Generating predicted follow-on requests to a natural language request received by a natural language processing system

Patent metadata

Field	Value
Publication number	US-11670288-B1
Application number	US-202117328887-A
Country	US
Kind code	B1
Filing date	May 24, 2021
Priority date	Sep 28, 2018
Publication date	Jun 6, 2023
Grant date	Jun 6, 2023

Abstract

In various embodiments, a natural language (NL) application receives a partial NL request associated with a first context, and determining that the partial NL request corresponds to at least a portion of a first next NL request prediction included in one or more next NL request predictions generated based on a first natural language (NL) request, the first context associated with the first NL request, and a first sequence prediction model, where the first sequence prediction model is generated via a machine learning algorithm applied to a first data dependency model and a first request prediction model. In response to determining that the partial NL request corresponds to at least the portion of the first next NL request prediction, the NL application generates a complete NL request based on the first NL request and the partial NL request, and causes the complete NL request to be applied to a data storage system.

First claim

What is claimed is:

1. A computer-implemented method, comprising:
receiving a first natural language (NL) request associated with a first context;
generating a set of next NL request predictions based on the first NL request, the first context, a data dependency model, and a request prediction model, wherein the request prediction model is trained based on one or more mappings between a set of predefined intents and one or more domain specific language (DSL) templates;
generating a next NL request associated with the first NL request based on the set of NL request predictions and a DSL template; and
causing the next NL request to be applied to a data storage system.

2. The method of claim 1 , further comprising causing the set of next NL request predictions to be provided to a user.

3. The method of claim 1 , further comprising ranking the set of next NL request predictions based on the first sequence prediction model.

4. The method of claim 1 , wherein the data dependency model comprises a data dependency graph, the data dependency graph capturing dependencies between a first artifact associated with the first NL request and one or more additional artifacts within the data storage system.

5. The method of claim 1 , wherein the data dependency model and the request prediction model are cached by an application at a client device, and generating the set of next NL request predictions comprises generating the set of next NL request predictions in the application at the client device.

6. The method of claim 1 , wherein the first context includes at least one of a time associated with the first NL request, a device with which a user specified the first NL request, a job function of the user, a location of the user, a time of day, a user situation, a user context, or a user preference.

7. The method of claim 1 , wherein generating a next NL request comprises translating the next NL request into a DSL request and applying the DSL request based on the DSL template.

8. The method of claim 1 , further comprising caching the set of NL request predictions.

9. One or more non-transitory computer-readable storage media including instructions that, when executed by one or more processors, cause the one or more processors to perform the steps of:
receiving a first natural language (NL) request associated with a first context;
generating a set of next NL request predictions based on the first NL request, the first context, a data dependency model, and a request prediction model, wherein the request prediction model is trained based on one or more mappings between a set of predefined intents and one or more domain specific language (DSL) templates;
generating a next NL request associated with the first NL request based on the set of NL request predictions and a DSL template; and
causing the next NL request to be applied to a data storage system.

10. The one or more non-transitory computer-readable storage media of claim 9 , further comprising causing the set of next NL request predictions to be provided to a user.

11. The one or more non-transitory computer-readable storage media of claim 9 , further comprising ranking the set of next NL request predictions based on the first sequence prediction model.

12. The one or more non-transitory computer-readable storage media of claim 9 , wherein the data dependency model comprises a data dependency graph, the data dependency graph capturing dependencies between a first artifact associated with the first NL request and one or more additional artifacts within the data storage system.

13. The one or more non-transitory computer-readable storage media of claim 9 , wherein the data dependency model and the request prediction model are cached by an application at a client device, and generating the set of next NL request predictions comprises generating the set of next NL request predictions in the application at the client device.

14. The one or more non-transitory computer-readable storage media of claim 9 , wherein the first context includes at least one of a time associated with the first NL request, a device with which a user specified the first NL request, a job function of the user, a location of the user, a time of day, a user situation, a user context, or a user preference.

15. The one or more non-transitory computer-readable storage media of claim 9 , wherein generating a next NL request comprises translating the next NL request into a DSL request and applying the DSL request based on the DSL template.

16. The one or more non-transitory computer-readable storage media of claim 9 , further comprising caching the set of NL request predictions.

17. A computing device, comprising:
a memory that includes instructions; and
a processor that is coupled to the memory and, when executing the instructions, is configured to:
receive a first natural language (NL) request associated with a first context;
generate a set of next NL request predictions based on the first NL request, the first context, a data dependency model, and a request prediction model, wherein the request prediction model is trained based on one or more mappings between a set of predefined intents and one or more domain specific language (DSL) templates;
generate a next NL request associated with the first NL request based on the set of NL request predictions and a DSL template; and
cause the next NL request to be applied to a data storage system.

18. The computing device of claim 17 , further comprising causing the set of next NL request predictions to be provided to a user.

19. The computing device of claim 17 , further comprising ranking the set of next NL request predictions based on the first sequence prediction model.

20. The computing device of claim 17 , wherein the data dependency model comprises a data dependency graph, the data dependency graph capturing dependencies between a first artifact associated with the first NL request and one or more additional artifacts within the data storage system.

Assignees

• Splunk Inc

Inventors

• Das Dipock
• Pochugari Dayanand
• Radon Aungon Nag

Classifications

• G06F16/243

  Natural language query formulation · CPC title

• G10L2015/225

  Feedback of the input speech · CPC title

• G06F16/9032

  Query formulation · CPC title

• G06N3/08

  Learning methods · CPC title

• G06F16/90332 Primary

  Natural language query formulation or dialogue systems · CPC title