computing resources required increase significantly and visualization becomes difficult
Solution Approach 1:
The patent segments the complex transaction graph into multiple layers based on node importance and transaction patterns. By dividing the analysis into hierarchical layers (core layer, intermediate layer, peripheral layer), the system can focus computational resources on the most critical segments first, reducing overall computing requirements while maintaining detection capability.
Solution Approach 2:
The patent introduces a hierarchical dimension to the traditional flat transaction graph analysis. By organizing nodes into multiple layers with different importance levels and analyzing them in a hierarchical manner, the system adds a dimensional structure that enables more efficient resource allocation and reduces the computational complexity of analyzing all transactions uniformly.
2Productivity
If conventional transaction analysis methods are used to detect suspicious activities, then detection capability is maintained, but visualization of complex transaction flows becomes difficult
Solution Approach 1:
The patent segments the transaction graph into hierarchical layers that can be visualized in a structured manner. By organizing nodes into core, intermediate, and peripheral layers with clear spatial separation, the visualization becomes more manageable and interpretable, allowing analysts to easily understand complex transaction patterns without being overwhelmed by the full complexity.
Solution Approach 2:
The patent uses hierarchical layering to add a vertical dimension to the visualization of transaction flows. This hierarchical organization transforms the flat, complex graph into a multi-level structure that can be more effectively displayed and analyzed visually, improving the ease of operation for detecting suspicious activities.
3Ease of operation
If graph transformation is performed to minimize reverse edges, then visualization clarity is improved, but processing complexity increases
Solution Approach 1:
The patent performs preliminary graph transformation operations during the graph construction phase, organizing nodes into hierarchical layers and minimizing reverse edges before the analysis begins. By preparing the graph structure in advance with optimal orientation and layer assignment, the system reduces the need for complex processing during actual transaction analysis, balancing visualization clarity with processing complexity.
Data Source
AI summary
Methods and systems are presented for analyzing transactions conducted through user accounts with an online service provider based on graph analysis. A graph is generated based on a set of seed accounts that are determined to be involved in suspicious activities. The graph includes a set of seed nodes representing the seed accounts, and a set of nodes representing user accounts that are connected to the set of seed accounts in downstream transactions. A random walk traversal based on multiple dimensions is performed on the graph to determine nodes that are closely related to the set of seed nodes. Transactions conducted through the seed accounts and accounts corresponding to nodes that are closely related to the set of seed nodes are analyzed to detect any malicious activities. The graph may also be presented according to a layered hierarchical structure for better visualization of transaction flows through the accounts.