Cryptocurrency Network Analysis Through Diffusion Risk Propagation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods struggle to effectively analyze networks with large numbers of nodes and edges, particularly in cryptocurrency systems, to detect illicit activities such as money laundering and fraud, due to incomplete transactional information and hidden links between nodes.
Innovation Solution
A computer-implemented method using a diffusion algorithm to update a graph dataset with risk parameters, calculating similarity scores, and identifying illicit nodes by applying a diffusion algorithm to propagate risk through the network, utilizing both classical and quantum computing for faster updates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional network analysis methods are used on cryptocurrency networks, then some insight into money laundering behaviour can be obtained, but the analysis becomes difficult to implement in networks having large numbers of nodes and edges
Solution Approach 1:
The patent transforms the network analysis problem by changing parameters from traditional topological properties to quantum mechanical analogs (Hamiltonian, energy eigenvalues, diffusion coefficients). This allows complex network structures to be analyzed through mathematical transformations that reduce computational complexity while maintaining detection accuracy.
Solution Approach 2:
The patent replaces classical network analysis methods with quantum-inspired computational approaches. By using quantum diffusion equations and Hamiltonian mechanics analogs, the system can process large-scale cryptocurrency networks more efficiently than traditional graph theory methods.
2Measurement precision
If traditional network analysis methods are used on cryptocurrency networks, then some insight into money laundering behaviour can be obtained, but the analysis becomes difficult when transactional information is incomplete
Solution Approach 1:
The patent implements iterative diffusion processes where risk parameters are continuously updated based on network propagation. The system uses feedback loops where initial risk assessments are refined through multiple diffusion steps, allowing the analysis to converge even with incomplete initial information about transactional relationships.
Solution Approach 2:
The patent applies preliminary risk parameter assignments to nodes based on available information before the diffusion process begins. This preliminary action provides an initial state that guides the subsequent diffusion analysis, enabling the system to work effectively even when complete transactional information is not available.
3Measurement precision
If diffusion algorithm is applied to update risk parameters across the network, then hidden links between nodes can be detected, but computational resources and time increase
Solution Approach 1:
The patent applies the diffusion algorithm iteratively with a predetermined number of steps rather than requiring full convergence. This partial action approach provides sufficient detection accuracy for identifying hidden links and illicit activities while significantly reducing the computational time and resources required compared to running the diffusion process to complete convergence.
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
A computer-implemented method for analysing a network of information comprising a plurality of interconnected nodes. The method comprises: accessing a graph dataset (402) representative of a graph comprising a plurality of nodes and a plurality of directed edges; updating the graph dataset (404); and, for each pair of nodes in the plurality of nodes, calculating a similarity score (406) using the updated graph dataset. The similarity score is indicative of the similarity between the nodes for which the similarity score is calculated. The graph dataset comprises an initial risk parameter for each node and an initial weight for each directed edge. Updating the graph dataset (404) comprises applying a diffusion algorithm to the graph dataset until a predetermined condition is met. The updated graph dataset comprises an updated risk parameter for each node.