Systems and methods for validating interpolation results using monte carlo simulations on interpolated data inputs

• US 8,290,969 B2
• Filed: 02/28/2011
• Issued: 10/16/2012
• Est. Priority Date: 02/28/2011
• Status: Active Grant

• Current Assignee: Red Hat, Inc. (International Business Machines Corporation)
• Original Assignee: Red Hat, Inc. (International Business Machines Corporation)
• Inventors: Williamson, Eric
• Primary Examiner(s): Vy, Hung T
• Application Number:
  US13/037,344
• Publication Number:
  US 20120221500A1
• Time in Patent Office: 596 Days
• Field of Search: 707/756, 706/12, 706/13, 706/45, 706/47, 704/205
• US Class Current: 707/756
• CPC Class Codes: G06F 17/17 Function evaluation by approximation methods, e.g. inter- or extrapolation, smoothing, least mean square method G06F17/18 takes precedence ; interpolation for numerical control G05B19/18

Abstract

Embodiments relate to systems and methods for validating interpolation results using Monte Carlo simulations on interpolated data inputs. A database can store sets of operational data, such as financial, medical, climate or other information. For given data, a portion of the input data can be known or predetermined, while for a second portion can be unknown and subject to interpolation. The interpolation engine can generate a conformal interpolation function and interpolated input sets that map to a set of target output data. In aspects, in order to test the interpolated input data, the operator can initiate a Monte Carlo or other variational analysis using access a validation dialog. The Monte Carlo process can apply randomized perturbations to the values of the interpolated input variables, and track the results of that perturbation on the other interpolated inputs. A set of validation rules can be applied to those randomized results, to determine whether the remaining interpolated variables remain in conformance with expected ranges or values or demonstrate anomalous responses.

Claims

1. A method, comprising:

accessing, by an interpolation engine running on a computing device, a set of combined input data comprising a set of predetermined input data and a set of interpolated input data, the set of interpolated input data being generated to conformally map the set of combined input data to a set of target output data and comprising a set of variables;

performing, by the interpolation engine, a variational analysis on at least a first variable in the set of variables, wherein the variational analysis comprises applying a perturbation to the at least first variable;

determining a change in at least one remaining variable in the set of variables caused by the variational analysis performed on the at least first variable; and

applying a set of validation rules to the change in the at least one remaining variable to validate or reject a value of the at least first variable.

2. The method of claim 1, wherein the variational analysis comprises a Monte Carlo simulation performed on the at least first variable.

3. The method of claim 2, wherein the perturbation is a random perturbation, and wherein the Monte Carlo simulation comprises applying the random perturbation to the at least first variable.

4. The method of claim 3, wherein the random perturbation comprises an increase or decrease in the at least first variable based on a randomization range.

5. The method of claim 4, wherein the set of validation rules comprises a limit to the change in the at least one of the remaining variables.

6. The method of claim 5, wherein the limit to the change in the at least one of the remaining variables comprises a function of the randomization range.

7. The method of claim 1, further comprising repeating the variational analysis on the remainder of the variables in the set of interpolated input data.

8. The method of claim 7, further comprising repeating the determining a change in at least one of the remaining variables in the set of variables in the set of interpolated input data caused by variational analysis performed on the remainder of the variables in the set of interpolated input data.

9. The method of claim 8, further comprising storing changed variables caused by the variational analysis across all variables in the set of interpolated input data to generate a set of Monte Carlo interpolation results.

10. The method of claim 1, wherein the set of predetermined input data comprises at least one of a set of financial data, a set of medical data, a set of demographic data, a set of engineering data, a set of network operations data, or a set of geographic data.

11. A system, comprising:

an interface to a database storing a set of target output data and a set of combined input data, the set of combined input data comprising—

a set of predetermined input data, and

a set of interpolated input data, the set of interpolated input data being generated to conformally map the set of combined input data to the set of target output data and comprising a set of variables; and

a processor, communicating with the database via the interface, the processor being configured to—

access the set of combined input data,

perform a variational analysis on at least a first variable in the set of variables, wherein the variational analysis comprises applying a perturbation to the at least first variable,

determine a change in at least one remaining variable in the set of variables caused by the variational analysis performed on the at least first variable, and

apply a set of validation rules to the change in the at least one remaining variable to validate or reject the value of the at least first variable.

12. The system of claim 11, wherein the variational analysis comprises a Monte Carlo simulation performed on the at least first variable.

13. The system of claim 12, wherein the perturbation is a random perturbation, and wherein the Monte Carlo simulation comprises applying the random perturbation to the at least first variable.

14. The system of claim 13, wherein the random perturbation comprises an increase or decrease in the at least first variable based on a randomization range.

15. The system of claim 14, wherein the set of validation rules comprises a limit to the change in the at least one of the remaining variables.

16. The system of claim 15, wherein the limit to the change in the at least one of the remaining variable comprises a function of the randomization range.

17. The system of claim 11, wherein the processor is further configured to repeat the variational analysis on the remainder of the variables in the set of interpolated input data.

18. The system of claim 17, wherein the processor is further configured to repeat the determining a change in at least one of the remaining variables in the set of variables in the set of interpolated input data caused by the variational analysis performed on the remainder of the variables in the set of interpolated input data.

19. The system of claim 18, wherein the processor is further configured to store changed variables caused by the variational analysis across all variables in the set of interpolated input data to generate a set of Monte Carlo interpolation results.

20. The system of claim 11, wherein the set of predetermined input data comprises at least one of a set of financial data, a set of medical data, a set of demographic data, a set of engineering data, a set of network operations data, or a set of geographic data.

21. A non-transitory computer readable medium having instructions that, when executed by a computing device, cause the computing device to perform a method comprising:

accessing, by an interpolation engine running on the computing device, a set of combined input data comprising a set of predetermined input data and a set of interpolated input data, the set of interpolated input data being generated to conformally map the set of combined input data to a set of target output data and comprising a set of variables;

performing, by the interpolation engine, a variational analysis on at least a first variable in the set of variables, wherein the variational analysis comprises applying a perturbation to the at least first variable;

determining a change in at least one remaining variable in the set of variables caused by the variational analysis performed on the at least first variable; and

applying a set of validation rules to the change in the at least one remaining variable to validate or reject a value of the at least first variable.

22. The non-transitory computer readable medium of claim 21, wherein the perturbation is a random perturbation, and wherein the variational analysis comprises a Monte Carlo simulation performed on the at least first variable using the random perturbation of the at least first variable.