Systems and methods for natural language question answering and analysis

• US 10,409,846 B2
• Filed: 02/16/2016
• Issued: 09/10/2019
• Est. Priority Date: 02/13/2015
• Status: Active Grant

• Current Assignee: Thomson Reuters Enterprise Centre GmbH (The Woodbridge Co. Ltd.)
• Original Assignee: Thomson Reuters Global Resources Unlimited Company (The Woodbridge Co. Ltd.)
• Inventors: Schilder, Frank, Smiley, Charese, Song, Dezhao, Brew, Christopher Hardie
• Primary Examiner(s): Al-Hashemi, Sana A
• Application Number:
  US15/044,188
• Publication Number:
  US 20160239562A1
• Time in Patent Office: 1,302 Days
• Field of Search: 707609, 707687, 707709, 707750, 707813, 707821
• CPC Class Codes: G06F 16/3322 using system suggestions G06F16/3325 takes precedence
• CPC Class Codes: G06F 16/3329 Natural language query formulation or dialogue systems
• CPC Class Codes: G06F 16/3344 using natural language analysis
• CPC Class Codes: G06F 16/338 Presentation of query results
• CPC Class Codes: G06F 16/90332 Natural language query formulation or dialogue systems
• CPC Class Codes: G06F 40/205 Parsing
• CPC Class Codes: G06F 40/284 Lexical analysis, e.g. tokenisation or collocates

Abstract

The present invention is directed towards systems and methods for querying a data repository using a flexible natural language interface, which comprises receiving an initiated user question at a graphical user interface and generating automatically one or more suggested completed questions in response to the receipt of the initiated user question. A selected completed question at the graphical user interface is received and subsequently parsed into a logic representation. The logic representation is translated into an executable query, which is executed against the data repository. One or more search results in response to the executed translated query and presented at the graphical user interface. The present invention further comprises generating one or more analytic results corresponding to the one or more search responsive to the translated query.

Claims

1. A computer-implemented method for querying a data repository using a flexible natural language interface comprising:

receiving an initiated user question at a graphical user interface;

generating automatically one or more suggested completed questions in response to the receipt of the initiated user question, said generation based on a defined grammar and linguistic constraints associated with the defined grammar, wherein said generating comprises incrementally parsing a grammar tree, said grammar tree comprising grammar rules relating to the defined grammar on branch nodes and lexical rules relating to the linguistic constraints on leaf nodes;

receiving a selected completed question at the graphical user interface;

parsing the selected completed question into a logic representation;

translating the logic representation into an executable query;

executing the translated query against the data repository;

generating one or more search results in response to the executed translated query;

generating one or more analytic results corresponding to the one or more search results responsive to the executed translated query, wherein said one or more analytic results comprises one or more of a descriptive analytic result, a comparative analytic result, a temporal analytic result and a content-based analytic result; and

presenting the one or more search results and the one or more analytic results corresponding to one or more search results at the graphical user interface.

4. The computer-implemented method of claim 1 wherein generating automatically one or more suggested completed questions in response to the receipt of the initiated user question further comprises:

upon incrementally parsing the grammar tree, determining whether the grammar rules and linguistic constraints are satisfied;

generating automatically one or more suggested completed questions upon determination that the grammar rules and linguistic constraints are satisfied.

5. The computer-implemented method of claim 1 wherein generating automatically one or more suggested completed questions in response to the receipt of the initiated user question further comprises ranking the one or more suggested completed questions based on a relational quantity.

6. The computer-implemented method of claim 1 wherein parsing the selected question into a logic representation further comprises:

parsing the logic representation to a parse tree; and

traversing the parse tree in order to identify one or more query constraints for the translated query language, said query constraints utilized to generate a translated query.

2. Non-transitory computer readable media comprising program code stored thereon for execution by a programmable processor to perform a method for querying a data repository using a flexible natural language interface, the computer readable media comprising:

program code for receiving an initiated user question at a graphical user interface;

program code for generating automatically one or more suggested completed questions in response to the receipt of the initiated user question, said generation based on a defined grammar and linguistic constraints associated with the defined grammar, wherein said program code for generating comprises program code for incrementally parsing a grammar tree, said grammar tree comprising grammar rules relating to the defined grammar on branch nodes and lexical rules relating to the linguistic constraints on leaf nodes;

program code for receiving a selected completed question at the graphical user interface;

program code for parsing the selected completed question into a logic representation;

program code for translating the logic representation into an executable query;

program code for executing the translated query against the data repository;

program code for generating one or more search results in response to the executed translated query;

program code for generating one or more analytic results corresponding to the one or more search results responsive to the executed translated query, wherein said one or more analytic results comprises one or more of a descriptive analytic result, a comparative analytic result, a temporal analytic result and a content-based analytic result; and

program code for presenting the one or more search results and the one or more analytic result corresponding to one or more search results at the graphical user interface.

7. The computer readable media of claim 2 wherein the program code for generating automatically one or more suggested completed questions in response to the receipt of the initiated user question further comprises:

program code for determining whether the grammar rules and linguistic constraints are satisfied upon incrementally parsing the grammar tree; and

program code for generating automatically one or more suggested completed questions upon determination that the grammar rules and linguistic constraints are satisfied.

8. The computer readable media of claim 2 wherein program code for generating automatically one or more suggested completed questions in response to the receipt of the initiated user question further comprises program code for ranking the one or more suggested completed queries based on a relational quantity.

9. The computer readable media of claim 2 wherein program code for parsing the selected question into a logic representation further comprises:

program code for parsing the logic representation to a parse tree; and

program code for traversing the parse tree in order to identify one or more query constraints for the translated query language, said query constraints utilized to generate a translated query.

3. A system for querying a data repository using a flexible natural language interface comprising:

a server including a processor configured to;

receive an initiated user question at a graphical user interface;

generate automatically one or more suggested completed questions in response to the receipt of the initiated user question, said generation based on a defined grammar and linguistic constraints associated with the defined grammar, wherein said generating comprises incrementally parsing a grammar tree, said grammar tree comprising grammar rules relating to the defined grammar on branch nodes and lexical rules relating to the linguistic constraints on leaf nodes;

receive a selected completed question at the graphical user interface;

parse the selected completed question into a logic representation;

translate the logic representation into an executable query;

execute the translated query against the data repository;

generate one or more search results in response to the executed translated query;

generate one or more analytic results corresponding to the one or more search results responsive to the executed translated query, wherein said one or more analytic results comprises one or more of a descriptive analytic result, a comparative analytic result, a temporal analytic result and a content-based analytic result; and

present the one or more search results and the one or more analytic results corresponding to one or more search results at the graphical user interface.

10. The system of claim 3 wherein in generating automatically one or more suggested completed questions in response to the receipt of the initiated user question, the server is further configured to:

upon incrementally parsing the grammar tree, determine whether the grammar rules and linguistic constraints are satisfied; and

generate automatically one or more suggested completed questions upon determination that the grammar rules and linguistic constraints are satisfied.

11. The system of claim 3 wherein in generating automatically one or more suggested completed questions in response to the receipt of the initiated user question, the server is further configured to rank the one or more suggested completed questions based on a relational quantity.

12. The system of claim 3 wherein in parsing the selected query into a logic representation, the server is further configured to:

parse the logic representation to a parse tree; and

traverse the parse tree in order to identify one or more query constraints for the translated query language, said query constraints utilized to generate a translated query.