Supervisory layer for model exchange
Systems and methods for models utilizing siloed data are disclosed. For example, data stored with and/or available to one or more systems may be siloed such that it may not be aggregated and/or shared with other systems. The presently-disclosed systems and methods utilize predictive layers and models to allow each system to predict outcomes using its own data and then models are shared between systems to allow each associated system to gain the benefits of the data of other systems without aggregating such data or otherwise sharing the data.
1. A system comprising:
one or more processors; and
computer-readable media storing computer-executable instructions that, when executed by the one or more processors, cause the one or more processors to perform operations comprising:
storing account data associated with multiple remote systems, the account data indicating at least one of:
data types associated with the remote systems;
predictive layers associated with the remote systems; or
data constraints associated with the remote systems;
receiving, from a remote system, a request to utilize a model with a predictive layer, wherein the model is configured in a first language and the remote system is configured in a second language;
determining a mapping between the first language and the second language;
determining to provide the remote system with the model based at least in part on the mapping;
determining a condition to account for in association with providing the model to the remote system, the condition including at least one of:
a change in a degree of a confidence score associated with an outcome predicted by the predictive layer utilizing the model; or
a data type associated with the model corresponding to a data type associated with the predictive layer; and
sending the model to the remote system based at least in part on the condition being satisfied.
2. The system of claim 1 , further comprising configuring the model based at least in part on the data types associated with the remote systems.
3. The system of claim 1 , further comprising determining an additional condition to account for in association with providing the model to an additional remote system and sending a notification to the additional remote system indicating that the model is available in response to the additional condition being satisfied.
4. The system of claim 1 , further comprising a graphical user interface (GUI) and presenting, via the GUI, a status of a plurality of models associated with the multiple remote systems.
5. The system of claim 1 , further comprising accessing a database storing data associated with the remote system and generating an API based on the data associated with the remote system.
6. The system of claim 5 , further comprising sending the API to the remote system, wherein the model is provided to the remote system via the API.
7. The system of claim 6 , further comprising determining information to surface on a device associated with the remote system via the API based at least in part on historical data associated with the first data type and the second data type.
8. A system comprising:
one or more processors; and
computer-readable media storing first computer-executable instructions that, when executed by the one or more processors, cause the one or more processors to perform operations comprising:
storing account data associated with multiple remote systems, the account data indicating at least one of:
data types associated with the remote systems;
predictive layers associated with the remote systems; or
data constraints associated with the remote systems;
receiving, from a remote system, a request to utilize a model with a predictive layer, wherein the model is configured in a first language and the remote system is configured in a second language;
determining a mapping between the first language and the second language;
determining to provide the remote system with the model based at least in part on the mapping;
determining that a data type associated with the model corresponds to a data type associated with the predictive layer; and
sending the model to the remote system in response to determining that the data type associated with the model corresponds to the data type associated with the predictive layer, the model being configured to generate a feature configured as an input to the predictive layer of the remote system.
9. The system of claim 8 , further comprising receiving an indication of an outcome that the predictive layer is configured to determine and configuring the model to correspond to the outcome prior to sending.
10. The system of claim 8 , further comprising configuring the model based at least in part on the data types associated with the remote systems.
11. The system of claim 8 , further comprising determining an additional condition to account for in association with providing the model to an additional remote system and sending a notification to the additional remote system indicating that the model is available in response to the additional condition being satisfied.
12. The system of claim 8 , further comprising a graphical user interface (GUI) and presenting, via the GUI, a status of a plurality of models associated with the multiple remote systems.
13. The system of claim 8 , further comprising:
determining an outcome that occurs based at least in part on utilizing the feature as in input to the predictive layer;
determining a confidence associated with determining whether the outcome occurs based at least in part on utilizing the feature as an input to the predictive layer;
determining that the confidence satisfies a threshold confidence; and
providing the model to the remote system in response to determining that the confidence satisfies the threshold confidence.
14. A method comprising:
storing account data associated with multiple remote systems, the account data indicating at least one of:
data types associated with the remote systems;
predictive layers associated with the remote systems; or
data constraints associated with the remote systems;
receiving, from a remote system, a request to utilize a model with a predictive layer, the model being configured to generate a feature configured as an input to the predictive layer, wherein the model is configured in a first language and the remote system is configured in a second language;
determining a mapping between the first language and the second language;
determining to provide the remote system with the model based at least in part on the mapping;
determining a change in a degree of a confidence associated with an outcome predicted by the predictive layer utilizing the model; and
sending the model to the remote system in response to determining that the change in the degree of the confidence is within a predefined threshold.
15. The method of claim 14 , further comprising configuring the model based at least in part on the data types associated with the remote systems.
16. The method of claim 14 , further comprising determining an additional condition to account for in association with providing the model to an additional remote system and sending a notification to the additional remote system indicating that the model is available in response to the additional condition being satisfied.
17. The method of claim 16 , further sending an application programming interface (API) to the remote system, wherein the model is provided to the remote system via the API.
18. The method of claim 17 , further comprising determining information to surface on a device associated with the remote system via the API based at least in part on historical data associated with the first data type and the second data type.
19. The method of claim 14 , further comprising determining that the confidence is greater than a threshold confidence and providing the model to the remote system in response to determining that the confidence being greater than the threshold confidence.
20. The method of claim 14 , further comprising presenting, via a graphical user interface (GUI), a status of a plurality of models associated with the multiple remote systems.