Generating representative unstructured data to test artificial intelligence services for bias

1. A computer-implemented bias detection method to evaluate bias in an artificial intelligence service, the method comprising:

selecting a bias context, the bias context having a bias specification associated with the bias context;

generating test data for determining the bias in the artificial intelligence service based on the bias specification and the bias context; and

testing the artificial intelligence service for the bias with the generated test data.

2. The computer-implemented method of claim 1 , wherein the bias specification is modified for a specific type of bias.

3. The computer-implemented method of claim 1 , wherein the bias specification identifies one or more variables and corresponding bias values.

4. The computer-implemented method of claim 1 , wherein the bias context identifies a decision-making scenario and corresponding entities, attributes, regulations, and scope.

5. The computer-implemented method of claim 1 , further comprising learning a correlation between the bias specification and the bias context from the generated test data.

6. The computer-implemented method of claim 1 , wherein the bias specification is hierarchical and extendable.

7. The computer-implemented method of claim 1 , embodied in a cloud-computing environment.

8. A computer program product for bias detection to evaluate bias in an artificial intelligence service, the computer program product comprising a computer-readable storage medium having program instructions embodied therewith, the program instructions executable by a computer to cause the computer to perform:

selecting a bias context, the bias context having a bias specification associated with the bias context;

generating test data for determining the bias in the artificial intelligence service based on the bias specification and the bias context; and

testing the artificial intelligence service for the bias with the generated test data.

9. The computer program product of claim 8 , wherein the bias specification is modified for a specific type of bias.

10. The computer program product of claim 8 , wherein the bias specification identifies one or more variables and corresponding bias values.

11. The computer program product of claim 8 , wherein the bias context identifies a decision-making scenario and corresponding entities, attributes, regulations, and scope.

12. The computer program product of claim 8 , further comprising learning a correlation between the bias specification and the bias context from the generated test data.

13. The computer program product of claim 8 , wherein the bias specification is hierarchical and extendable.

14. A bias detection system to evaluate bias in an artificial intelligence service, the system comprising:

a processor, and

a memory, the memory storing instructions to cause the processor to perform:

selecting a bias context, the bias context having a bias specification associated with the bias context;

generating test data for determining the bias in the artificial intelligence service based on the bias specification and the bias context; and

testing the artificial intelligence service for the bias with the generated test data.

15. The system of claim 14 , wherein the bias specification is modified for a specific type of bias.

16. The system of claim 14 , wherein the bias specification identifies one or more variables and corresponding bias values.

17. The system of claim 14 , wherein the bias context identifies a decision-making scenario and corresponding entities, attributes, regulations, and scope.

18. The system of claim 14 , further comprising learning a correlation between the bias specification and the bias context from the generated test data.

19. The system of claim 14 , wherein the bias specification is hierarchical and extendable.

20. The system of claim 14 , embodied in a cloud-computing environment.