Systems and methods for providing combined prediction scores
A prediction system provides combined prediction scores for a target geographic zone within a hierarchical geographical model. The combined prediction scores are generated based on correlations between sales data and household attributes. The combined prediction scores are pre-calculated, stored on a database, and transmitted to a client device based on a request from the client device.
1. A method, comprising:
identifying a hierarchical geographical model including a plurality of geographic zones, wherein the hierarchical geographical model includes a first tier and a second tier, the first tier including a first subset of the plurality of geographic zones and the second tier including a second subset of the plurality of geographic zones wherein the first subset of the plurality of geographic zones have a larger geographic area than the second subset of the plurality of geographic zones;
receiving a plurality of household prediction scores for a plurality of households based on a plurality of attributes;
for each geographic zone of the plurality of geographic zones, determining a combined prediction score based on household prediction scores of the plurality of households that reside within the geographic zone;
based on a request from a client device associated with a geographic zone from the first subset of the plurality of geographic zones, transmitting the prepared combined prediction score for the geographic zone from the first subset of the plurality of geographic zones to the client device;
receiving the prepared combined prediction score for the geographic zone from the first subset of the plurality of geographic zones;
generating, by the client device, a prediction map GUI on the client device comprising user-selectable node icons representing geographic zones from the first subset of the plurality of geographic zones;
detecting, by the client device, a user manipulation of a user-selectable node icon representing a first geographic zone from the first subset of the plurality of geographic zones that decreases a displayed geographic area within the prediction map GUI, wherein the user manipulation of the user-selectable node icon comprises a zoom-in interaction;
determining one or more geographic zones from the second subset of the plurality of geographic zones that are within a geographic area covered by the first geographic zone from the first subset of the plurality of geographic zones;
requesting the prepared combined prediction scores for the one or more geographic zones from the second subset of the plurality of geographic zones that are within the geographic area covered by the first geographic zone from the first subset of the plurality of geographic zones; and
changing a resolution of the prediction map GUI such that the user-selectable node icon representing the first geographic zone from the first subset of the plurality of geographic zones within the prediction map GUI is broken into a plurality of additional user-selectable node icons representing the one or more geographic zones from the second subset of the plurality of geographic zones and the prepared combined prediction scores are overlaid on the plurality of additional user-selectable node icons representing one or more geographic zones from the second subset of the plurality of geographic zones within the prediction map GUI.
2. The method of claim 1 , wherein the at least one of the prepared combined prediction scores is transmitted from a server on a cloud computing system to the client device, and wherein the client device does not calculate the combined prediction score.
3. The method of claim 1 , further comprising preparing the plurality of household prediction scores by inputting the plurality of attributes into a machine learning model, the machine learning model being trained to output a model prediction score for a given household based on attributes associated with the given household.
4. The method of claim 1 , wherein the plurality of attributes include publicly accessible data about the plurality of households.
5. The method of claim 1 , wherein the plurality of attributes include personal information, the personal information including name, age, race, gender, phone number, email address, mortgage status, home type, or home composition.
6. The method of claim 1 , wherein determining the combined prediction score includes pre-calculating a first combined prediction score for the first subset separately from a second combined prediction score for the second subset.
7. The method of claim 1 , wherein the request from the client device includes an index value associated with a node of a target geographic zone of the plurality of geographic zones, and wherein transmitting the at least one of the prepared combined prediction scores includes transmitting the prepared combined prediction scores associated with the index value.
8. The method of claim 1 , wherein each geographic zone of the second subset is encompassed within a single geographic zone of the first subset.
9. A system comprising:
a server device comprising:
at least one processor; and
at least one non-transitory computer-readable storage medium storing instructions that, when executed by the at least one processor, cause the server device to:
identify a hierarchical geographical model including a plurality of geographic zones, wherein the hierarchical geographical model includes a first tier and a second tier, the first tier including a first subset of the plurality of geographic zones and the second tier including a second subset of the plurality of geographic zones wherein the first subset of the plurality of geographic zones have a larger geographic area than the second subset of the plurality of geographic zones;
receive a plurality of household prediction scores for a plurality of households based on a plurality of attributes;
for each geographic zone of the plurality of geographic zones, prepare a combined prediction score based on household prediction scores of the plurality of households that reside within the geographic zone; and
based on a request associated with a geographic zone from the first subset of the plurality of geographic zones, transmit the prepared combined prediction score for the geographic zone from the first subset of the plurality of geographic zones; and
a client device comprising:
at least one processor; and
at least one non-transitory computer-readable storage medium storing instructions that, when executed by the at least one processor, cause the server device to:
receive the prepared combined prediction score for the geographic zone from the first subset of the plurality of geographic zones;
generate a prediction map GUI on the client device comprising user-selectable node icons representing geographic zones from the first subset of the plurality of geographic zones;
detect a user manipulation of a user-selectable node icon representing a first geographic zone from the first subset of the plurality of geographic zones that decreases a displayed geographic area within the prediction map GUI, wherein the user manipulation of the user-selectable node icon comprises a zoom-in interaction;
determine one or more geographic zones from the second subset of the plurality of geographic zones that are within a geographic area covered by the first geographic zone from the first subset of the plurality of geographic zones;
request the prepared combined prediction scores for the one or more geographic zones from the second subset of the plurality of geographic zones that are within the geographic area covered by the first geographic zone from the first subset of the plurality of geographic zones; and
change a resolution of the prediction map GUI such that the user-selectable node icon representing the first geographic zone from the first subset of the plurality of geographic zones within the prediction map GUI is broken into a plurality of additional user-selectable node icons representing the one or more geographic zones from the second subset of the plurality of geographic zones and the prepared combined prediction scores are overlaid on the plurality of additional user-selectable node icons representing one or more geographic zones from the second subset of the plurality of geographic zones within the prediction map GUI.
