Universal reporting using natural language queries

Patent metadata

Field	Value
Publication number	US-12248466-B2
Application number	US-202318361476-A
Country	US
Kind code	B2
Filing date	Jul 28, 2023
Priority date	Jul 28, 2023
Publication date	Mar 11, 2025
Grant date	Mar 11, 2025

Abstract

Methods, systems, and devices for managing report requests are described. A device associated with an identity management platform may receive a message from a user of an organization via a client device. The message may include a natural language and may indicate a request for information associated with the organization. In response to the request, the device may generate a first query based on translating the message into an intermediary language using a machine learning model. The intermediary language may be associated with the identity management platform. The device may transmit a report to the user via the client device. The report may be based on the first query and include the information associated with the organization.

First claim

What is claimed is:

1. A method for managing report requests at a device associated with an identity management platform, comprising:
receiving a natural language message from a user of an organization via a client device, wherein the natural language message indicates a request for information associated with the organization;
generating, via a machine learning model, a first query based at least in part on translating the natural language message into an intermediary language using the machine learning model, wherein the intermediary language is associated with the identity management platform;
validating the first query within the intermediary language;
translating the first query into a second intermediary language associated with a database comprising the information associated with the organization that is requested via the natural language message, wherein translating the first query into the second intermediary language is based at least in part on validating the first query; and
transmitting a report to the user via the client device, wherein the report comprises the information associated with the organization, and wherein the report comprises the information based at least in part on generating the first query, validating the first query, and translating the first query from the intermediary language into the second intermediary language.

2. The method of claim 1 , wherein the intermediary language comprises a first domain-specific language and the second intermediary language comprises a second domain-specific language, the method further comprising:
generating a second query based at least in part on translating the first query into the second domain-specific language using a compiler, wherein the second domain-specific language is associated with the database comprising the information associated with the organization; and
generating the report based at least in part on executing the second query.

3. The method of claim 2 , wherein validating the first query comprises:
transmitting the first query to the user of the organization via the client device; and
receiving a second natural language message from the user of the organization via the client device, wherein the second natural language message comprises an indication of a modification to the first query, and wherein translating the first query into the second domain-specific language comprises:
translating the modification of the first query into the second domain-specific language.

4. The method of claim 3 , wherein the second natural language message indicates the modification to the request for information associated with the organization, or the second natural language message comprises the modification of the first query.

5. The method of claim 2 , wherein validating the first query comprises:
generating an intermediate representation of the first query; and
validating the intermediate representation based at least in part on a schema pre-defined by the database, wherein translating the first query into the second domain-specific language is based at least in part on the validating.

6. The method of claim 2 , further comprising:
determining that the first query satisfies one or more constraints configured by the identity management platform, wherein translating the first query into the second domain-specific language is based at least in part on the determining.

7. The method of claim 6 , wherein determining that the first query satisfies the one or more constraints comprises:
determining that the first query only requests information associated with the organization.

8. The method of claim 1 , further comprising:
training the machine learning model to translate the natural language message into the intermediary language, wherein translating the natural language message into the first query is based at least in part on the training.

9. The method of claim 8 , wherein the information associated with the organization is based at least in part on data obtained by the identity management platform, and wherein training the machine learning model exposes the machine learning model to a structure of data.

10. The method of claim 1 , wherein the machine learning model comprises a large language model.

11. The method of claim 1 , wherein the information associated with the organization comprises information associated with resources of the organization that are managed by the identity management platform, information associated with users of the resources, information associated with security events pertaining to the users, or information associated with security events pertaining to the resources, or any combination thereof.

12. A device for managing report requests, comprising:
one or more memories storing processor-executable code; and
one or more processors coupled with the one or more memories and individually or collectively operable to execute the code to cause the device to:
receive a natural language message from a user of an organization via a client device, wherein the natural language message indicates a request for information associated with the organization;
generate, via a machine learning model, a first query based at least in part on translating the natural language message into an intermediary language using the machine learning model, wherein the intermediary language is associated with an identity management platform;
validate the first query within the intermediary language;

translate the first query into a second intermediary language associated with a database comprising the information associated with the organization that is requested via the natural language message, wherein translating the first query into the second intermediary language is based at least in part on validating the first query; and
transmit a report to the user via the client device, wherein the report comprises the information associated with the organization, and wherein the report comprises the information based at least in part on generating the first query, validating the first query, and translating the first query from the intermediary language into the second intermediary language.

13. The device of claim 12 , wherein the intermediary language comprises a first domain-specific language and the second intermediary language comprises a second domain-specific language, and the one or more processors are individually or collectively further operable to execute the code to cause the device to:
generate a second query based at least in part on translating the first query into the second domain-specific language using a compiler, wherein the second domain-specific language is associated with the database comprising the information associated with the organization; and
generate the report based at least in part on executing the second query.

14. The device of claim 13 , wherein, to validate the first query, the one or more processors are individually or collectively operable to execute the code to cause the device to:
transmit the first query to the user of the organization via the client device; and
receive a second natural language message from the user of the organization via the client device, wherein the second natural language message comprises an indication of a modification to the first query, and wherein, to translate the first query into the second domain-specific language, the one or more processors are individually or collectively operable to execute the code to cause the device to:
translate the modification of the first query into the second domain-specific language.

15. The device of claim 14 , wherein the se

Assignees

• Okta Inc

Inventors

• Wang Zhao
• Huang Brian
• Akhundov Murad
• Du Zhi Rui
• Siblini Jad

Classifications

• G06F16/24522 Primary

  Translation of natural language queries to structured queries · CPC title