Reinforcement Learning Model for Simulated Transaction Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Financial crime detection systems face challenges in training predictive models due to the sensitivity and limited availability of real customer data, necessitating the use of simulated transaction data that resembles real data to effectively detect fraudulent activities.
Innovation Solution
A computer-implemented method utilizing a reinforcement learning model with an intelligent agent and policy engine to simulate transaction data, identify behavioral patterns, and generate new data that mimics standard customer transactions, ensuring the simulated data aligns with real data characteristics through iterative adjustments based on feedback.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If real customer data is used to train predictive models, then the accuracy of financial crime detection is improved, but data availability is limited due to sensitivity and privacy concerns
Solution Approach 1:
The patent creates synthetic copies of real customer transaction data through simulation. The system generates artificial transaction records that replicate the statistical properties, patterns, and behavioral characteristics of real data without containing actual sensitive information. This allows unlimited availability of training data while maintaining privacy protection.
Solution Approach 2:
The simulation system adjusts key parameters such as transaction amounts, frequencies, types, and temporal patterns to match the statistical distributions observed in real customer data. By calibrating these parameters, the synthetic data achieves fidelity to real-world patterns while being completely artificial in origin.
2Quantity of substance
If simulated transaction data is generated to increase data availability, then data quantity is improved, but the realism and similarity to actual customer behavior deteriorates
Solution Approach 1:
The system employs feedback loops where simulated data is continuously compared against real data patterns. The simulation model learns from discrepancies between generated and actual customer behavior, adjusting its parameters and algorithms to improve realism. This iterative refinement process ensures synthetic data progressively converges toward authentic transaction patterns.
Solution Approach 2:
The system performs preliminary analysis of real customer data to extract behavioral patterns, statistical distributions, and transaction characteristics before generating synthetic data. By pre-characterizing the target patterns, the simulation can be configured to reproduce these features accurately from the outset, improving the initial realism of generated data.
3Adaptability or versatility
If more simulated data is generated to detect different types of financial crimes, then the versatility of detection is improved, but the complexity of the simulation system increases
Solution Approach 1:
The simulation system is designed as a multi-functional platform that can generate diverse transaction scenarios for various crime types using a unified framework. By implementing configurable transaction templates and behavioral models that can be adjusted for different crime patterns (money laundering, fraud, terrorist financing), the system achieves broad detection coverage without requiring separate specialized systems for each crime type.
Data Source
AI summary
Embodiments can provide a method for identifying a behavioral pattern from simulated transaction data, including: simulating transaction data using a reinforcement learning model; identifying a behavioral pattern from the simulated transaction data; comparing the behavioral pattern with standard customer transaction data to determine whether the behavioral pattern is present in the standard customer transaction data. If the behavioral pattern is present in the standard customer transaction data, the behavioral pattern is applied in a model implemented on the cognitive system. The step of simulating transaction data further includes: providing standard customer transaction data representing a group of customers having similar transaction characteristics as a goal; and performing a plurality of iterations to simulate the standard customer transaction data, wherein the plurality of iterations is performed until a degree of similarity of simulated customer transaction data relative to the standard customer transaction data is higher than a first predefined threshold.