10. The system of claim 9 , wherein the at least one of the prepared combined prediction scores is transmitted from a server on a cloud computing system to the client device, and wherein the client device does not calculate the combined prediction score.
11. The system of claim 9 , further comprising instructions that, when executed by the at least one processor, cause the system to prepare the plurality of household prediction scores by inputting the plurality of attributes into a machine learning model, the machine learning model being trained to output a model prediction score for a given household based on attributes associated with the given household.
12. The system of claim 9 , wherein the plurality of attributes include publicly accessible data about the plurality of households.
13. The system of claim 9 , wherein preparing the combined prediction score includes preparing a first combined prediction score for the first subset separately from a second combined prediction score for the second subset.
14. The system of claim 9 , wherein the request associated with the geographic zone from the first subset of the plurality of geographic zones includes an index value associated with a node of a target geographic zone of the plurality of geographic zones, and wherein transmitting the at least one of the prepared combined prediction scores includes transmitting the prepared combined prediction scores associated with the index value.
15. A non-transitory computer readable medium storing instructions thereon that, when executed by at least one processor, causes one or more computing devices to:
identify a hierarchical geographical model including a plurality of geographic zones, wherein the hierarchical geographical model includes a first tier and a second tier, the first tier including a first subset of the plurality of geographic zones and the second tier including a second subset of the plurality of geographic zones wherein the first subset of the plurality of geographic zones have a larger geographic area than the second subset of the plurality of geographic zones;
receive a plurality of household prediction scores for a plurality of households based on a plurality of attributes;
for each geographic zone of the plurality of geographic zones, prepare a combined prediction score based on household prediction scores of the plurality of households that reside within the geographic zone;
based on a request from a client device associated with a geographic zone from the first subset of the plurality of geographic zones, transmit the prepared combined prediction score for the geographic zone from the first subset of the plurality of geographic zones;
detect a user manipulation of a user-selectable node icon associated with the first subset of the plurality of geographic zones in a prediction map GUI displayed on the client device that decreases a displayed geographic area within the prediction map GUI, wherein the user manipulation of the user-selectable node icon comprises a zoom-in interaction;
determine one or more geographic zones from the second subset of the plurality of geographic zones that are within a geographic area covered by the geographic zone from the first subset of the plurality of geographic zones;
transmit the prepared combined prediction scores for the one or more geographic zones from the second subset of the plurality of geographic zones that are within the geographic area covered by the geographic zone from the first subset of the plurality of geographic zones; and
change a resolution of the prediction map GUI such that the geographic zone from the first subset of the plurality of geographic zones within the prediction map GUI is broken into the one or more geographic zones from the second subset of the plurality of geographic zones and the prepared combined prediction scores are overlaid on the one or more geographic zones from the second subset of the plurality of geographic zones.
16. The non-transitory computer readable medium of claim 15 , wherein the prepared combined prediction score for the geographic zone from the first subset of the plurality of geographic zones is transmitted from a server on a cloud computing system to the client device.
17. The non-transitory computer readable medium of claim 16 , further storing instructions thereon that, when executed by at least one processor, causes one or more computing devices to, in response to an additional detected user manipulation of the user-selectable node icon associated with the first subset of the plurality of geographic zones in the prediction map GUI displayed on the client device that increases the displayed geographic area within the prediction map GUI:
transmit an updated combined prediction score for the geographic zone from the first subset of the plurality of geographic zones; and
change the resolution of the prediction map GUI such that the one or more geographic zones from the second subset of the plurality of geographic zones within the prediction map GUI are collapsed into the geographic zone from the first subset of the plurality of geographic zones and the updated combined prediction score for the geographic zone from the first subset of the plurality of geographic zones are overlaid on the geographic zone from the first subset of the plurality of geographic zones.
18. The non-transitory computer readable medium of claim 15 , wherein the plurality of attributes include publicly accessible data about the plurality of households.
19. The non-transitory computer readable medium of claim 15 , wherein the plurality of attributes include personal information, the personal information including name, age, race, gender, phone number, email address, mortgage status, home type, or home composition.
20. The non-transitory computer readable medium of claim 15 , wherein determining the combined prediction score includes pre-calculating a first combined prediction score for the first subset separately from a second combined prediction score for the second subset.