US 12,743,454 B1Granted

Fuzzy matching natural language query from user specific metadata

AssigneeAmazon Technologies, Inc.

InventorsAmjad Al-Rikabi, Stephen Michael Ash, Anuj Chauhan, Rishav Chakravarti, Wuwei Lan, Jerry Lin, Gregory David Adams

Life of this patent

Estimating the term…

patentgrep

Loading patent…

Abstract

Systems and methods are provided for an improved business intelligence platform that maintains and uses personalized data cache for processing natural language queries with personalized fuzzy matching in lieu of using predefined indexes of large amounts of data. The systems and methods provide for performing fuzzy matching using a natural language query and a personalized data cache storing a copy of data previously provided in response to interactions associated with a user profile, wherein the personalized data cache is separate from a data source storing the response data, and wherein the fuzzy matching does not use any fuzzy index of the data source generated prior to receiving the natural language query; obtaining, from results of the fuzzy matching, a set of query terms; obtaining the response data from the data source using the set of query terms; and generating a response to the natural language query using the response data.

Claims20

Dependent claims

A system comprising:

computer-readable memory storing computer-executable instructions;
and
one or more processors in communication with the computer-readable memory, wherein the computer-executable instructions, when executed by the one or more processors, configure the one or more processors to:
provide, to a client device associated with a user profile, a data visualization based on a first data set stored at a data source;
store, in a personalized data cache associated with the user profile, a copy of data in the data visualization, wherein the personalized data cache is separate from the data source;
receive a natural language query from the client device, wherein the natural language query comprises a request for response data associated with a least one item of data in the data visualization;
perform in-memory fuzzy matching without using a full fuzzy index of the data source, using the natural language query and items of data retrieved from the personalized data cache;
obtain, from results of the fuzzy matching, a set of query terms for querying the data source;
generate an executable query using the set of query terms and a large language model;
execute the executable query to obtain the response data from the data source;
and
generate a response to the natural language query using the response data.

9 more elements

One or more non-transitory computer-readable media comprising computer-executable instructions that, when executed by a computing system, cause the computing system to:

receive a natural language query associated with a user profile, wherein the natural language query comprises a request for response data;
perform fuzzy matching using the natural language query and a personalized data cache storing a copy of data previously provided in response to interactions associated with the user profile, wherein the personalized data cache is separate from a data source storing the response data, and wherein the fuzzy matching does not use a full fuzzy index of the data source;
obtain, from results of the fuzzy matching, a set of query terms for querying the data source, wherein the data source stores the response data and the data previously provided in response to interactions associated with the user profile;
obtain the response data from the data source using the set of query terms;
and
generate a response to the natural language query using the response data.

3 more elements

Dependencies are inferred from the claim text.