IP Library Granted Patent US 11,449,785
Granted Patent B2
US 11,449,785 · App. 15/434,558 · Granted Sep 20, 2022

Method and system for artificial intelligence learning using messaging service and method and system for relaying answer using artificial intelligence

Inventors: Seok Ho Kang (Seongnam-si, KR); Jae Gwang Lee (Seongnam-si, KR); Jea Seung Jung (Seongnam-si, KR); Hee-Cheol Seo (Seongnam-si, KR); JoongJae Lee (Seongnam-si, KR); Jeehyun Lee (Seongnam-si, KR); Young-sik Lim (Seongnam-si, KR); Injae Lee (Seongnam-si, KR)
Assignee: LINE Corporation
G06N20/00G06N3/006H04L51/046H04L51/216H04L67/306
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 11,449,785
App. No.
15/434,558
Granted
Sep 20, 2022
Kind
B2
Abstract

Provided are an artificial intelligence (AI) learning method and system and an answer relay method and system using an AI. According to an AI learning method, an AI may transmit a question to users through a messaging service and may acquire learning data for the AI through reactions of the user to the transmitted question.

Claims (84)

1. An artificial intelligence (AI) learning method of an AI learning system implemented on a computer, the method comprising:

selecting, by a processor of a server that includes or is connected to the AI learning system, question data from a database associated with the server;

verifying, by the processor, a communication session between an account of at least one user among users registered to a messaging service and an AI account, the AI account associated with an AI engine of the AI learning system and registered to the messaging service;

causing, by the processor, the AI account to transmit questions to the at least one user;

receiving, by the processor, from the account of the at least one user answers to the questions through the verified communication session;

generating, by the processor, a profile of the at least one user based on the questions from the AI account and the answers from the account of the at least one user;

designating, by the processor, one of the at least one user as an answerer based on the profile of the at least one user;

transmitting, by the processor, the question data to the answerer through the verified communication session as an instant message of the AI account;

receiving, by the processor, answer data to the question data from the answerer through the verified communication session;

generating, by the processor, learning data by mutually associating and storing the question data and the received answer data, to train the AI engine based on the question data and the received answer data;

receiving question data of a first user through a first communication session, the first communication session between an account of the first user among users registered to a messaging service and an AI account;

selecting a second user to which the question data of the first user is to be transmitted from among the users registered to the messaging service;

transmitting the question data to the second user through a second communication session between an account of the second user and the AI account;

receiving answer data to the question data of the first user from the second user through the second communication session; and

transmitting the received answer data from the second user to the first user as an answer for the question data of the first user through the first communication session, to train the AI engine based on the received answer data and the question data.

2. The method of claim 1 , wherein

the server is a messaging server, in which the AI learning system is included, or to which the AI learning system is connected over a network,

the messaging server is configured to provide the messaging service, and

the verifying comprises,

identifying the at least one user among the users registered to the messaging service, and

setting the communication session between the account of the at least one user and the AI account through the messaging server.

3. The method of claim 1 , wherein

the selecting question data includes selecting or generating the question data using the database, which is included in the AI learning system, or is connected to the AI learning system over a network, and

the database is generated based on at least one of instant messages transmitted and received between the users registered to the messaging service, instant messages transmitted and received between the users registered to the messaging service and the AI account, or previously generated learning data.

4. The method of claim 1 , wherein the selecting question data comprises generating the question data using an instant message received through a first communication session between an account of a first user among the users registered to the messaging service and the AI account.

5. The method of claim 4 , further comprising:

transmitting the received answer data through the first communication session for the account of the first user.

6. The method of claim 5 , wherein the generating the learning data comprises generating the learning data by further using response data, the response data being data received from the first user in response to the received answer data transmitted through the communication session between the account of the first user and the AI account.

7. The method of claim 1 , wherein

the selecting question data comprises selecting the question data based on a question message received through a first communication session between an account of a first user among the users registered to the messaging service and the AI account,

the verifying comprises selecting the at least one user based on a previous history of each of the users registered to the messaging service, and

the previous history includes at least one of

a location of a corresponding user registered to the messaging service,

a time at which the question message is transmitted through the messaging service,

a time at which the answer data to the question data is transmitted through the messaging service, and

a time at which an answer to the question data is refused.

