System and method for unsupervised abstraction of sensitive data for consortium sharing
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.
1. A computer-implemented method for generating an artificial customer in a data processing system comprising a processing device and a memory comprising instructions which are executed by the processing device, the method comprising:
receiving customer data from a plurality of computing devices over a network, the customer data including information for a plurality of customers to a plurality of entities;
performing, by the processing device, unsupervised learning on the customer data to produce a plurality of clusters of customers with a first feature in common;
performing, by the processing device, unsupervised learning on the plurality of clusters of customers to produce a plurality of sub-clusters of customers with a second feature in common;
repeating, by the processing device, the unsupervised learning on the plurality of sub-clusters to produce further sub-clusters with a plurality of features in common;
determining, by the processing device, that a sub-cluster represents a group of real customers;
determining a data distribution for each of the plurality of features in common present in the sub-cluster;
generating an artificial customer profile by combining randomly selected information including an address, a first name, a second name, a phone number, an email address, a credit score, a revenue, and a wage from the customer data;
generating, by a cognitive system, an artificial customer by combining the artificial customer profile with the data distributions;
providing the artificial customer to each of the plurality of computing devices for generating simulated transaction data based on the group of real customers;
generating, by the cognitive system of at least one of the plurality of computing devices, simulated transaction data in imitation of real transaction data through a reinforcement learning model, wherein the simulated transaction data cannot be traced to the customer data, wherein the step of generating the simulated transaction data further comprises:
providing the artificial customer as a goal;
performing a plurality of iterations to generate simulated transaction data, wherein the plurality of iterations is performed until a degree of similarity of the simulated transaction data relative to the artificial customer is higher than a first predefined threshold,
in each iteration:
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, and
adjusting a policy based on the feedback; and
training a predictive model for identifying an abnormal customer behavior, using the simulated customer data.
2. The method of claim 1 , wherein the information for the plurality of customers comprises identifying information and transaction information.
3. The method of claim 1 , further comprising filtering the customer data prior to performing unsupervised learning.
4. The method of claim 3 , wherein the filtering comprises a recency, frequency, monetary value (RFM) analysis to group customers.
5. The method of claim 1 , wherein determining that the sub-cluster represents a standard the group of real customers comprises applying one or more rules.
6. The method of claim 5 , wherein the one or more rules comprise a size determination indicating a minimum or maximum number of customers in a sub-cluster that is determined to represent the group of real customers.
7. An abstraction system comprising a processing device and a memory comprising instructions which are executed by the processing device for generating an artificial customer in a data processing system configured to:
receiving customer data from a plurality of computing devices over a network, the customer data including information for a plurality of customers to a plurality of entities;
performing, by the processing device, unsupervised learning on the customer data to produce a plurality of clusters of customers with a first feature in common;
performing, by the processing device, unsupervised learning on the plurality of clusters of customers to produce a plurality of sub-clusters of customers with a second feature in common;
repeating, by the processing device, the unsupervised learning on the plurality of sub-clusters to produce further sub-clusters with a plurality of features in common;
determining, by the processing device, that a sub-cluster represents a group of real customers;
determining a data distribution for each of the plurality of features in common present in the sub-cluster;
generating an artificial customer profile by combining randomly selected information including an address, a first name, a second name, a phone number, an email address, a credit score, a revenue, and a wage;
generating, by a cognitive system, an artificial customer by combining the artificial customer profile with the data distributions;
providing the artificial customer to each of the plurality of computing devices for generating simulated transaction data based on the group of real customers;
generating, by the cognitive system of at least one of the plurality of computing devices, simulated transaction data in imitation of real transaction data through a reinforcement learning model, wherein the simulated transaction data cannot be traced to the customer data, wherein the step of generating the simulated transaction data further comprises:
providing the artificial customer as a goal;
performing a plurality of iterations to generate simulated transaction data, wherein the plurality of iterations is performed until a degree of similarity of the simulated transaction data relative to the artificial customer is higher than a first predefined threshold,
in each iteration:
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, and
adjusting a policy based on the feedback; and
training a predictive model for identifying an abnormal customer behavior, using the simulated customer data.
8. The abstraction system of claim 7 , wherein the information for the plurality of customers comprises identifying information and transaction information.
9. The abstraction system of claim 7 , further comprising filtering the customer data prior to performing unsupervised learning.
10. The abstraction system of claim 9 , wherein the filtering comprises a recency, frequency, monetary value (RFM) analysis to group customers.
11. The abstraction system of claim 7 , wherein determining that the sub-cluster represents the group of real customers comprises applying one or more rules.
12. The abstraction system of claim 11 , wherein the one or more rules comprise a size determination indicating a minimum or maximum number of customers in a sub-cluster that is determined to represent the group of real customers.
13. A computer program product comprising a non-transitory memory storing software that when executed by a processor performs a method comprising:
receiving customer data from a plurality of computing devices over a network, the customer data including information for a plurality of customers to a plurality of entities;
performing, by the processor, unsupervised learning on the customer data to produce a plurality of clusters of customers with a first feature in common;
performing, by the processor, unsupervised learning on the plurality of clusters of customers to produce a plurality of sub-clusters of customers with a second feature in common;
repeating, by the processor, the unsupervised learning on the plurality of sub-clusters to produce further sub-clusters with a plurality of features in common;
determining, by the processor, that a sub-cluster represents a group of real customers;
determining a data distribution for each of the plurality of features in common present in the sub-cluster;
generating an artificial customer profile by combining randomly selected information including an address, a first name, a second name, a phone number, an email address, a credit score, a revenue, and a wage;
generating, by a cognitive system, an artificial customer by combining the artificial customer profile with the data distributions;
providing the artificial customer to each of the plurality of computing devices for generating simulated transaction data based on the group of real customers;
generating, by the cognitive system of at least one of the plurality of computing devices, simulated transaction data in imitation of real transaction data through a reinforcement learning model, wherein the simulated transaction data cannot be traced to the customer data, wherein the step of generating the simulated transaction data further comprises:
providing the artificial customer as a goal;
performing a plurality of iterations to generate simulated transaction data, wherein the plurality of iterations is performed until a degree of similarity of the simulated transaction data relative to the artificial customer is higher than a first predefined threshold,
in each iteration:
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, and
adjusting a policy based on the feedback; and
training a predictive model for identifying an abnormal customer behavior, using the simulated customer data.
14. The computer program product of claim 13 , wherein the information for the plurality of customers comprises identifying information and transaction information.
15. The computer program product of claim 13 , further comprising filtering the customer data prior to performing unsupervised learning.
16. The computer program product of claim 15 , wherein the filtering comprises a recency, frequency, monetary value (RFM) analysis to group customers.
17. The computer program product of claim 13 , wherein determining that the sub-cluster represents the group of real customers comprises applying one or more rules.
18. The computer program product of claim 17 , wherein the one or more rules comprise a size determination indicating a minimum or maximum number of customers in a sub-cluster that is determined to represent the group of real customers.