Facilitating use of select hyper-local data sets for improved modeling

1. A computer-implemented method comprising:

receiving, by a workbench platform, customer data associated with a customer;

determining, using the customer data, an accuracy of a customer model by training the customer model against a first portion of the customer data and testing the customer model using a second portion of the customer data to determine how accurate the customer model is in predicting results;

identifying a location associated with the customer data;

filtering all data sets accessible by the workbench platform by the location to generate a plurality of hyper-local data sets, wherein a hyper-local data set is a data set that comprises data specific to a predetermined radius of the location;

for each of the plurality of hyper-local data sets, determining an increase in model accuracy attributed to a supplemental use of the respective hyper-local data set with the customer data as additional training data, wherein determining the increase in model accuracy comprises:

training the customer model against the first portion of the customer data and the respective hyper-local data set; and

testing the customer model using the second portion of the customer data to determine how accurate the customer model is in predicting results when supplemented with the respective hyper-local data set;

clustering at least one group of hyper-local data sets of the plurality of hyper-local data sets into one or more buckets corresponding to a predetermined range of accuracy improvement; and

for each bucket, facilitating use of a select one or more of the hyper-local data sets of the at least one group of hyper-local data sets in generating an improved accuracy customer model.

2. The computer-implemented method of claim 1 , wherein the customer model comprises a machine learning model that is generated using the first portion of the customer data as training data.

3. The computer-implemented method of claim 1 , wherein facilitating use of a select one or more of the hyper-local data sets of the at least one group of hyper-local data sets comprises:

notifying an owner of a first hyper-local data set of the at least one group of hyper-local data sets of the difference in model accuracy associated with the first hyper-local data set;

notifying the owner of the difference in model accuracy associated with one or more second hyper-local data sets of the at least one group of hyper-local data sets that are associated with one or more other owners; and

receiving an offer from the owner, wherein the offer represents a price of the first hyper-local data set that is offered for sale to the customer.

4. The computer-implemented method of claim 1 , wherein the workbench platform prevents export or copying of any hyper-local data set by the customer.

5. The computer-implemented method of claim 1 , wherein facilitating use of a select one or more of the hyper-local data sets of the at least one group of hyper-local data sets comprises:

providing, to the customer, one or more offers associated with one or more corresponding hyper-local data sets of the at least one group of hyper-local data sets; and

receiving, from the customer, an acceptance of one of the one or more of the offers.

6. The computer-implemented method of claim 1 , wherein facilitating use of a select one or more of the hyper-local data sets of the at least one group of hyper-local data sets comprises automatically accepting on behalf of the customer, one of a lowest price offer or a best value offer associated with a hyper-local data set of the at least one group of hyper-local data sets.

7. A system comprising:

a memory having computer readable instructions; and

one or more processors for executing the computer readable instructions, the computer readable instructions controlling the one or more processors to perform operations comprising:

receiving, by a workbench platform, customer data associated with a customer;

determining, using the customer data, an accuracy of a customer model by training the customer model against a first portion of the customer data and testing the customer model using a second portion of the customer data to determine how accurate the customer model is in predicting results;

identifying a location associated with the customer data;

filtering all data sets accessible by the workbench platform by the location to generate a plurality of hyper-local data sets, wherein a hyper-local data set is a data set that comprises data specific to a predetermined radius of the location;

for each of the plurality of hyper-local data sets, determining an increase in model accuracy attributed to a supplemental use of the respective hyper-local data set with the customer data as additional training data, wherein determining the increase in model accuracy comprises:

training the customer model against the first portion of the customer data and the respective hyper-local data set; and

testing the customer model using the second portion of the customer data to determine how accurate the customer model is in predicting results when supplemented with the respective hyper-local data set;

clustering at least one group of hyper-local data sets of the plurality of hyper-local data sets into one or more buckets corresponding to a predetermined range of accuracy improvement; and

for each bucket, facilitating use of a select one or more of the hyper-local data sets of the at least one group of hyper-local data sets in generating an improved accuracy customer model.

8. The system of claim 7 , wherein the customer model comprises a machine learning model that is generated using the first portion of the customer data as training data.

9. A computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a computer processor to cause the computer processor to perform a method comprising:

receiving, by a workbench platform, customer data associated with a customer;

determining, using the customer data, an accuracy of a customer model by training the customer model against a first portion of the customer data and testing the customer model using a second portion of the customer data to determine how accurate the customer model is in predicting results;

identifying a location associated with the customer data;

filtering all data sets accessible by the workbench platform by the location to generate a plurality of hyper-local data sets, wherein a hyper-local data set is a data set that comprises data specific to a predetermined radius of the location;

for each of the plurality of hyper-local data sets, determining an increase in model accuracy attributed to a supplemental use of the respective hyper-local data set with the customer data as additional training data, wherein determining the increase in model accuracy comprises:

training the customer model against the first portion of the customer data and the respective hyper-local data set; and

testing the customer model using the second portion of the customer data to determine how accurate the customer model is in predicting results when supplemented with the respective hyper-local data set;

clustering at least one group of hyper-local data sets of the plurality of hyper-local data sets into one or more buckets corresponding to a predetermined range of accuracy improvement; and

for each bucket, facilitating use of a select one or more of the hyper-local data sets of the at least one group of hyper-local data sets in generating an improved accuracy customer model.

10. The computer program product of claim 9 , wherein the customer model comprises a machine learning model that is generated using the first portion of the customer data as training data.