User behavior confidence level of automation

1. A method executing on a computing processor of a computing system, the method comprising operations of:

sharing, by a first computing system, an extrapolation model for predicting inventory in at least one product with a second computing system accessible by a user computing device, wherein sharing the extrapolation model by the first computing system comprises:

identifying a plurality of product categories by applying a categorization model to product data that includes product names and product descriptions, wherein the product data is obtained from internal and external sources;

creating a categorization graph for a plurality of business entities from inputs that include the plurality of product categories, published data that includes merchant categorizations, and geographical data;

generating cohort clusters of similar business entities by applying a clustering algorithm to the categorization graph;

using the cohort clusters to learn an extrapolation model for predicting inventory in at least one product; and

providing access to the extrapolation model via a merchant application that runs on an online merchant platform on the second computing system;

making a prediction as to the inventory level in the at least one product at a specific point in time using the extrapolation model for a user of the merchant application, based on input to the extrapolation model obtained from the user via a graphical user interface (GUI) on the user computing device;

based on the prediction, determining that the inventory level in the at least one product is low;

determining a confidence level for the user, wherein the confidence level for the user is based on how many inventory orders, resulting from previous predictions, have been approved by the user; and

autonomously, by the processor, ordering additional inventory for the user without further input from the user, if the confidence level is above a specified threshold, to restock the at least one product.

2. The method of claim 1 , further comprising an operation of:

displaying a view in the GUI of the merchant application notifying the user that the inventory level is low, if the confidence level is below the specified threshold.

3. The method of claim 1 , wherein the inputs further include a plurality of product suppliers.

4. The method of claim 1 , wherein the categorization model uses a naïve Bayesian algorithm.

5. The method of claim 1 , wherein the clustering algorithm uses a K-means algorithm.

6. The method of claim 1 , wherein the extrapolation model is a hybrid model that includes linear extrapolation and seasonal extrapolation.

7. The method of claim 1 , wherein the merchant categorizations include one or both of MCC (merchant category code) codes and NAICS (North American Industry Classification System) codes.

8. The method of claim 1 , wherein the product categorization model is updated using feedback in the form of manual product categorizations.

9. The method of claim 1 , wherein the merchant application is a third-party application distributed through a store hosted by the online merchant platform.

10. A non-transitory computer-readable media configured to store a program, which when executed by a processor, performs:

sharing, by a first computing system, an extrapolation model for predicting inventory in at least one product with a second computing system accessible by a user computing device, wherein sharing the extrapolation model by the first computing system comprises:

identifying a plurality of product categories by applying a categorization model to product data that includes product names and product descriptions, wherein the product data is obtained from internal and external sources;

creating a categorization graph for a plurality of business entities from inputs that include the plurality of product categories, published data that includes merchant categorizations, and geographical data;

generating cohort clusters of similar business entities by applying a clustering algorithm to the categorization graph;

using the cohort clusters to learn an extrapolation model for predicting inventory in at least one product;

providing access to the extrapolation model via a merchant application that runs on an online merchant platform;

making a prediction as to the inventory level in the at least one product at a specific point in time using the extrapolation model for a user of the merchant application, based on input to the extrapolation model obtained from the user via graphical user interface (GUI) on the user computing device;

based on the prediction, determining that the inventory level in at least one product is low;

determining a confidence level for the user, wherein the confidence level for the user is based on how many inventory orders, resulting from previous predictions, have been approved by the user; and

autonomously, by the processor, ordering additional inventory for the user without further input from the user, if the confidence level is above a specified threshold, to restock the at least one product.

11. The computer-readable media of claim 10 , wherein the program, when executed by the processor, performs:

displaying a view in the GUI of the merchant application notifying the user that the inventory level is low, if the confidence level is below the specified threshold.

12. The computer-readable media of claim 10 , wherein the inputs further include a plurality of product suppliers.

13. The computer-readable media of claim 10 , wherein the categorization model uses a naïve Bayesian algorithm.

14. The computer-readable media of claim 10 , wherein the clustering algorithm uses a K-means algorithm.

15. The computer-readable media of claim 10 , wherein the extrapolation model is a hybrid model that includes linear extrapolation and seasonal extrapolation.

16. The computer-readable media of claim 10 , wherein the merchant categorizations include one or both of MCC (merchant category code) codes and NAICS (North American Industry Classification System) codes.

17. The computer-readable media of claim 10 , wherein the product categorization model is updated using feedback in the form of manual product categorizations.

18. The computer-readable media of claim 10 , wherein the merchant application is a third-party application distributed through a store hosted by the online merchant platform.

19. A system, comprising:

a hardware processor and memory; and

software instructions stored in the memory, which when executed by the hardware processor, cause the hardware processor to:

share, by a first computing system, an extrapolation model for predicting inventory in at least one product with a second computing system accessible by a user computing device, wherein sharing the extrapolation model by the first computing system comprises:

identifying a plurality of product categories by applying a categorization model to product data that includes product names and product descriptions, wherein the product data is obtained from internal and external sources;

creating a categorization graph for a plurality of business entities from inputs that include the plurality of product categories, published data that includes merchant categorizations, and geographical data;

generating cohort clusters of similar business entities by applying a clustering algorithm to the categorization graph;

using the cohort clusters to learn an extrapolation model for predicting inventory in at least one product;

providing access to the extrapolation model via a merchant application that runs on an online merchant platform;

make a prediction as to the inventory level in the at least one product at a specific point in time using the extrapolation model for a user of the merchant application, based on input to the extrapolation model obtained from the user via a graphical user interface on the user computing device;

based on the prediction, determine that the inventory level in the at least one product is low;

determine a confidence level for the user, wherein the confidence level for the user is based on how many inventory orders, resulting from previous predictions, have been approved by the user; and

autonomously, by the processor, order additional inventory for the user without further input from the user, if the confidence level is above a specified threshold, to restock the at least one product.

20. The system of claim 19 , further comprising software instructions which cause the hardware processor to:

display a view in the GUI of the merchant application notifying the user that the inventory level is low, if the confidence level is below the specified threshold.