Synthetic Customer Profile Generation for Privacy-Safe Financial Modeling
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 real customer data is sensitive and limited, making it challenging to generate sufficient training data without exposing sensitive information.
Innovation Solution
A computer-implemented method using unsupervised learning to create standard customer profiles, which are then used to generate synthetic transaction data that mimics real customer behavior, ensuring the data is realistic yet untraceable and does not expose sensitive information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If real customer data is used to train predictive models, then model accuracy is improved, but customer privacy is compromised
Solution Approach 1:
The patent creates synthetic customer profiles that copy the statistical patterns and behavioral characteristics of real customers without containing actual sensitive information. These synthetic profiles are generated by learning from real data and then reproducing only the necessary patterns for model training, thereby maintaining accuracy while protecting privacy.
Solution Approach 2:
The patent introduces an intermediary process that transforms real customer data into synthetic profiles. This intermediary layer (the data generation system with privacy constraints) allows the benefits of real data to be captured while filtering out sensitive information, serving as a mediator between data utility and privacy protection.
2Productivity
If more simulated customer data is generated, then predictive model performance is improved, but data sensitivity risks increase
Solution Approach 1:
The system generates multiple synthetic customer profiles by copying statistical patterns from real data rather than duplicating actual records. This allows scaling training data volume while maintaining privacy, as each synthetic profile is a pattern-based reproduction而非 a copy of sensitive information.
Solution Approach 2:
The patent changes the parameters of customer profiles by learning statistical distributions from real data and then generating new instances with varied parameters. This allows generating diverse synthetic data for model training while ensuring no actual sensitive parameters are exposed.
3Object-affected harmful factors
If real customer data is restricted for privacy reasons, then customer privacy is protected, but training data quality decreases
Solution Approach 1:
The system copies the essential statistical patterns and behavioral characteristics from real customer data into synthetic profiles. This copying process preserves the quality aspects needed for training (transaction patterns, spending behaviors, temporal patterns) while excluding sensitive personal information.
Solution Approach 2:
The data generation system acts as an intermediary that processes real data to extract quality patterns while filtering sensitive information. This intermediary transformation maintains training data quality by preserving statistical properties while protecting privacy through controlled data synthesis.
Data Source
AI summary
An abstraction system for generating a standard customer profile in a data processing system has a processing device and a memory. The abstraction system may receive customer data from a computing device over a network and perform unsupervised learning on the customer data to produce a plurality of clusters of customers with a first feature in common. The abstraction system performs unsupervised learning on the plurality of clusters of customers to produce a plurality of sub-clusters of customers with a second feature in common, and repeats the unsupervised learning on the plurality of sub-clusters produced to produce further sub-clusters with a plurality of features in common. The abstraction system determines that a sub-cluster represents a standard customer and stores a plurality of standard customer profiles based on the determined standard customers. The abstraction system provides the standard customer profiles to a cognitive system for generating synthetic transaction data.