Granted Patent B1

US 12,602,386 · App. 18/962,878 · Granted Apr 14, 2026

Inventors: Ofir Ezrielev (Be'er Sheva, IL); Amihai Savir (Newton, MA); Tsehsin Jason Liu (Wellesley, MA)

Assignee: Dell Products L.P.

G06F16/24575G06F16/285G06F16/951G06F16/9535G06F16/24578

Abstract

Claims (75)

1 . A method for providing customized services to users of data processing systems, the method comprising:

making an identification that sufficient contextual information for a query from a first user of the users is unavailable from a first data source;

based on the identification:

identifying, based on the first user, a plurality of other data sources;

obtaining, from the plurality of other data sources and based on the query, corresponding portions of supplemental contextual information for the query;

obtaining, based on performing a similarity analysis of the corresponding portions of the supplemental contextual information, a consensus response for a portion of the sufficient contextual information that is unavailable from the first data source to obtain the sufficient contextual information, the consensus response including metadata indicating the data sources used to obtain the consensus response;

generating an ingest data package that is a data structure comprising: the query; and the consensus response for the portion of the sufficient contextual information that is unavailable from the first data source; and

servicing the query using the ingest data package, with the query of the ingest data package as a prompt for an artificial intelligence model and with the consensus response of the ingest data package as the sufficient contextual information as context for the prompt to obtain a response usable to provision computer-implemented services to at least the first user.

2 . The method of claim 1 , wherein the supplemental contextual information comprises:

a first data chunk with a first information content obtained from a second data source of the plurality of the other data sources; and

a second data chunk with a second information content obtained from a third data source of the plurality of the other data sources.

3 . The method of claim 2 , wherein the first information content and the second information content relate to a same topic indicated by the query, and the first data source lacking sufficient information related to the same topic.

4 . The method of claim 2 , wherein the first data source is managed by a first personal agent assigned to a first user, the second data source is managed by a second personal agent assigned to a second user, the third data source is managed by a third personal agent assigned to a third user, and the first data source, the second data source, and the third data source serving as personal context libraries for the first user, the second user, and the third user, respectively, for queries submitted to the artificial intelligence model.

5 . The method of claim 4 , wherein a data source hierarchy defines an ordering of at least the first data source, the second data source, and the third data source based on:

first characteristics of the first user,

second characteristics of the second user,

third characteristics of the third user, and

a data source ranking schema; and

wherein the plurality of the data sources are identified using, at least in part, the data source hierarchy.

6 . The method of claim 5 , wherein the second characteristics of the second user comprise at least one characteristic selected from a list of characteristics consisting of:

a job title for the second user;

job duties of the second user; and

metadata for tasks previously performed by the second personal agent for the second user.

7 . The method of claim 5 , wherein the data source ranking schema comprises a rule set for ranking the second data source and the third data source with respect to the first data source based on degrees of similarity between the first characteristics, the second characteristics, and the third characteristics.

8 . The method of claim 7 , wherein obtaining the consensus response comprises:

for a topic indicated by the query and for which the first data source is unable to provide the sufficient contextual information:

identifying a most common response from the corresponding portions of supplemental contextual information for the topic.

9 . The method of claim 8 , wherein obtaining the consensus response further comprises:

identifying a minority response from the corresponding portions of the supplemental contextual information for the topic;

designating the most common response as a required part of the sufficient contextual information; and

designating the minority response as an optional part of the sufficient contextual information.

10 . A non-transitory machine-readable medium having instructions stored therein, which when executed by a processor, cause the processor to perform operations for providing customized services to users of data processing systems, the operations comprising:

making an identification that sufficient contextual information for a query from a first user of the users is unavailable from a first data source;

based on the identification:

identifying, based on the first user, a plurality of other data sources;

obtaining, from the plurality of other data sources and based on the query, corresponding portions of supplemental contextual information for the query;

obtaining, based on performing a similarity analysis of the corresponding portions of the supplemental contextual information, a consensus response for a portion of the sufficient contextual information that is unavailable from the first data source to obtain the sufficient contextual information, the consensus response including metadata indicating the data sources used to obtain the consensus response;

generating an ingest data package that is a data structure comprising: the query; and the consensus response for the portion of the sufficient contextual information that is unavailable from the first data source; and

servicing the query using the ingest data package, with the query of the ingest data package as a prompt for an artificial intelligence model and with the consensus response of the ingest data package as the sufficient contextual information as context for the prompt to obtain a response usable to provision computer-implemented services to at least the first user.

