Reinforcement Learning Agent for Synthetic Customer Data Generation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Financial crime detection systems require large amounts of realistic simulated transaction data to build effective predictive models, but sensitive real customer data is limited, making it challenging to simulate fraudulent situations effectively.
Innovation Solution
A computer-implemented method using a reinforcement learning model with an intelligent agent, policy engine, and environment to generate simulated customer data by combining randomly selected real customer profile information with standard transaction data, iteratively adjusting policies based on feedback to achieve similarity with real customer transaction data, ensuring privacy protection through composite information splitting.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If more real customer data is used to train predictive models, then model accuracy improves, but customer privacy protection deteriorates
Solution Approach 1:
The patent creates synthetic copies of customer transaction data that preserve the statistical patterns and relationships of real data while containing no actual personal information. The generated data mimics the distribution, correlations, and complexity of real customer behavior, enabling accurate model training without exposing sensitive customer records.
Solution Approach 2:
The patent introduces an intermediate synthetic data generation process that acts as a mediator between real customer data and predictive models. Instead of directly using real data, the system generates intermediate synthetic datasets that capture essential patterns while eliminating privacy risks, thus resolving the contradiction between model accuracy and privacy protection.
2Object-affected harmful factors
If limited real customer data is used for training, then privacy protection is maintained, but model training quality deteriorates
Solution Approach 1:
The system generates abundant synthetic copies of customer transaction data that expand the limited real data into a comprehensive training dataset. These synthetic copies preserve the statistical properties and behavioral patterns of real customers, enabling high-quality model training while maintaining privacy protection through composite information splitting.
3Object-affected harmful factors
If simulated transaction data is generated to replace real data, then privacy protection improves, but data realism deteriorates
Solution Approach 1:
The patent employs advanced copying techniques that generate synthetic data with realistic statistical patterns, transaction behaviors, and contextual relationships. The generated data maintains the complexity and variability of real customer transactions, ensuring high realism while complete anonymization through composite information splitting ensures privacy protection.
Data Source
AI summary
Embodiments can provide a computer implemented method for simulating customer data using a reinforcement learning model, including: generating an artificial customer profile by combining randomly selected information from a set of real customer profile data; providing standard customer transaction data representing a group of real customers having similar transaction characteristics as a goal; performing a plurality of iterations to simulate the standard customer transaction data; and combining the artificial customer profile with the simulated customer transaction data to form simulated customer data. In each iteration, the method includes conducting an action including a plurality of simulated transactions; comparing the action with the goal; providing a feedback associated with the action based on a degree of similarity relative to the goal; adjusting a policy based on the feedback.