Granted Patent B2

US 12,380,175 · App. 18/227,423 · Granted Aug 5, 2025

Inventor: Shravan Rayanchu (Mountain View, CA)

Assignee: GOOGLE LLC

G06F16/9537G06F16/2228G06F16/248G06F16/29G06F16/3331G06F16/954G06F40/205G06N20/00G06Q30/0256G06Q30/0261H04L43/16H04L51/222H04L67/52H04W4/021H04W4/025

Abstract

Claims (42)

1. A method of selecting content for a computing device, comprising:

receiving, by one or more processors of a data processing system, from a client computing device, location data of the client computing device;

identifying, by the data processing system, an entity around a location of the client computing device;

generating, by the data processing system, a database indicating a plurality of radius thresholds associated with a plurality of respective feature combinations, at least in part by (i) using a machine learning model to predict performance metric values associated with different distance values and different feature combinations and (ii) comparing the predicted performance metric values to a performance metric criterion to identify the radius thresholds;

identifying, by the data processing system, a radius threshold corresponding to the entity based on (i) a particular feature combination associated with the entity and the client computing device and (ii) a particular radius threshold that is associated with the particular feature combination in the database;

determining, by the data processing system, that a distance between the client computing device and the entity is less than the radius threshold corresponding to the entity; and

transmitting, by the data processing system, responsive to the determination, to the client computing device, a content item associated with the entity.

2. The method of claim 1 , wherein:

at least one feature combination of the plurality of respective feature combinations includes a set of features associated with at least one of a plurality of entities, a plurality of client computing devices, or search queries from the plurality of client computing devices; and

the performance metric values associated with the different distance values and the different feature combinations include indications of interest in content items associated with the plurality of entities.

3. The method of claim 2 , wherein the particular feature combination includes at least one of a popularity of the entity, a search query vertical associated with the search query, query data associated with the search query, a location of the client computing device, or an activity of the client computing device.

4. The method of claim 3 , wherein the particular feature combination includes the popularity of the entity, and wherein the popularity of the entity is positively correlated to a distance from the entity to a client computing device that interacts with a reference associated with the entity.

5. The method of claim 1 , wherein the performance metric values associated with the different distance values and the different feature combinations include values for at least one of click through rate, conversion rate, rate of requests for directions to a business location, predicted click through rate, cost per click, predicted cost per click, or return on investment.

6. The method of claim 1 , wherein the performance metric criterion includes a requirement that a predicted performance metric value associated with the identified radius threshold exceeds a performance metric threshold.

7. The method of claim 1 , wherein the performance metric criterion includes a requirement that a predicted performance metric value associated with the identified radius threshold have a greatest value among a plurality of performance metric values associated with the entity.

8. The method of claim 1 , wherein transmitting the content item associated with the entity includes:

identifying a first set of candidate content items based on a search query from the client computing device;

selecting a second set of candidate content items from the first set of candidate content items based on indications of interest associated with the first set of candidate content items;

identifying one or more candidate content items of the second set of candidate content items, the one or more candidate content items being associated with the entity;

selecting the content item from the one or more candidate content; and

transmitting, to the client computing device, the content item of the one or more candidate content items.

9. A system for selecting content for a computing device, comprising:

a data processing system comprising one or more processors, and a network interface in communication with a client computing device;

wherein the one or more processors are configured to:

receive, from the client computing device, location data of the client computing device;

identify an entity around a location of the client computing device;

generate a database indicating a plurality of radius thresholds associated with a plurality of respective feature combinations, at least in part by (i) using a machine learning model to predict performance metric values associated with different distance values and different feature combinations and (ii) comparing the predicted performance metric values to a performance metric criterion to identify the radius thresholds;

identify a radius threshold corresponding to the entity based on (i) a particular feature combination associated with the entity and the client computing device and (ii) a particular radius threshold that is associated with the particular feature combination in the database;

determine that a distance between the client computing device and the entity is less than the radius threshold corresponding to the entity; and

transmit, responsive to the determination, to the client computing device, a content item associated with the entity.

10. The system of claim 9 , wherein:

at least one feature combination of the plurality of respective feature combinations includes a set of features associated with at least one of a plurality of entities, a plurality of client computing devices, or search queries from the plurality of client computing devices; and

the performance metric values associated with the different distance values and the different feature combinations include indications of interest in content items associated with the plurality of entities.

11. The system of claim 10 , wherein the particular feature combination includes at least one of a popularity of the entity, a search query vertical associated with the search query, query data associated with the search query, a location of the client computing device, or an activity of the client computing device.

12. The system of claim 9 , wherein the performance metric criterion includes a requirement that a predicted performance metric value associated with the identified radius threshold exceeds a performance metric threshold.

13. The system of claim 9 , wherein the performance metric criterion includes a requirement that a predicted performance metric value associated with the identified radius threshold have a greatest value among a plurality of performance metric values associated with the entity.

14. The system of claim 9 , wherein to transmit the content item associated with the entity, the one or more processors are further configured to:

identify a first set of candidate content items based on a search query from the client computing device;

select a second set of candidate content items from the first set of candidate content items based on indications of interest associated with the first set of candidate content items;

identify one or more candidate content items of the second set of candidate content items, the one or more candidate content items being associated with the entity;

select the content item from the one or more candidate content; and

transmit, to the client computing device, the content item.

Assignments (2)

Continuity (4)

References Cited (81)