System and method for determining a decision-making strategy
One embodiment of the present disclosure provides a system and method for determining a decision-making strategy. During operation, the system can obtain sample data corresponding to a service. The system can pre-determine, based on the sample data and the service, a basic parametric shape model and can generate a plurality of shape parameters for the basic parametric shape mode. The system can then generate one or more parametric shape models with different shapes based on the plurality of shape parameters. The system can determine goodness-of-fit of the one or more parametric shape models with respect to a pre-defined decision-making strategy. Next, the system can in response to determining that the goodness-of-fit of at least one parametric shape model satisfies a set of objectives, outputting the at least one parametric shape model as an optimum decision-making strategy for the service.
1. A computer-implemented method, comprising:
obtaining sample data corresponding to a service;
pre-determining, based on the sample data and the service, a basic parametric shape model;
training, through machine learning, the basic parametric shape model by iteratively performing acts including:
generating a plurality of shape parameters for the basic parametric shape model;
generating, based on the plurality of shape parameters, one or more parametric shape models with different shapes;
determining goodness-of-fit of the one or more parametric shape models with respect to a predefined decision-making strategy;
in response to determining that the goodness-of-fit of the one or more parametric shape models does not satisfy the set of objectives:
one or more of hybridizing the plurality of shape parameters to generate a hybridized set of shape parameters or mutating the plurality of shape parameters to generate a mutated set of shape parameters; and
updating the one or more parametric shape models based on one or more of the hybridized set of shape parameters or the mutated set of shape parameters;
in response to determining that the goodness-of-fit of at least one parametric shape model satisfies a set of objectives, outputting the at least one parametric shape model as a decision-making strategy for the service; and
applying the decision-making strategy to real-time transaction data to provide enhanced risk-control for the service.
2. The method of claim 1 , wherein the goodness-of-fit corresponds to a measure of similarity between the pre-defined decision-making strategy and a respective decision-making strategy corresponding to the one or more parametric shape models.
3. The method of claim 1 , wherein determining the goodness-of-fit of the one or more parametric shape models with respect to the predefined decision-making strategy further comprises:
determining, based on each parametric shape model and the sample data, a disturb rate and a coverage rate, wherein the disturb rate is associated with a first portion of sample data generated by a reliable service, and wherein the coverage rate is associated with a second portion of the sample data generated by an unreliable entity;
determining a difference between the disturb rate and a pre-defined disturb rate associated with the pre-defined decision-making strategy; and
determining a difference between the coverage rate and a pre-defined coverage rate associated with the pre-defined decision-making strategy.
4. The method of claim 1 , wherein the set of objectives include one or more of:
a pre-defined goodness-of-fit;
a pre-defined disturb rate; and
a pre-defined coverage rate.
5. The method of claim 1 , further comprising:
in response to determining that the goodness-of-fit of the one or more parametric shape models does not satisfy the set of objectives and the one or more parametric shape models corresponds to a last update of the plurality of shape parameters, outputting a parametric shape model with best goodness-of-fit as an optimum decision-making strategy for the service.
6. The method of claim 1 , wherein the one or more parametric shape models describe a relationship between a service parameter of the service and a risk assessment value of the service.
7. A computer system, comprising:
one or more processors; and
one or more storage devices coupled to the one or more processors and, individually or collectively storing instructions which when executed by the one or more processors cause the one or more processors to, individually or collectively, perform a method, the method comprising
obtaining sample data corresponding to a service;
pre-determining, based on the sample data and the service, a basic parametric shape model;
training, through machine learning, the basic parametric shape model by iteratively performing acts including:
generating a plurality of shape parameters for the basic parametric shape model;
generating, based on the plurality of shape parameters, one or more parametric shape models with different shapes;
determining goodness-of-fit of the one or more parametric shape models with respect to a predefined decision-making strategy; and
in response to determining that the goodness-of-fit of the one or more parametric shape models does not satisfy the set of objectives;
one or more of hybridizing, the plurality of shape parameters to generate a hybridized set of shape parameters or mutating the plurality of shape parameters to generate a mutated set of shape parameters; and
updating the one or more parametric shape models based on one or more of the hybridized set of shape parameters or the mutated set of shape parameters;
in response to determining that the goodness-of-fit of at least one parametric shape model satisfies a set of objectives, outputting the at least one parametric shape model as a decision-making strategy for the service; and
applying the decision-making strategy to real-time transaction data to provide enhanced risk-control for the service.
