Systems and processes for bias removal in a predictive performance model

1. A system comprising:

a memory device; and

at least one hardware processor coupled to the memory device, wherein the at least one hardware processor is configured to at least:

receive a plurality of sets of input data individually describing a respective asset of a plurality of assets associated with an entity;

receive a plurality of inputs responsive to a subset of a plurality of queries of a user;

generate a plurality of predictive models individually generated based at least in part on a respective set of the plurality of sets of input data;

calculate a plurality of predicted outcomes for the user by applying each of the plurality of predictive models to the plurality of inputs; and

generate a user interface comprising the plurality of predictive outcomes for the user for each of the plurality of predictive models.

2. The system of claim 1 , wherein a particular predictive model of the plurality of predictive models comprise an artificial intelligence neural network predictive of performance.

3. The system of claim 2 , wherein the at least one hardware processor is further configured to:

determine that a validity value for the artificial intelligence neural network meets a validity threshold;

responsive to determining that the validity threshold for the artificial intelligence neural network meets the validity threshold, compute a predictive bias of the artificial intelligence neural network associated with at least one non-factored inputs;

in response to the predictive bias of the artificial intelligence neural network exceeding a bias threshold, score a plurality of nodes of the artificial intelligence neural network according to an effect on the predictive bias;

store a rule preventing a combination of parameters associated with a highest scored node of the artificial intelligence neural network;

generate a second artificial intelligence neural network predictive of performance based at least in part on the input data, the plurality of inputs, and the rule; and

calculate one of the plurality of predicted outcomes based on the second artificial intelligence neural network.

4. The system of claim 1 , wherein the plurality of predicted outcomes individually correspond to a respective one of the plurality of predictive models.

5. The system of claim 1 , wherein the at least one hardware processor is further configured to:

identify a plurality of additional users; and

calculate a plurality of sets of additional predicted outcomes individually corresponding to a respective one of the plurality of additional users for the user by applying each of the plurality of predictive models.

6. The system of claim 1 , wherein the at least one hardware processor is further configured to generate a graph comprising a grade associated with each of the plurality of predictive outcomes.

7. The system of claim 1 , wherein each of the plurality of predictive models comprise a respective plurality of compartmentalized variables.

8. A process comprising:

receiving, via at least one hardware processor, a plurality of sets of input data individually describing a respective asset of a plurality of assets associated with an entity;

receiving, via the at least one hardware processor, a plurality of inputs responsive to a subset of a plurality of queries of a user;

generating, via the at least one hardware processor, a plurality of predictive models individually generated based at least in part on a respective set of the plurality of sets of input data;

calculating, via the at least one hardware processor, a plurality of predicted outcomes for the user by applying each of the plurality of predictive models to the plurality of inputs; and

generating, via the at least one hardware processor, a list ranked according to the plurality of predictive outcomes for the user.

9. The process of claim 8 , wherein each of the plurality of predictive models comprise a respective plurality of compartmentalized variables.

10. The process of claim 8 , further comprising receiving, via the at least one hardware processor, a selection of the user for evaluation, wherein the plurality of predicted outcomes are generated for the user in response to the selection.

11. The process of claim 8 , further comprising:

calculating, via the at least one hardware processor, a plurality of additional predicted outcomes for a plurality of additional users using a particular predictive model that corresponds to a particular asset, wherein a plurality of particular users comprises the user and the plurality of additional users; and

generating, via the at least one hardware processor, a second ranked list of the plurality of particular users based on the plurality of additional predictive outcomes and one of the plurality of predictive outcomes that corresponds to the particular predictive model.

12. The process of claim 8 , wherein the plurality of assets are a plurality of job positions and the list is a ranking of the plurality of job positions for the user.

13. The process of claim 8 , further comprising:

identifying, via the at least one hardware processor, a plurality of additional users; and

calculating, via the at least one hardware processor, a plurality of sets of additional predicted outcomes individually corresponding to a respective one of the plurality of additional users for the user by applying each of the plurality of predictive models.

14. The process of claim 8 , further comprising generating, via the at least one hardware processor, a graph comprising a grade associated with each of the plurality of predictive outcomes.

15. A non-transitory computer-readable medium embodying a program that, when executed by at least one hardware processor, causes the at least one hardware processor to:

receive a plurality of sets of input data individually describing a respective asset of a plurality of assets associated with an entity;

receive a plurality of inputs responsive to a subset of a plurality of queries of a user;

generate a plurality of predictive models individually generated based at least in part on a respective set of the plurality of sets of input data;

calculate a plurality of predicted outcomes for the user by applying each of the plurality of predictive models to the plurality of inputs; and

generate a user interface comprising the plurality of predictive outcomes for the user for each of the plurality of predictive models.

16. The non-transitory computer-readable medium of claim 15 , wherein the program further causes the at least one hardware processor to generate a particular one of the predictive models by determining a plurality of coefficients individually corresponding to a plurality of attribute types based on a respective set of the plurality of sets of input data.

17. The non-transitory computer-readable medium of claim 15 , wherein the program further causes the at least one hardware processor to:

identify a plurality of additional users; and

calculate a plurality of sets of additional predicted outcomes individually corresponding to a respective one of the plurality of additional users for the user by applying each of the plurality of predictive models.

18. The non-transitory computer-readable medium of claim 15 , wherein the program further causes the at least one hardware processor to generate a graph comprising a grade associated with each of the plurality of predictive outcomes.

19. The non-transitory computer-readable medium of claim 15 , wherein a particular predictive model of the plurality of predictive models comprise an artificial intelligence neural network predictive of performance.

20. The non-transitory computer-readable medium of claim 19 , wherein the program further causes the at least one hardware processor

determine that a validity value for the artificial intelligence neural network meets a validity threshold;

responsive to determining that the validity threshold for the artificial intelligence neural network meets the validity threshold, compute a predictive bias of the artificial intelligence neural network associated with at least one non-factored inputs;

in response to the predictive bias of the artificial intelligence neural network exceeding a bias threshold, score a plurality of nodes of the artificial intelligence neural network according to an effect on the predictive bias;

store a rule preventing a combination of parameters associated with a highest scored node of the artificial intelligence neural network;

generate a second artificial intelligence neural network predictive of performance based at least in part on the input data, the plurality of inputs, and the rule; and

calculate one of the plurality of predicted outcomes based on the second artificial intelligence neural network.