Back to results
US 12,585,645 B1Granted
Natural language query processing
AssigneeAmazon Technologies, Inc.
InventorsHanbo Li, Patrick Ng, Zhiguo Wang, Rishav Chakravarti, Stephen Michael Ash, Bing Xiang, Gregory David Adams
Life of this patent
Estimating the term…
Abstract
Techniques for handling natural language query processing are described. In some examples, semantic meanings of words are determined during the natural language query processing. These semantic meanings are generated from metadata associated with the query and are to be used by an entity linker to help the linker link candidates to columns.
Claims20
A computer-implemented method comprising:
- receiving a natural language query (NLQ);
- performing a lexical query to retrieve metadata for the NLQ from an index of cell values and column names generated from topic metadata and structured datasets;
- performing named entity recognition (NER) on the NLQ to determine one or more entities of the NLQ;
- performing a lexical query to retrieve a first set of linkable candidates associated with the determined one or more entities;
- performing a semantic search to generate a second set of linkable candidates of a database schema, wherein the second set of linkable candidates is associated with the determined one or more entities;
- performing named entity linking (NEL) using the first and second sets of linkable candidates, the NEL to identify candidate linkages between at least a subset of the determined one or more entities and associated linkable candidates from at least one of the first and second sets of linkable candidates and rank the identified candidate linkages;
- ranking datasets identified by the ranked candidate linkages for the determined one or more entities of the NLQ;
- selecting one of multiple intent representations generated for the NLQ according to the ranked candidate linkages and the ranked datasets to be an intent representation for the natural language query according to a score for the intent representation;
- causing the intent representation to be used to execute the natural language query;
- and
- returning, via an interface, a result for the execution of the NLQ.
8 more elements
A computer-implemented method comprising:
- receiving a natural language query (NLQ);
- performing a lexical query to retrieve metadata for the NLQ;
- performing named entity recognition (NER) on the NLQ to determine one or more entities of the NLQ;
- performing a lexical query to retrieve a first set of linkable candidates associated with the determined one or more entities;
- performing a semantic search to generate a second set of linkable candidates of a database schema, wherein the second set of linkable candidates is associated with the determined one or more entities;
- performing named entity linking (NEL) using the first and second sets of linkable candidates, the NEL to identify candidate linkages between at least a subset of the determined one or more entities and associated linkable candidates from at least one of the first and second sets of linkable candidates and rank the identified candidate linkages;
- ranking datasets identified by the ranked candidate linkages for the determined one or more entities of the NLQ;
- selecting one of multiple intent representations generated for the NLQ according to the ranked candidate linkages and the ranked datasets to be an intent representation for the natural language query according to a score for the intent representation;
- causing the intent representation to be used to execute the natural language query;
- and
- returning, via an interface, a result for the execution of the NLQ.
8 more elements
A system comprising:
- a first one or more electronic devices to implement a database storage service in a multi-tenant provider network;
- and
- a second one or more electronic devices to implement a natural language query (NLQ) processing service in the multi-tenant provider network, the NLQ processing service including instructions that upon execution cause the NLQ processing service to:
- receive an NLQ;
- performing a lexical query to retrieve metadata from the database storage service for the NLQ;
- performing named entity recognition (NER) on the NLQ to determine one or more entities of the NLQ;
- performing a lexical query to retrieve a first set of linkable candidates associated with the determined one or more entities;
- performing a semantic search to generate a second set of linkable candidates of a database schema, wherein the second set of linkable candidates is associated with the determined one or more entities;
- performing named entity linking (NEL) using the first and second sets of linkable candidates, the NEL to identify candidate linkages between at least a subset of the determined one or more entities and associated linkable candidates from at least one of the first and second sets of linkable candidates and rank the identified candidate linkages;
- ranking datasets identified by the ranked candidate linkages for the determined one or more entities of the NLQ;
- selecting one of multiple intent representations generated for the NLQ according to the ranked candidate linkages and the ranked datasets to be an intent representation for the natural language query according to a score for the intent representation;
- causing the intent representation to be used to execute the natural language query;
- and
- returning, via an interface, a result for the execution of the NLQ.
11 more elements
Dependencies are inferred from the claim text.