Financial Crime Detection Using Graphical Network Features
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for detecting financial crimes and fraud in financial institutions do not adequately utilize network effects and relationships between financial entities, leading to missed detections.
Innovation Solution
A machine learning method that generates graphical network features by applying financial entity risk indicators to a network model of financial entities and their relationships, and feeds these features into a predictive model to detect financial crimes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional classification and outlier identification approaches are used for financial crime detection, then the detection process is simple and fast, but financial crimes are missed because network relationships are not considered
Solution Approach 1:
The patent transitions from traditional single-entity feature analysis to multi-dimensional network-based analysis by constructing graphical representations of financial entities and their relationships. This dimensional expansion allows the system to capture network effects and relational patterns that traditional approaches miss, thereby improving detection accuracy without excessive complexity increase
Solution Approach 2:
The system performs preliminary construction of network models and graphical features before actual crime detection. By pre-establishing the network structure, entity relationships, and graphical representations, the system prepares the analytical framework in advance, enabling more accurate and efficient crime detection when actual transactions are analyzed
2Measurement precision
If network-based features are incorporated into financial crime detection, then detection accuracy improves by revealing hidden connections, but computational complexity increases
Solution Approach 1:
The patent segments the complex network analysis task into distinct components: network construction, graphical feature generation, and predictive modeling. Each component processes specific aspects of network data independently, allowing the system to manage computational complexity through modular processing while maintaining high detection precision through comprehensive network analysis
Solution Approach 2:
The system introduces graphical features as an intermediary representation between raw network data and final crime predictions. These graphical features serve as compressed, informative summaries of network relationships, reducing the dimensionality and computational burden of analyzing complete network structures while preserving the essential patterns needed for accurate detection
Data Source
AI summary
Disclosed is an example approach in which network and non-network features are used to train a predictive machine learning model that is implemented to predict financial crime and fraud. Graphical network features may be generated by applying financial entity risk vectors to a network model with representations of various types of networks. The network model may comprise transactional, non-social, and/or social networks, with edges corresponding to linkages that may be weighted according to various characteristics (such as frequency, amount, type, recency, etc.). The graphical network features may be fed to the predictive model to generate a likelihood and/or prediction with respect to a financial crime. A perceptible alert is generated on one or more computing devices if a financial crime is predicted or deemed sufficiently likely. The alert may identify a subset of the set of financial entities involved in the financial crime and present graphical representations of networks and linkages.