Granted Patent B2

US 12,602,438 · App. 18/635,644 · Granted Apr 14, 2026

Inventors: Muralidhar Krishnaprasad (Seattle, WA); Kristen Chapman (Hopedale, MA)

Assignee: Salesforce, Inc.

G06F16/951G06F16/9024G06F16/9535G06F16/9574

Abstract

Claims (39)

1 . A computer-implemented method comprising:

providing, to a user application running on a user client device, a privacy sandbox comprising one or more application programming interfaces (APIs);

providing, to the privacy sandbox, a query model for generating a topic identifier for each of a plurality of market segment topics;

in response to a first call to one of the one or more APIs, using the query model and first data provided by the user application to generate at least one topic identifier, wherein the first data is stored on the client user device in a storage location only accessible by the user application;

storing the generated at least one topic identifier, wherein the at least one topic identifier can be read by the privacy sandbox; and

providing the stored at least one topic identifier to the privacy sandbox.

2 . The computer-implemented method of claim 1 , further comprising:

receiving second data at the user application,

wherein the second data comprises an advertisement based on the at least one topic identifier.

3 . The computer-implemented method of claim 2 , wherein the user application displays the received second data.

4 . The computer-implemented method of claim 2 , wherein the second data comprises a listing of matched topic identifiers.

5 . The computer-implemented method of claim 1 , the method further comprises:

changing the first data over a period of time.

6 . The computer-implemented method of claim 5 , the method further comprises:

utilizing the changed first data by the query model to generate a second topic identifier for a corresponding market segment topic.

7 . The computer-implemented method of claim 1 , wherein the query model includes a plurality of market segment topic-based queries,

wherein each market segment topic-based query of the plurality of queries matches targeted content within the application.

8 . The computer-implemented method of claim 7 , wherein the query model is configured to determine that a query of the plurality of market segment topic-based queries matches user-specific data using only data stored in the user application.

9 . The computer-implemented method of claim 8 , wherein the query matching the user-specific data stored in the user application is configured to cause the user application to indicate a matched market segment topic.

10 . The computer-implemented method of claim 7 , wherein the targeted content contains user-specific personalized content.

11 . The computer-implemented method of claim 7 , wherein the targeted content contains user-specific content.

12 . The computer-implemented method of claim 1 , wherein the first data comprises user application history data stored in a privacy sandbox of an application data repository, wherein the user application history data indicates a behavior of the user interacting with the user application.

13 . The computer-implemented method of claim 1 , wherein the first data stored in the user application comprises user-specific attribute data.

14 . The computer-implemented method of claim 1 , further comprising replying to a request to provide an updated query model from the user application.

15 . The computer-implemented method of claim 1 , wherein an application-executable library is configured to cause the user application to locally process at least one query from the query model against locally stored at least one of user attribute data and historical user application activity.

16 . The computer-implemented method of claim 15 , wherein the at least one query from the query model is configured to provide instructions to the application-executable library, the instructions configured to collect at least one of the user attribute data and the historical user application activity.

17 . The computer-implemented method of claim 1 , where the query model comprises a prioritized decision tree graph query model.

18 . A computer-implemented system comprising:

a non-transient, tangible memory configured to store computer-executable instructions; and

a processor implemented with physical computing resources configured to retrieve the stored computer-executable instructions and execute the computer-executable instructions on the processor, the computer-executable instructions configured to:

providing, to a user application running on a user client device, a privacy sandbox comprising one or more application programming interfaces (APIs);

providing, to the privacy sandbox, a query model for generating a topic identifier for each of a plurality of market segment topics;

in response to a first call to one of the one or more APIs, using the query model and first data provided by the user application to generate at least one topic identifier, wherein the first data is stored on the client user device in a storage location only accessible by the user application;

storing the generated at least one topic identifier, wherein the at least one topic identifier can be read by the privacy sandbox; and

providing the stored at least one topic identifier to the privacy sandbox.

19 . The computer-implemented system of claim 18 , further comprising:

receiving second data at the user application,

wherein the second data comprises an advertisement based on the at least one topic identifier.

20 . The computer-implemented system of claim 19 , wherein the user application displays the received second data.

Assignments (2)

Continuity (3)

References Cited (12)