11 . The non-transitory machine-readable medium of claim 10 , wherein the supplemental contextual information comprises:

a first data chunk with a first information content obtained from a second data source of the plurality of the other data sources; and

a second data chunk with a second information content obtained from a third data source of the plurality of the other data sources.

12 . The non-transitory machine-readable medium of claim 11 , wherein the first information content and the second information content relate to a same topic indicated by the query, and the first data source lacking sufficient information related to the same topic.

13 . The non-transitory machine-readable medium of claim 11 , wherein the first data source is managed by a first personal agent assigned to a first user, the second data source is managed by a second personal agent assigned to a second user, the third data source is managed by a third personal agent assigned to a third user, and the first data source, the second data source, and the third data source serving as personal context libraries for the first user, the second user, and the third user, respectively, for queries submitted to the artificial intelligence model.

14 . The non-transitory machine-readable medium of claim 13 , wherein a data source hierarchy defines an ordering of at least the first data source, the second data source, and the third data source based on:

first characteristics of the first user,

second characteristics of the second user,

third characteristics of the third user, and

a data source ranking schema; and

wherein the plurality of the data sources are identified using, at least in part, the data source hierarchy.

15 . The non-transitory machine-readable medium of claim 14 , wherein the second characteristics of the second user comprise at least one characteristic selected from a list of characteristics consisting of:

a job title for the second user;

job duties of the second user; and

metadata for tasks previously performed by the second personal agent for the second user.

16 . A data processing system, comprising:

a processor; and

a memory coupled to the processor to store instructions, which when executed by the processor, cause the processor to perform operations for providing customized services to users of data processing systems, the operations comprising:

making an identification that sufficient contextual information for a query from a first user of the users is unavailable from a first data source;

based on the identification:

identifying, based on the first user, a plurality of other data sources;

obtaining, from the plurality of other data sources and based on the query, corresponding portions of supplemental contextual information for the query;

obtaining, based on performing a similarity analysis of the corresponding portions of the supplemental contextual information, a consensus response for a portion of the sufficient contextual information that is unavailable from the first data source to obtain the sufficient contextual information, the consensus response including metadata indicating the data sources used to obtain the consensus response;

generating an ingest data package that is a data structure comprising: the query; and the consensus response for the portion of the sufficient contextual information that is unavailable from the first data source; and

servicing the query using the ingest data package, with the query of the ingest data package as a prompt for an artificial intelligence model and with the consensus response of the ingest data package as the sufficient contextual information as context for the prompt to obtain a response usable to provision computer implemented services to at least the first user.

17 . The data processing system of claim 16 , wherein the supplemental contextual information comprises:

a first data chunk with a first information content obtained from a second data source of the plurality of the other data sources; and

a second data chunk with a second information content obtained from a third data source of the plurality of the other data sources.

18 . The data processing system of claim 17 , wherein the first information content and the second information content relate to a same topic indicated by the query, and the first data source lacking sufficient information related to the same topic.

19 . The data processing system of claim 17 , wherein the first data source is managed by a first personal agent assigned to a first user, the second data source is managed by a second personal agent assigned to a second user, the third data source is managed by a third personal agent assigned to a third user, and the first data source, the second data source, and the third data source serving as personal context libraries for the first user, the second user, and the third user, respectively, for queries submitted to the artificial intelligence model.

20 . The data processing system of claim 19 , wherein a data source hierarchy defines an ordering of at least the first data source, the second data source, and the third data source based on:

first characteristics of the first user,

second characteristics of the second user,

third characteristics of the third user, and

a data source ranking schema; and

wherein the plurality of the data sources are identified using, at least in part, the data source hierarchy.

Assignments (1)

References Cited (33)