8. The computer system of claim 7 , wherein the goodness-of-fit corresponds to a measure of similarity between the pre-defined decision-making strategy and a respective decision-making strategy corresponding to the one or more parametric shape models.
9. The computer system of claim 7 , wherein determining the goodness-of-fit of the one or more parametric shape models with respect to the predefined decision-making strategy further comprises:
determining, based on each parametric shape model and the sample data, a disturb rate and a coverage rate, wherein the disturb rate is associated with a first portion of sample data generated by a reliable service, and wherein the coverage rate is associated with a second portion of the sample data generated by an unreliable entity;
determining a difference between the disturb rate and a pre-defined disturb rate associated with the pre-defined decision-making strategy; and
determining a difference between the coverage rate and a pre-defined coverage rate associated with the pre-defined decision-making strategy.
10. The computer system of claim 7 , wherein the method further comprises:
in response to determining that the goodness-of-fit of the one or more parametric shape models does not satisfy the set of objectives and the one or more parametric shape models corresponds to a last update of the plurality of shape parameters, outputting a parametric shape model with best goodness-of-fit as an optimum decision-making strategy for the service.
11. A non-transitory computer-readable storage medium storing instructions that when executed by a computer cause the computer to perform a method, the method comprising:
obtaining sample data corresponding to a service;
pre-determining, based on the sample data and the service, a basic parametric shape model;
training, through machine learning, the basic parametric shape model by iteratively performing acts including:
generating a plurality of shape parameters for the basic parametric shape model;
generating, based on the plurality of shape parameters, one or more parametric shape models with different shapes;
determining goodness-of-fit of the one or more parametric shape models with respect to a predefined decision-making strategy; and
in response to determining that the goodness-of-fit of the one or more parametric shape models does not satisfy the set of objectives:
one or more of hybridizing the plurality of shape parameters to generate a hybridized set of shape parameters or mutating the plurality of shape parameters to generate a mutated set of shape parameters; and
updating the one or more parametric shape models based on one or more of the hybridized set of shape parameters or the mutated set of shape parameters;
in response to determining that the goodness-of-fit of at least one parametric shape model satisfies a set of objectives, outputting the at least one parametric shape model as a decision-making strategy for the service; and
applying the decision-making strategy to real-time transaction data to provide enhanced risk-control for the service.
12. The non-transitory computer-readable storage medium of claim 11 , wherein the goodness-of-fit corresponds to a measure of similarity between the pre-defined decision-making strategy and a respective decision-making strategy corresponding to the one or more parametric shape models.
13. The non-transitory computer-readable storage medium of claim 11 , wherein determining the goodness-of-fit of the one or more parametric shape models with respect to the predefined decision-making strategy further comprises:
determining, based on each parametric shape model and the sample data, a disturb rate and a coverage rate, wherein the disturb rate is associated with a first portion of sample data generated by a reliable service, and wherein the coverage rate is associated with a second portion of the sample data generated by an unreliable entity;
determining a difference between the disturb rate and a pre-defined disturb rate associated with the pre-defined decision-making strategy; and
determining a difference between the coverage rate and a pre-defined coverage rate associated with the pre-defined decision-making strategy.
14. The non-transitory computer-readable storage medium of claim 11 , the method further comprising:
in response to determining that the goodness-of-fit of the one or more parametric shape models does not satisfy the set of objectives and the one or more parametric shape models corresponds to a last update of the plurality of shape parameters, outputting a parametric shape model with best goodness-of-fit as an optimum decision-making strategy for the service.
15. The computer system of claim 7 , wherein the set of objectives include one or more of:
a pre-defined goodness-of-fit;
a pre-defined disturb rate; and
a pre-defined coverage rate.
16. The computer system of claim 7 , wherein the one or more parametric shape models describe a relationship between a service parameter of the service and a risk assessment value of the service.
17. The non-transitory computer-readable storage medium of claim 11 , wherein the set of objectives include one or more of:
a pre-defined goodness-of-fit;
a pre-defined disturb rate; and
a pre-defined coverage rate.
18. The non-transitory computer-readable storage medium of claim 11 , wherein the one or more parametric shape models describe a relationship between a service parameter of the service and a risk assessment value of the service.