AUTOMATIC GENERATION OF STRUCTURED QUERIES FROM NATURAL LANGUAGE INPUT

• US 20180165330A1
• Filed: 12/08/2016
• Published: 06/14/2018
• Est. Priority Date: 12/08/2016
• Status: Active Grant

• Current Assignee: SAP SE
• Original Assignee: SAP SE
• Inventors: Halbani, Ronen, Tarshish, Shahar
• Granted Patent: US 10,657,124 B2
• Time in Patent Office: Days
• CPC Class Codes: G06F 16/24522 Translation of natural language queries to structured queries

Abstract

The present disclosure relates to computer systems and computer-implemented methods for automatically enabling natural language rules in a structured rule engine. In one example, the method includes receiving an input identifying an unstructured query corresponding to a particular application associated with a rules framework associated with a pre-defined query structure format, analyzing the received input based on a first set of design time artifacts associated with the particular application and a second set of design time artifacts associated with the rules framework, the first set of artifacts uniquely associated with the particular application and the second set of artifacts generically associated with the rules framework, wherein the second set of design time artifacts are applied commonly to a plurality of different applications. A structured query for the particular application is generated in the pre-defined query structure format of the rules framework based on the first and second set of artifacts.

Claims

1. A computer-implemented method performed by at least one processor, the method comprising:

receiving an input identifying an unstructured query corresponding to a particular application associated with a rules framework associated with a pre-defined query structure format;

analyzing the received input based on a first set of design time artifacts associated with the particular application and a second set of design time artifacts associated with the rules framework, the first set of design time artifacts uniquely associated with the particular application and the second set of design time artifacts generically associated with the rules framework, wherein the second set of design time artifacts are applied commonly to a plurality of different applications; and

generating a structured query for the particular application in the pre-defined query structure format of the rules framework based on the first and second set of design time artifacts.

2. The method of claim 1, wherein the first set of design time artifacts are generated prior to receiving the input, and comprise a set of data objects, attributes, data object relations, and terms specifically associated with the particular application.

3. The method of claim 1, wherein the second set of design time artifacts are generated prior to receiving the input, and comprise artifacts associated with a syntax and grammar associated with the pre-defined query structure format of the rules framework.

4. The method of claim 3, wherein the artifacts include tokens and patterns satisfying the pre-defined query structure format of the rules framework.

5. The method of claim 1, wherein analyzing the received input comprises:

performing linguistic and semantic processing on the received input to create an output of key tokens from the received input, the key tokens corresponding to a set of extracted entities and relationships between the extracted entities identified based on first and second set of design time artifacts; and

identifying at least one term from a term container included in the first set of design time artifacts matching a combination of one or more of the key tokens, where the at least one identified terms are used for generating the structured query for the particular application.

6. The method of claim 5, wherein generating the structured query for the particular application comprises:

identifying at least one expression created by a combination of the at least one identified terms where the at least one expression corresponds to a pattern defined within the second set of design time artifacts.

7. The method of claim 6, wherein the pattern is defined as a combination of entity types, and wherein each of the at least one identified terms are associated with an entity type.

8. The method of claim 1, wherein the received input identifying the unstructured query comprises a natural language query associated with the particular application.

9. A non-transitory, computer-readable medium storing one or more instructions executable by a computer system to perform operations comprising:

receiving an input identifying an unstructured query corresponding to a particular application associated with a rules framework associated with a pre-defined query structure format;

analyzing the received input based on a first set of design time artifacts associated with the particular application and a second set of design time artifacts associated with the rules framework, the first set of design time artifacts uniquely associated with the particular application and the second set of design time artifacts generically associated with the rules framework, wherein the second set of design time artifacts are applied commonly to a plurality of different applications; and

generating a structured query for the particular application in the pre-defined query structure format of the rules framework based on the first and second set of design time artifacts.

10. The computer-readable medium of claim 9, wherein the first set of design time artifacts are generated prior to receiving the input, and comprise a set of data objects, attributes, data object relations, and terms specifically associated with the particular application.

11. The computer-readable medium of claim 9, wherein the second set of design time artifacts are generated prior to receiving the input, and comprise artifacts associated with a syntax and grammar associated with the pre-defined query structure format of the rules framework.

12. The computer-readable medium of claim 11, wherein the artifacts include tokens and patterns satisfying the pre-defined query structure format of the rules framework.

13. The computer-readable medium of claim 9, wherein analyzing the received input comprises:

performing linguistic and semantic processing on the received input to create an output of key tokens from the received input, the key tokens corresponding to a set of extracted entities and relationships between the extracted entities identified based on first and second set of design time artifacts; and

identifying at least one term from a term container included in the first set of design time artifacts matching a combination of one or more of the key tokens, where the at least one identified terms are used for generating the structured query for the particular application.

14. The computer-readable medium of claim 13, wherein generating the structured query for the particular application comprises:

identifying at least one expression created by a combination of the at least one identified terms where the at least one expression corresponds to a pattern defined within the second set of design time artifacts.

15. The computer-readable medium of claim 14, wherein the pattern is defined as a combination of entity types, and wherein each of the at least one identified terms are associated with an entity type.

16. The computer-readable medium of claim 9, wherein the received input identifying the unstructured query comprises a natural language query associated with the particular application.

17. A computer system, comprising:

a computer memory; and

a hardware processor interoperably coupled with the computer memory and configured to perform operations comprising;

receiving an input identifying an unstructured query corresponding to a particular application associated with a rules framework associated with a pre-defined query structure format, the received input comprising a natural language query associated with the particular application;

analyzing the received input based on a first set of design time artifacts associated with the particular application and a second set of design time artifacts associated with the rules framework, the first set of design time artifacts uniquely associated with the particular application and the second set of design time artifacts generically associated with the rules framework, wherein the second set of design time artifacts are applied commonly to a plurality of different applications; and

generating a structured query for the particular application in the pre-defined query structure format of the rules framework based on the first and second set of design time artifacts.

18. The system of claim 17, wherein the first set of design time artifacts are generated prior to receiving the input, and comprise a set of data objects, attributes, data object relations, and terms specifically associated with the particular application.

19. The system of claim 17, wherein the second set of design time artifacts are generated prior to receiving the input, and comprise artifacts associated with a syntax and grammar associated with the pre-defined query structure format of the rules framework, further wherein the artifacts include tokens and patterns satisfying the pre-defined query structure format of the rules framework.

20. The system of claim 17, wherein analyzing the received input comprises:

performing linguistic and semantic processing on the received input to create an output of key tokens from the received input, the key tokens corresponding to a set of extracted entities and relationships between the extracted entities identified based on first and second set of design time artifacts; and

identifying at least one term from a term container included in the first set of design time artifacts matching a combination of one or more of the key tokens, where the at least one identified terms are used for generating the structured query for the particular application; and

wherein generating the structured query for the particular application comprises identifying at least one expression created by a combination of the at least one identified terms where the at least one expression corresponds to a pattern defined within the second set of design time artifacts, wherein the pattern is defined as a combination of entity types, and wherein each of the at least one identified terms are associated with an entity type.