Method, device, and non-transitory computer-readable recording medium for creating chatbot scenario considering response rate

Patent metadata

Field	Value
Publication number	US-12250182-B2
Application number	US-202318334916-A
Country	US
Kind code	B2
Filing date	Jun 14, 2023
Priority date	Jun 20, 2022
Publication date	Mar 11, 2025
Grant date	Mar 11, 2025

Abstract

A method, a device, and a non-transitory computer-readable recording medium are provided for creating a chatbot scenario in consideration of a response rate. A chatbot scenario creation method may include obtaining chatbot sentence components and at least one chatbot sentence including the chatbot sentence components from a chatbot; predicting a response rate of a user for at least one chatbot sentence based on a predicted response rate preset for each of chatbot sentence components; and providing the predicted response rate of the user for the at least one chatbot sentence.

First claim

What is claimed is:

1. A method of creating a chatbot scenario performed by a computer device comprising at least one processor, the method comprising:
obtaining chatbot sentence components and at least one chatbot sentence including the chatbot sentence components from a chatbot;
outputting the at least one chatbot sentence to a user;
predicting a response rate of the user for the at least one chatbot sentence based on a predicted response rate that is predetermined for each of the chatbot sentence components, the predicted response rate based on probabilities that a response event occurs from the user for each of the chatbot sentence components;
providing the predicted response rate for the at least one chatbot sentence; and
updating the predicted response rate for each of the chatbot sentence components that constitute the at least one chatbot sentence based on a result of the response event occurring from the user as the at least one chatbot sentence is outputted to the user.

2. The method of claim 1 , wherein the predicted response rate for each of the chatbot sentence components is calculated in advance based on a probability that the response event occurred from users for each of the chatbot sentence components in a past scenario provided from the chatbot or a probability that the response event occurred from the users for each of the chatbot sentence components included in advertising texts.

3. The method of claim 1 , wherein the providing further comprises recommending at least one recommended chatbot sentence or at least one recommended chatbot sentence component based on the predicted response rate for the at least one chatbot sentence or the predicted response rate for each of the chatbot sentence components.

4. The method of claim 1 , wherein the predicting comprises predicting the predicted response rate of the user for the at least one chatbot sentence by applying a Naïve bayes method based on a predicted response rate preset for each of the chatbot sentence components.

5. The method of claim 1 , wherein the predicted response rate is represented by a probability that each of response sentence components is to be input as the response event from the user.

6. The method of claim 5 , wherein the predicted response rate for each of the chatbot sentence components is calculated in advance based on a probability that each of the response sentence components is input as the response event from users for each of the chatbot sentence components in a past scenario provided from the chatbot or a probability that each of the response sentence components is input as the response event from the users for each of the chatbot sentence components included in advertising texts.

7. The method of claim 6 , wherein the providing further comprises updating the predicted response rate for each of chatbot sentence components that constitute the at least one chatbot sentence based on a result of inputting a specific response sentence component as the response event from the user as the at least one chatbot sentence is provided to the user.

8. The method of claim 5 , wherein the predicting comprises predicting the predicted response rate of each of the response sentence components capable of being input as the response event from the user in response to the at least one chatbot sentence, and
the providing comprises providing the predicted response rate of each of the response sentence components that respond to the at least one chatbot sentence.

9. The method of claim 8 , wherein the providing further comprises providing at least one specific sentence component having the predicted response rate greater than or equal to a preset value among the response sentence components or a combination thereof.

10. The method of claim 8 , wherein the providing further comprises recommending at least one recommended chatbot sentence or at least one recommended chatbot sentence component based on the predicted response rate of each of the response sentence components for the at least one chatbot sentence or the predicted response rate of each of the response sentence components for each of the chatbot sentence components.

11. A non-transitory computer-readable recording medium storing instructions that, when executed by a processor, cause the processor to execute a method of creating a chatbot scenario, the method comprising:
obtaining chatbot sentence components and at least one chatbot sentence including the chatbot sentence components from a chatbot;
outputting the at least one chatbot sentence to a user;
predicting a response rate of the user for the at least one chatbot sentence based on a predicted response rate that is predetermined for each of the chatbot sentence components, the predicted response rate based on probabilities that a response event occurs from the user for each of the chatbot sentence components;
providing the predicted response rate of the user for the at least one chatbot sentence; and
updating the predicted response rate for each of the chatbot sentence components that constitute the at least one chatbot sentence based on a result of the response event occurring from the user as the at least one chatbot sentence is outputted to the user.

12. A computer device for performing a method of creating a chatbot scenario, the computer device comprising:
at least one processor configured to execute computer-readable instructions to:
obtain chatbot sentence components and at least one chatbot sentence including the chatbot sentence components from a chatbot;
output the at least one chatbot sentence to a user;
predict a response rate of the user for the at least one chatbot sentence based on a predicted response rate that is predetermined for each of the chatbot sentence components, the predicted response rate based on probabilities that a response event occurs from the user for each of the chatbot sentence components;
provide the predicted response rate for the at least one chatbot sentence; and
update the predicted response rate for each of the chatbot sentence components that constitute the at least one chatbot sentence based on a result of the response event occurring from the user as the at least one chatbot sentence is outputted to the user.

13. The computer device of claim 12 , wherein the predicted response rate for each of the chatbot sentence components is calculated in advance based on a probability that the response event occurred from users for each of the chatbot sentence components in a past scenario provided from the chatbot or a probability that the response event occurred from the users for each of the chatbot sentence components included in advertising texts.

14. The computer device of claim 12 , wherein the at least one processor is configured to recommend at least one recommended chatbot sentence or at least one recommended chatbot sentence component based on the predicted response rate for the at least one chatbot sentence or the predicted response rate for each of the chatbot sentence components.

15. The computer device of claim 12 , wherein the predicted response rate is represented by a probability that each of response sentence components is to be input as the response event from the user.

16. The computer device of claim 15 , wherein the predicted response rate for each of the chatbot sentence components is calculated in advance based on a probability that each of the response sentence components is input as the response event from users for each of the chatbot sentence components in a past scenario provided from the chatbot or a probability that each of the response sente

Assignees

• Line Plus Corp

Inventors

• Lee Sunho

Classifications

• G06Q50/10

  Services · CPC title

• G06Q10/04

  Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem" (market predictions or forecasting for commercial activities G06Q30/0202) · CPC title

• G06N20/00

  Machine learning · CPC title

• G06F17/18

  for evaluating statistical data {, e.g. average values, frequency distributions, probability functions, regression analysis (forecasting specially adapted for a specific administrative, business or logistic context G06Q10/04)} · CPC title

• G06F16/3329

  Natural language query formulation · CPC title