Synthetic and traditional data stewards for selecting, optimizing, verifying and recommending one or more datasets
Confirmation of data selection in a zero-trust environment is provided. In some embodiments, a synthetic data steward and/or a traditional data steward can receive the dataset(s). Additionally, a script is received from the algorithm developer. The dataset(s) and script(s) reside within a secure computing node and are therefore inaccessible by any party. The script(s) are executed, resulting in at least one confirmation about the data within the dataset(s). The script(s) complete any of confirming a format for data in the at least one dataset, the expected class values for data within the at least one dataset, an overall characterization and completeness of the at least one dataset, and/or an expected class membership for different data attributes within the at least one dataset.
1. A computerized method for dataset selection confirmation within a zero-trust computing system, the method comprising:
receiving at a sequestered computing node at least one dataset;
receiving at the sequestered computing node a set of requirements from an algorithm developer;
applying the set of requirements to the at least one dataset to generate a tabular format for each at least one dataset;
converting each of the tabular formatted at least one dataset into a vector set;
receive a target vector;
generating various combinations of the at least one dataset to generate combination vector sets;
applying a cost function to minimize a difference between the target vector and one of the combination vector sets; and
selecting the datasets forming the minimized combination vector set for further processing.
2. The method of claim 1 , wherein the set of requirements confirms a format for data in the at least one dataset.
3. The method of claim 1 , wherein the set of requirements confirms expected class values for data within the at least one dataset.
4. The method of claim 1 , wherein the set of requirements confirms an overall characterization and completeness of the at least one dataset.
5. The method of claim 1 , wherein the set of requirements confirms an expected class membership for different data attributes within the at least one dataset.
6. The method of claim 5 , wherein the expected class membership is the target vector.
7. The method of claim 1 , wherein the sequestered computing node is within one of a data steward or a synthetic data steward.
8. The method of claim 7 , wherein the synthetic data steward aggregated the at least one dataset from a plurality of data stewards.
9. The method of claim 1 , wherein the cost function applies a penalty to conditions responsive to the algorithm developer.
10. A zero-trust computing system for dataset selection confirmation comprising:
a data store for receiving at a sequestered computing node at least one dataset, and receiving at the sequestered computing node a set of requirements from an algorithm developer; and
a processor configured to apply the set of requirements to the at least one dataset to generate a tabular format for each at least one dataset, convert each of the tabular formatted at least one dataset into a vector set, receive a target vector, generate various combinations of the at least one dataset to generate combination vector sets, apply a cost function to minimize a difference between the target vector and one of the combination vector sets, and selecting the datasets forming the minimized combination vector set for further processing.
11. The zero-trust computing system of claim 10 , wherein the set of requirements confirms a format for data in the at least one dataset.
12. The zero-trust computing system of claim 10 , wherein the set of requirements confirms expected class values for data within the at least one dataset.
13. The zero-trust computing system of claim 10 , wherein the set of requirements confirms an overall characterization and completeness of the at least one dataset.
14. The zero-trust computing system of claim 10 , wherein the set of requirements confirms an expected class membership for different data attributes within the at least one dataset.
15. The zero-trust computing system of claim 14 , wherein the expected class membership is the target vector.
16. The zero-trust computing system of claim 10 , wherein the sequestered computing node is within one of a data steward or a synthetic data steward.
17. The zero-trust computing system of claim 16 , wherein the synthetic data steward aggregated the at least one dataset from a plurality of data stewards.
18. The zero-trust computing system of claim 10 , wherein the cost function applies a penalty to conditions responsive to the algorithm developer.