Methods and systems for managing chatbots with respect to rare entities

Patent metadata

Field	Value
Publication number	US-11397857-B2
Application number	US-202016743810-A
Country	US
Kind code	B2
Filing date	Jan 15, 2020
Priority date	Jan 15, 2020
Publication date	Jul 26, 2022
Grant date	Jul 26, 2022

Abstract

Embodiments for managing chatbots are provided. A set of documents is received. A plurality of entities are identified within the set of documents. At least one of the plurality of entities is selected based on a rareness criteria. Contextual data associated with each of the selected at least one of the plurality of entities is identified within the set of documents. At least one question-answer (QA) pair associated with each of the selected at least one of the plurality of entities is generated based on the identified contextual data.

First claim

The invention claimed is:

1. A method for managing a chatbot, by a processor, comprising:
receiving a set of documents;
identifying a plurality of entities within the set of documents, wherein the identifying of the plurality of entities includes determining a number of references to each of the plurality of entities within the set of documents;
selecting at least one of the plurality of entities based on a rareness criteria, wherein the selecting of the at least one of the plurality of entities based on a rareness criteria includes:
associating a first of the plurality of entities with a second of the plurality of entities based on a fuzzy matching algorithm, wherein the first of the plurality of entities has a first number of references in the set of documents, and the second of the plurality of entities has a second number of references in the set of documents;
adding the second number of references to the first number of references to calculate a composite number of references for the first of the plurality of entities; and
utilizing the composite number of references to determine if the first of the plurality of entities meets the rareness criteria;

identifying contextual data associated with each of the selected at least one of the plurality of entities within the set of documents; and
generating at least one question-answer (QA) pair associated with each of the selected at least one of the plurality of entities based on the identified contextual data.

2. The method of claim 1 , wherein the selecting of the at least one of the plurality of entities based on the rareness criteria includes selecting those of the plurality of entities for which the number of references within the set of documents is less than a first predetermined threshold and greater than a second predetermined threshold.

3. The method of claim 1 , further comprising providing a user interface that allows a user to modify the selecting of the at least one of the plurality of entities, and wherein the identifying of the contextual data associated with the selected at least one of the plurality of entities within the set of documents includes identifying contextual data associated with said modified at least one of the plurality of entities, and the generating of the at least one QA pair includes generating a QA pair associated with each of said modified at least one of the plurality of entities.

4. The method of claim 1 , further comprising causing a chatbot system to be trained utilizing the generated at least one QA pair.

5. The method of claim 1 , wherein each of the plurality of entities includes at least one of an individual, an object, and a location.

6. A system for managing a chatbot comprising:
a processor executing instructions stored in a memory device, wherein the processor:
receives a set of documents;
identifies a plurality of entities within the set of documents, wherein the identifying of the plurality of entities includes determining a number of references to each of the plurality of entities within the set of documents;
selects at least one of the plurality of entities based on a rareness criteria, wherein the selecting of the at least one of the plurality of entities based on a rareness criteria includes:
associating a first of the plurality of entities with a second of the plurality of entities based on a fuzzy matching algorithm, wherein the first of the plurality of entities has a first number of references in the set of documents, and the second of the plurality of entities has a second number of references in the set of documents;
adding the second number of references to the first number of references to calculate a composite number of references for the first of the plurality of entities; and
utilizing the composite number of references to determine if the first of the plurality of entities meets the rareness criteria;

identifies contextual data associated with each of the selected at least one of the plurality of entities within the set of documents; and
generates at least one question-answer (QA) pair associated with each of the selected at least one of the plurality of entities based on the identified contextual data.

7. The system of claim 6 , wherein the selecting of the at least one of the plurality of entities based on the rareness criteria includes selecting those of the plurality of entities for which the number of references within the set of documents is less than a first predetermined threshold and greater than a second predetermined threshold.

8. The system of claim 6 , wherein the processor further provides a user interface that allows a user to modify the selecting of the at least one of the plurality of entities, and wherein the identifying of the contextual data associated with the selected at least one of the plurality of entities within the set of documents includes identifying contextual data associated with said modified at least one of the plurality of entities, and the generating of the at least one QA pair includes generating a QA pair associated with each of said modified at least one of the plurality of entities.

9. The system of claim 6 , wherein the processor further causes a chatbot system to be trained utilizing the generated at least one QA pair.

10. The system of claim 6 , wherein each of the plurality of entities includes at least one of an individual, an object, and a location.

11. A computer program product for managing a chatbot, by a processor, the computer program product embodied on a non-transitory computer-readable storage medium having computer-readable program code portions stored therein, the computer-readable program code portions comprising:
an executable portion that receives a set of documents;
an executable portion that identifies a plurality of entities within the set of documents, wherein the identifying of the plurality of entities includes determining a number of references to each of the plurality of entities within the set of documents;
an executable portion that selects at least one of the plurality of entities based on a rareness criteria, wherein the selecting of the at least one of the plurality of entities based on a rareness criteria includes:
associating a first of the plurality of entities with a second of the plurality of entities based on a fuzzy matching algorithm, wherein the first of the plurality of entities has a first number of references in the set of documents, and the second of the plurality of entities has a second number of references in the set of documents;
adding the second number of references to the first number of references to calculate a composite number of references for the first of the plurality of entities; and
utilizing the composite number of references to determine if the first of the plurality of entities meets the rareness criteria;

an executable portion that identifies contextual data associated with each of the selected at least one of the plurality of entities within the set of documents; and
an executable portion that generates at least one question-answer (QA) pair associated with each of the selected at least one of the plurality of entities based on the identified contextual data.

12. The computer program product of claim 11 , wherein the selecting of the at least one of the plurality of entities based on the rareness criteria includes selecting those of the plurality of entities for which the number of references within the set of documents is less than a first predetermined threshold and greater than a second predetermined threshold.

13. The computer program product of claim 11 , wherein the computer-readable program code portions furthe

Assignees

• IBM

Inventors

• Drzewucki Michael
• Decker Renee F
• Dixius James M
• Sagal Ii David Kenneth

Classifications

• G06F40/216

  using statistical methods · CPC title

• G06F40/35

  Discourse or dialogue representation · CPC title

• G06F40/284

  Lexical analysis, e.g. tokenisation or collocates · CPC title

• G06F40/56

  Natural language generation · CPC title

• G06F40/295 Primary

  Named entity recognition · CPC title