Intelligent agent to simulate customer data

1. A computer implemented method in a data processing system comprising a processor and a memory comprising instructions, which are executed by the processor to cause the processor to implement the method for simulating customer data using a reinforcement learning model including an intelligent agent, a policy engine, and an environment, the method comprising:

generating, by the processor, an artificial customer profile by:

splitting into a plurality of parts, composite information from a set of real customer data including a name, contact information, and finance information;

randomly selecting at least two of the plurality of parts;

combining the randomly selected at least two of the plurality of parts; and

updating the artificial customer profile when a degree of similarity between the combination of the randomly selected at least two of the plurality of parts and the composite information from a set of real customer data is greater than a predefined threshold value for the artificial customer profile;

generating, by the processor using the reinforcement learning model, simulated transaction data in imitation of real transaction data of a group of real customers having similar transaction characteristics, wherein generating the simulated transaction data further comprises:

providing, by the processor, statistic data representing the group of real customers having similar transaction characteristics as a goal;

performing, by the processor, a plurality of iterations to simulate the real transaction data, wherein the plurality of iterations is performed until a degree of similarity of simulated transaction data relative to the statistic data is higher than a first predefined threshold;

in each iteration:

conducting, by the intelligent agent, an action including a plurality of simulated transactions;

comparing, by the environment, the action with the goal;

providing, by the environment, a feedback associated with the action based on a degree of similarity relative to the goal; and

updating, by the policy engine, a policy based on the feedback for determining a next action,

wherein using the statistic data as the goal results in an inability to trace the simulated transaction data to a real customer from the group of real customers; and

generating, by the processor, simulated customer data by combining the artificial customer profile with the last action to form simulated customer data; and

training a predictive model for identifying an abnormal customer behavior, using the simulated customer data.

2. The method as recited in claim 1 , wherein the real customer profile data includes one or more of an address of a customer, a name of a customer, contact information, credit information, and income information.

3. The method as recited in claim 1 , wherein each simulated transaction includes transaction type, transaction amount, transaction time, transaction location, transaction medium, a second party associated with the simulated transaction.

4. The method as recited in claim 1 , wherein the environment includes a set of all previous actions conducted by the intelligent agent.

5. The method as recited in claim 4 , further comprising:

removing, by the processor, a plurality of previous actions having the degree of similarity lower than a second predefined threshold.

6. The method as recited in claim 1 , further comprising:

acquiring, by the processor, the statistic data from raw customer transaction data through an unsupervised clustering approach.

7. The method as recited in claim 1 , wherein the feedback is a reward or a penalty.

8. A computer program product for simulating customer data using a reinforcement learning model including an intelligent agent, a policy engine, and an environment, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor to cause the processor to:

generate an artificial customer profile by:

splitting into a plurality of parts, composite information from a set of real customer data including a name, contact information, and finance information;

randomly selecting at least two of the plurality of parts;

combining the randomly selected at least two of the plurality of parts; and

updating the artificial customer profile when a degree of similarity between the combination of the randomly selected at least two of the plurality of parts and the composite information from a set of real customer data is greater than a predefined threshold value for the artificial customer profile;

generate, using the reinforcement learning model, simulated transaction data in imitation of real transaction data of a group of customers having similar transaction characteristics, wherein generating the simulated transaction data further comprises:

providing statistic data representing the group of customers having similar transaction characteristics as a goal;

performing a plurality of iterations to simulate the real transaction data, wherein the plurality of iterations is performed until a degree of similarity of simulated transaction data relative to the statistic data is higher than a first predefined threshold;

in each iteration:

conducting, by the intelligent agent, an action including a plurality of simulated transactions;

comparing, by the environment, the action with the goal;

providing, by the environment, a feedback associated with the action based on a degree of similarity relative to the goal; and

update, by the policy engine, a policy based on the feedback for determining a next action,

wherein using the statistic data as the goal results in an inability to trace the simulated transaction data to a real customer from the group of real customers; and

generate simulated customer data by combining the artificial customer profile with the last action to form simulated customer data; and

training a predictive model for identifying an abnormal customer behavior, using the simulated customer data.

9. The computer program product of claim 8 , wherein the real customer profile data includes one or more of an address of a customer, a name of a customer, contact information, credit information, and income information.

10. The computer program product of claim 8 , wherein each simulated transaction includes transaction type, transaction amount, transaction time, transaction location, transaction medium, a second party associated with the simulated transaction.

11. The computer program product of claim 8 , wherein the environment includes a set of all previous actions conducted by the intelligent agent.

12. The computer program product of claim 11 , wherein the program instructions executable by the processor further cause the processor to remove a plurality of previous actions having the degree of similarity lower than a second predefined threshold.

13. The computer program product of claim 8 , wherein the program instructions executable by the processor further cause the processor to:

acquire the statistic data from raw customer transaction data through an unsupervised clustering approach.

14. The computer program product of claim 8 , wherein the feedback is a reward or a penalty.

15. A system for simulating customer data using a reinforcement learning model including an intelligent agent, a policy engine, and an environment, the system comprising:

a processor configured to:

generate an artificial customer profile by:

splitting into a plurality of parts, composite information from a set of real customer data including a name, contact information, and finance information;

randomly selecting at least two of the plurality of parts;

combining the randomly selected at least two of the plurality of parts; and

updating the artificial customer profile when a degree of similarity between the combination of the randomly selected at least two of the plurality of parts and the composite information from a set of real customer data is greater than a predefined threshold value for the artificial customer profile;

generate, using the reinforcement learning model, simulated transaction data in imitation of real transaction data of a group of customers having similar transaction characteristics, wherein generating the simulated transaction data further comprises:

providing statistic data representing the group of customers having similar transaction characteristics as a goal;

performing a plurality of iterations to simulate the real transaction data, wherein the plurality of iterations is performed until a degree of similarity of simulated transaction data relative to the statistic data is higher than a first predefined threshold;

in each iteration:

 conducting, by the intelligent agent, an action including a plurality of simulated transactions;

 comparing, by the environment, the action with the goal;

 providing, by the environment, a feedback associated with the action based on a degree of similarity relative to the goal; and

 updating, by the policy engine, a policy based on the feedback for determining a next action,

 wherein using the statistic data as the goal results in an inability to trace the simulated transaction data to a real customer from the group of real customers; and

generate simulated customer data by combining the artificial customer profile with the last action to form simulated customer data; and

training a predictive model for identifying an abnormal customer behavior, using the simulated customer data.

16. The system of claim 15 , wherein the real customer profile data includes one or more of an address of a customer, a name of a customer, contact information, credit information, and income information.

17. The system of claim 15 , wherein the environment includes a set of all previous actions conducted by the intelligent agent.

18. The system of claim 17 , wherein prior to the step of adjusting a policy, the processor is further configured to add the action into the environment.

19. The system of claim 17 , wherein the processor is further configured to remove a plurality of previous actions having the degree of similarity lower than a second predefined threshold.

20. The system of claim 15 , wherein the feedback is a reward or a penalty.