Granted Patent B2

US 12,524,620 · App. 18/345,905 · Granted Jan 13, 2026

Inventors: Yu Jia (Kirkland, WA); Kevin M. Chuang (San Francisco, CA); Dixon Lo (San Francisco, CA); Praveen Kumar Bodigutla (San Jose, CA); Sandeep Kumar Jha (San Jose, CA)

Assignee: Microsoft Technology Licensing, LLC

G06F40/35G06F16/3329G06F40/20H04L51/52

Abstract

Claims (77)

1 . A method comprising:

receiving input via a client device comprising (i) a graphical user interface (GUI) associated with a profile and (ii) a profile interface, wherein the input is generated by an interaction with the profile interface;

extracting attribute data from the profile in response to receiving the input;

determining an identifier for the profile based on the attribute data, wherein the identifier comprises a characteristic of a user associated with the input;

determining a subset of attributes of the attribute data using the identifier;

determining an optional attribute for the user based on the characteristic of the user;

based on the input, including the optional attribute in the subset of attributes or excluding the optional attribute from the subset of attributes;

mapping the subset of attributes to a set of prompt inputs;

creating a prompt using the set of prompt inputs mapped to the subset of attributes, wherein the prompt comprises an instruction to a generative machine learning model to cause the generative machine learning model to generate a suggestion for adding content to the profile using the subset of attributes;

applying the generative machine learning model to the prompt;

outputting, by the generative machine learning model, based on the prompt, the suggestion; and

sending the suggestion to the client device for presentation via the profile interface.

2 . The method of claim 1 , further comprising:

receiving historical activity data for the profile, wherein determining the identifier is further based on the historical activity data.

3 . The method of claim 1 , further comprising:

in response to receiving the input, sending a plurality of profile section update options to the client device for presentation via the profile interface; and

receiving, from the client device, a profile section update selection of the plurality of profile section update options based on an interaction with the profile interface.

4 . The method of claim 3 , wherein sending the suggestion comprises:

sending the suggestion to the client device for presentation via a section of the profile interface based on the profile section update selection.

5 . The method of claim 3 , further comprising:

generating a performance parameter based on the suggestion.

6 . The method of claim 5 , wherein generating the performance parameter comprises:

receiving, from the client device, feedback on the suggestion.

7 . The method of claim 5 , further comprising:

training a machine learning model using the set of prompt inputs and the performance parameter; and

generating an updated set of prompt inputs using the trained machine learning model.

8 . The method of claim 5 , further comprising:

mapping an updated subset of attributes of the attribute data to the set of prompt inputs based on the performance parameter; and

creating an updated prompt using the set of prompt inputs mapped to the updated subset of attributes.

9 . The method of claim 5 , further comprising:

in response to the performance parameter satisfying a threshold, generating a suggestion example based on the subset of attributes and the profile section update selection, wherein creating the prompt further uses the suggestion example.

10 . The method of claim 9 , further comprising:

creating a training set using the subset of attributes and a set of examples; and

applying the training set to a suggestion example machine learning model, wherein generating the suggestion example uses an output of the suggestion example machine learning model.

11 . The method of claim 5 , further comprising:

applying an inference machine learning model to the suggestion, wherein generating the performance parameter is based on an output of the inference machine learning model.

12 . The method of claim 1 , wherein the attribute data comprises a set of mandatory attributes and a set of optional attributes,

and wherein mapping the subset of attributes to the set of prompt inputs comprises mapping the set of mandatory attributes and the set of optional attributes to the set of prompt inputs.

13 . The method of claim 1 , wherein the client device provides the profile interface displayed concurrently with the profile, and the profile interface is distinct from the GUI.

14 . A system comprising:

at least one memory device; and

at least one processing device, operatively coupled with the at least one memory device, to:

receive input via a client device comprising (i) a graphical user interface (GUI) associated with a profile and (ii) a profile interface, wherein the input is generated by an interaction with the profile interface;

extract attribute data from the profile in response to receiving the input;

determine an identifier for the profile based on the attribute data, wherein the identifier comprises a characteristic of a user associated with the input;

determine a subset of attributes of the attribute data using the identifier;

determine an optional attribute for the user based on the characteristic of the user;

include or exclude the optional attribute from the subset of attributes based on the input;

map the subset of attributes to a set of prompt inputs;

create a prompt using the set of prompt inputs mapped to the subset of attributes, wherein the prompt comprises an instruction to a generative machine learning model to cause the generative machine learning model to generate a suggestion for adding content to the profile using the subset of attributes;

apply the generative machine learning model to the prompt;

output, by the generative machine learning model, based on the prompt, the suggestion; and

send the suggestion to the client device for presentation via the profile interface.

15 . The system of claim 14 , wherein the at least one processing device further:

receives historical activity data for the profile, wherein determining the identifier is further based on the historical activity data.

16 . The system of claim 14 , wherein the at least one processing device further:

in response to receiving the input, sends a plurality of profile section update options to the client device for presentation via the profile interface; and

receives, from the client device, a profile section update selection of the plurality of profile section update options based on an interaction with the profile interface.

17 . The system of claim 16 , wherein sending the suggestion comprises:

sending the suggestion to the client device for presentation via a section of the profile interface based on the profile section update selection.

18 . At least one non-transitory computer-readable storage medium comprising at least one instruction that, when executed by at least one processing device, causes the at least one processing device to:

receive input via a client device comprising (i) a graphical user interface (GUI) associated with a profile and (ii) a profile interface, wherein the input is generated by an interaction with the profile interface;

extract attribute data from the profile in response to receiving the input;

determine an identifier for the profile based on the attribute data, wherein the identifier comprises a characteristic of a user associated with the input;

determine a subset of attributes of the attribute data using the identifier;

determine that an attribute of the attribute data is an optional attribute for the user based on the characteristic of the user;

based on the input, include the optional attribute in the subset of attributes or exclude the optional attribute from the subset of attributes;

map the subset of attributes to a set of prompt inputs;

create a prompt using the set of prompt inputs mapped to the subset of attributes, wherein the prompt comprises an instruction to a generative machine learning model to cause the generative machine learning model to generate a suggestion for adding content to the profile using the subset of attributes;

apply the generative machine learning model to the prompt;

output, by the generative machine learning model, based on the prompt, the suggestion; and

send the suggestion to the client device for presentation via the profile interface.

19 . The at least one non-transitory computer-readable storage medium of claim 18 , wherein the at least one processing device further:

receives historical activity data for the profile, wherein determining the identifier further uses the historical activity data.

20 . The at least one non-transitory computer-readable storage medium of claim 18 , wherein the at least one processing device further:

in response to receiving the input, sends a plurality of profile section update options to the client device for presentation via the profile interface; and

receives, from the client device, a profile section update selection of the plurality of profile section update options based on an interaction with the profile interface.

Assignments (1)

Continuity (3)

References Cited (30)