Managing contextual information for queries using data sources managed by personal agents

Patent metadata

Field	Value
Publication number	US-12468767-B1
Application number	US-202418962915-A
Country	US
Kind code	B1
Filing date	Nov 27, 2024
Priority date	Nov 27, 2024
Publication date	Nov 11, 2025
Grant date	Nov 11, 2025

Abstract

Methods and systems for providing customized services to users of data processing systems are disclosed. To do so, a query may be obtained from a user. Contextual information for the query may be attempted to be obtained using a first data source associated with the user. If the portion of the contextual information is stored as a static data chunk in the first data source, the static data chunk may be obtained. If the portion of the contextual information is stored as a reference in the first data source, the reference may be obtained. The reference may indicate that the portion of the contextual information is available as a dynamic data chunk from another data source. The portion of the contextual information may be obtained using the static data chunk and/or the dynamic data chunk and the query may be serviced using at least the portion of the contextual information.

First claim

What is claimed is:

1 . A method for providing customized services to users of data processing systems, the method comprising:
obtaining a query from a first user of the users; identifying that a portion of contextual information relevant to the query is able to be obtained using a first data source associated with the first user, the first data source being managed by a first personal agent assigned to the first user; making a determination regarding whether the portion of the contextual information is stored in the first data source as a static data chunk; in a first instance of the determination in which the portion of the contextual information is stored in the first data source as the static data chunk:
obtaining the static data chunk from the first data source; and
servicing the query using the query as a prompt for an artificial intelligence model and at least the portion of the contextual information obtained from the static data chunk 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 , further comprising:
in a second instance of the determination in which the portion of the contextual information is not stored in the first data source as a static data chunk:
obtaining a reference from the first data source, the reference indicating a second data source in which a dynamic data chunk associated with the portion of the contextual information is stored;
obtaining the dynamic data chunk from the second data source; and
servicing the query using the query as the prompt for the artificial intelligence model and at least the portion of the contextual information obtained from the dynamic data chunk as context for the prompt to obtain the response usable to provision the computer-implemented services to at least the first user.

3 . The method of claim 2 , wherein the first data source serves as a personal context library for the first user for queries submitted to the artificial intelligence model.

4 . The method of claim 1 , wherein an information content of the static data chunk is based on a dynamic data chunk from another data source at a point in time.

5 . The method of claim 1 , further comprising:
prior to obtaining the query:
initiating storage of the static data chunk in the first data source; and
notifying at least a second personal agent assigned to a second user that the portion of the contextual information is stored in the first data source as the static data chunk to enable the second user to obtain copies of the static data chunk as context for prompts submitted by the second user to artificial intelligence models.

6 . The method of claim 5 , wherein the second user has second user characteristics that meet similarity criteria with respect to first user characteristics of the first user.

7 . The method of claim 6 , wherein the first user characteristics comprise at least one type of characteristic selected from a list of types of characteristics consisting of:
a job title for the first user; job duties for the first user; a job division for the first user; and a job ranking for the first user.

8 . The method of claim 5 , further comprising:
prior to obtaining the query and after notifying the at least the second personal agent:
receiving a request for a copy of the static data chunk from the second personal agent; and
providing, in response to the request, a data package comprising an information content of the static data chunk to the second personal agent.

9 . The method of claim 1 , wherein servicing the query comprises:
obtaining, using at least the query and the portion of the contextual information, an ingest data package for an inference model; initiating generation of a response by the inference model using the ingest data package; and providing the response to the first user as a customized service.

10 . The method of claim 9 , wherein the ingest data package is a retrieval-augmented generation (RAG) output from a RAG pipeline process, and the inference model is a large language model (LLM).

11 . 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:
obtaining a query from a first user of the users; identifying that a portion of contextual information relevant to the query is able to be obtained using a first data source associated with the first user, the first data source being managed by a first personal agent assigned to the first user; making a determination regarding whether the portion of the contextual information is stored in the first data source as a static data chunk; in a first instance of the determination in which the portion of the contextual information is stored in the first data source as the static data chunk:
obtaining the static data chunk from the first data source; and
servicing the query using the query as a prompt for an artificial intelligence model and at least the portion of the contextual information obtained from the static data chunk as context for the prompt to obtain a response usable to provision computer-implemented services to at least the first user.

12 . The non-transitory machine-readable medium of claim 11 , wherein the operations further comprise:
in a second instance of the determination in which the portion of the contextual information is not stored in the first data source as a static data chunk:
obtaining a reference from the first data source, the reference indicating a second data source in which a dynamic data chunk associated with the portion of the contextual information is stored;
obtaining the dynamic data chunk from the second data source; and
servicing the query using the query as the prompt for the artificial intelligence model and at least the portion of the contextual information obtained from the dynamic data chunk as context for the prompt to obtain the response usable to provision the computer-implemented services to at least the first user.

13 . The non-transitory machine-readable medium of claim 12 , wherein the first data source serves as a personal context library for the first user for queries submitted to the artificial intelligence model.

14 . The non-transitory machine-readable medium of claim 11 , wherein an information content of the static data chunk is based on a dynamic data chunk from another data source at a point in time.

15 . The non-transitory machine-readable medium of claim 11 , wherein the operations further comprise:
prior to obtaining the query:
initiating storage of the static data chunk in the first data source; and
notifying at least a second personal agent assigned to a second user that the portion of the contextual information is stored in the first data source as the static data chunk to enable the second user to obtain copies of the static data chunk as context for prompts submitted by the second user to artificial intelligence models.

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:
obtaining a query from a first user of the users;
identifying that a portion of contextual information relevant to the query is able to be obtained using a first data source

Assignees

• Dell Products Lp

Inventors

• Ezrielev Ofir
• Liu Tsehsin Jason
• Savir Amihai

Classifications

• G06F16/90335 Primary

  Query processing · CPC title

• G06F16/9035 Primary

  Filtering based on additional data, e.g. user or group profiles · CPC title