Providing query restatements for explaining natural language query results
Query restatements may be provided for explaining natural language query results. A natural language query is received at a natural language query processing system. An intermediate representation of the natural language query is generated for executing the natural language query. The intermediate representation is translated into a natural language restatement of the natural language query. The natural language restatement is provided with a result of the natural language query via an interface of the natural language query processing system.
1. A system, comprising:
at least one processor; and
a memory, storing program instructions that when executed by the at least one processor, cause the at least one processor to implement a natural language query processing system that provides access to a plurality of fixed schema data sets, the natural language query processing system configured to:
receive a natural language query via an interface for the natural language query processing system;
process the natural language query through a natural query language processing pipeline to generate an intermediate representation of the natural language query used for executing the natural language query;
determine from the intermediate representation both a natural language restatement of the intermediate representation and a format for executing the intermediate representation of the natural language query that invokes a determined type of visualization as part of generating a result for the natural language query, wherein to determine the natural language restatement of the intermediate representation, the natural language query processing system is configured to translate the intermediate representation into a natural language restatement of the natural language query;
cause the natural language query to be executed according to the determined format for the intermediate representation; and
cause the natural language restatement to be displayed with the result, including the invoked type of visualization, of the natural language query via the interface of the natural language query processing system.
2. The system of claim 1 , wherein the query processing system is further configured to:
receive, via the interface, a selection of a portion of the natural language restatement; and
return, via the interface, an annotation that describes the selected portion of the natural language restatement.
3. The system of claim 1 , wherein the query processing system is further configured to:
receive, via the interface, an update to the natural language restatement;
update the intermediate representation according to the update to the natural language restatement;
execute the updated intermediate representation to generate an updated result for the natural language query; and
provide, via the interface, the updated result for the natural language query.
4. The system of claim 1 , wherein the natural language query processing system is implemented as part of a business intelligence service offered by a provider network.
5. A method, comprising:
receiving a natural language query via an interface for a natural language query processing system that provides access to a plurality of fixed schema data sets;
generating an intermediate representation of the natural language query used for executing the natural language query;
determining, from the intermediate representation, both a natural language restatement of the intermediate representation and a format for executing the intermediate representation of the natural language query that invokes a determined type of visualization as part of generating a result for the natural language query, wherein determining the natural language restatement of the intermediate representation comprises translating the intermediate representation into a natural language restatement of the natural language query; and
providing the natural language restatement with a result, including the determined type of visualization, of the natural language query from one of the fixed schema data sets via the interface of the natural language query processing system.
6. The method of claim 5 , wherein translating the intermediate representation into a natural language restatement of the natural language query comprises:
identifying one restatement template from a set of restatement templates that matches the intermediate representation; and
populating one or more fields in the restatement template from the intermediate representation.
7. The method of claim 5 , further comprising:
receiving, via the interface, a selection of a portion of the natural language restatement; and
returning, via the interface, an annotation that describes the selected portion of the natural language restatement.
8. The method of claim 7 , wherein the annotation includes a cell value as a source of the selected portion.
9. The method of claim 7 , wherein the annotation is stored as part of metadata for the one fixed schema data set accessed to perform the natural language query.
10. The method of claim 5 , further comprising:
receiving, via the interface, an update to the natural language restatement;
updating the intermediate representation according to the update to the natural language restatement;
executing the updated intermediate representation to generate an updated result for the natural language query; and
providing, via the interface, the updated result for the natural language query.
11. The method of claim 10 , wherein the update to the natural language restatement modifies an annotation of a portion of the natural language restatement provided via the interface.
12. The method of claim 5 , further comprising:
receiving, via the interface, an update to the natural language restatement; and
storing the update to the natural language restatement as part of training data for updating a machine learning model to generate the intermediate representation.
13. The method of claim 5 , wherein the natural language query processing system is implemented as part of a database system.
14. One or more non-transitory, computer-readable storage media, storing program instructions that when executed on or across one or more computing devices cause the one or more computing devices to implement:
receiving a natural language query via an interface for a natural language query processing system;
generating an intermediate representation of the natural language query used for executing the natural language query;
determining, from the intermediate representation, both a natural language restatement of the intermediate representation and a format for executing the intermediate representation of the natural language query that invokes a determined type of visualization as part of generating a result for the natural language query, wherein, in determining the natural language restatement of the intermediate representation, the program instructions cause the one or more computing devices to implement translating the intermediate representation into a natural language restatement of the natural language query; and
causing the natural language restatement to be displayed with a result, including the determined type of visualization, of the natural language query via the interface of the natural language query processing system.
15. The one or more non-transitory, computer-readable storage media of claim 14 , wherein, in translating the intermediate representation into a natural language restatement of the natural language query, the program instructions cause the one or more computing devices to implement:
identifying one restatement template from a set of restatement templates that matches the intermediate representation; and
populating one or more fields in the restatement template from the intermediate representation.
16. The one or more non-transitory, computer-readable storage media of claim 14 , storing further instructions that when executed by the one or more computing devices, cause the one or more computing devices to further implement:
receiving, via the interface, a selection of a portion of the natural language restatement; and
returning, via the interface, an annotation that describes the selected portion of the natural language restatement.
17. The one or more non-transitory, computer-readable storage media of claim 16 , wherein the annotation includes a column name as a source of the selected portion.
18. The one or more non-transitory, computer-readable storage media of claim 16 , wherein the annotation includes an operation of the selected portion.
19. The one or more non-transitory, computer-readable storage media of claim 14 , storing further instructions that when executed by the one or more computing devices, cause the one or more computing devices to further implement:
receiving, via the interface, an update to the natural language restatement;
updating the intermediate representation according to the update to the natural language restatement;
executing the updated intermediate representation to generate an updated result for the natural language query; and
providing, via the interface, the updated result for the natural language query.
20. The one or more non-transitory, computer-readable storage media of claim 14 , wherein the natural language query processing system is implemented as part of a service offered by a provider network.