Artificial multispectral metadata generator

1. A computer-implemented method comprising:

receiving, by a computer system, an input data set;

determining, by the computer system, attributes of the input data set to be transformed;

for the attribute to be transformed, generating, by the computer system, three or more random noise sets, using at least two noise generation methods;

creating, by the computer system, an amalgamated random set from the three or more random noise sets using a programmable ratio;

randomly sampling, by the computer system, from the amalgamated random set to create a raw noise set;

using, by the computer system, a standard deviation of the attribute to be transformed to create a noise base set;

multiplying, by the computer system, the noise base set with the raw noise set to obtain a final noise set for the attribute to be transformed; and

adding, by the computer system, the final noise set to the input data set to obtain an artificial data set.

2. The method of claim 1 , wherein the at least two noise generation methods include a pink noise generator having a power spectral density per unit of bandwidth proportional to 1/f β , wherein β is equal to 1.

3. The method of claim 1 , wherein the at least two noise generation methods include a violet noise generator having a power spectral density per unit of bandwidth proportional to 1/f β , wherein β is equal to 2.

4. The method of claim 1 , wherein the at least two noise generation methods include a brown noise generator having a power spectral density per unit of bandwidth proportional to f β , wherein β is equal to 2.

5. The method of claim 1 , further comprising using a random hexadecimal substitution to anonymize the attribute.

6. The method of claim 1 , wherein randomly sampling from the amalgamated random set includes randomly sampling with replacement from the amalgamated random set.

7. The method of claim 1 , further comprising using distributional attributes from the final noise set to gain insights from the input data set without having access to the input data set.

8. The method of claim 1 , wherein creating the amalgamated random set from the three or more random noise sets using the programmable ratio includes using a Fibonacci number sequence.

9. A system comprising:

a computing system comprising one or more processors and a data storage system in communication with the one or more processors, wherein the data storage system comprises instructions thereon that, when executed by the one or more processors, causes the one or more processors to:

receive an input data set;

determine an attribute of the input data set to be transformed;

for the attribute to be transformed, generate three or more random noise sets, using at least two noise generation methods;

create an amalgamated random set from the three or more random noise sets using a programmable ratio;

randomly sample from the amalgamated random set to create a raw noise set;

use a standard deviation of the attribute to be transformed to create a noise base set;

multiply the noise base set with the raw noise set to obtain a final noise set for the attribute to be transformed; and

add the final noise set to the input data set to obtain an artificial data set.

10. The system of claim 9 , wherein creating the amalgamated random set from the three or more random noise sets using the programmable ratio includes using a Fibonacci number sequence.

11. The system of claim 9 , wherein randomly sampling from the amalgamated random set includes randomly sampling with replacement from the amalgamated random set.

12. The system of claim 9 , further comprising using distributional attributes from the final noise set to gain insights from the input data set without having access to the input data set.

13. The system of claim 9 , wherein the at least two noise generation methods include a pink noise generator having a power spectral density per unit bandwidth proportional to 1/f β , wherein β is equal to 2.

14. The system of claim 9 , wherein the at least two noise generation methods include a violet noise generator having a power spectral density per unit of bandwidth proportional to f β , wherein β is equal to 2.

15. The system of claim 9 , wherein the at least two noise generation methods include a brown noise generator having a power spectral density per unit of bandwidth proportional to 1/f β , wherein β is equal to 2.

16. A non-transitory computer-readable storage medium, the non-transitory computer-readable storage medium including instructions that, when executed by computers, cause the computers to perform operations of:

receiving an input data set;

determining an attribute of the input data set to be transformed;

for the attribute to be transformed, generating three or more random noise sets, using at least two noise generation methods;

creating an amalgamated random set from the three or more random noise sets using a programmable ratio;

randomly sampling from the amalgamated random set to create a raw noise set;

using a standard deviation of the attribute to be transformed to create a noise base set;

multiplying the noise base set with the raw noise set to obtain a final noise set for the attribute to be transformed; and

adding the final noise set to the input data set to obtain an artificial data set.

17. The non-transitory computer-readable storage medium of claim 16 , further comprising using a random hexadecimal substitution to anonymize the attribute.

18. The non-transitory computer-readable storage medium of claim 16 , wherein randomly sampling from the amalgamated random set includes randomly sampling with replacement from the amalgamated random set.

19. The non-transitory computer-readable storage medium of claim 16 , further comprising using distributional attributes from the final noise set to gain insights from the input data set without having access to the input data set.

20. The non-transitory computer-readable storage medium of claim 16 , wherein creating the amalgamated random set from the three or more random noise sets using the programmable ratio includes using a Fibonacci number sequence.