8. The method of claim 1 , wherein the at least one user is randomly selected from among the users registered to the messaging service or selected from among the users based on the question data.

9. The method of claim 1 , further comprising:

analyzing the received question data of the first user and determining whether the trained AI engine is capable of providing an answer for the question data of the first user; and

in response to a result of the analyzing indicating that the trained AI engine is not capable of providing an answer for the question data of the first user, performing the selecting, the transmitting the question data to the second user, and the receiving answer data from the second user.

10. The method of claim 9 , wherein the analyzing comprises:

searching a database connected to the answer relay system or a document on a website using a keyword acquired by analyzing the question data; and

determining whether the second user is capable of providing the answer based on a result of the searching.

11. The method of claim 9 , wherein in response to the result of the analyzing indicating that the second user is not capable of providing the answer to the question data, the transmitting the question data to the second user includes transmitting the question data to a third user through a third communication session, the third communication session between an account of the third user and the AI account.

12. The method of claim 1 , wherein

the transmitting the question data to the second user includes transmitting the question data by selecting the second user based on a previous history of each of the users registered to the messaging service, and

the previous history includes at least one of

a location of a corresponding user registered to the messaging service,

a time at which a question message including the question data is transmitted through the messaging service,

a time at which the answer data to the question data is transmitted through the messaging service, and

a time at which an answer to the question data is refused.

13. The method of claim 1 , wherein the transmitting the question data to the second user includes transmitting the question data to the second user randomly selected from among the users registered to the messaging service or selected based on the question data.

14. An artificial intelligence (AI) learning system implemented on a computer, the system comprising:

a memory configured to store non-transitory computer-readable instructions; and

at least one processor configured to execute the non-transitory computer-readable instructions such that the at least one processor causes the system to,

select question data from a database associated with a server that includes or is connected to the AI learning system,

verify a communication session set between an account of at least one user among users registered to a messaging service and an AI account, the AI account associated with an AI engine of the AI learning system and registered to the messaging service,

cause the AI account to transmit questions to the at least one user,

receive from the account of the at least one user answers to the questions through the verified communication session,

generate a profile of the at least one user based on the questions from the AI account and the answers from the account of the at least one user,

designate one of the at least one user as an answerer based on the profile of the at least one user,

transmit the question data to the answerer through the verified communication session as an instant message of the AI account,

receive answer data to the question data from the answerer through the verified communication session,

generate learning data by mutually associating and storing the question data and the received answer data, for training the AI engine based on the question data and the received answer data,

receive question data of a first user through a first communication session, the first communication session between an account of the first user among users registered to a messaging service and an AI account,

select a second user to which the question data of the first user is to be transmitted from among the users registered to the messaging service,

transmit the question data to the second user through a second communication session between an account of the second user and the AI account,

receive answer data to the question data of the first user from the second user through the second communication session, and

transmit the received answer data from the second user to the first user as an answer for the question data of the first user through the first communication session, to train the AI engine based on the received answer data and the question data.

15. The system of claim 14 , wherein

the server is a messaging server, in which the AI learning system is included, or to which the AI learning system is connected over a network,

the messaging server is configured to provide the messaging service, and

the at least one processor is configured to verify the communication session by

identifying the at least one user among the users registered to the messaging service, and

setting a new communication session between the account of the at least one user and the AI account through the messaging server.

16. The method of claim 1 , further comprising:

storing, by the AI learning system, the learning data in a learning database included in or is in communication with the server over a network; and

acquiring, by the AI learning system, answer data having improved accuracy to the question data from the learning database through a process of continuously filtering the answer data suitable for the question data, to train the AI engine.

17. The method of claim 1 , further comprising:

storing, by the AI learning system, the learning data in a learning database included in or is in communication with the server over a network; and

acquiring, by the AI learning system, answer data having improved accuracy to the question data from the learning database through a process of continuously filtering the acquired answer data suitable for the question data, to train the AI engine.

18. The system of claim 14 , wherein the at least one processor is further configured to execute the non-transitory computer-readable instructions such that the at least one processor causes the system to

store the learning data in a learning database included in or is in communication with the server over a network; and

acquire answer data having improved accuracy to the question data from the learning database through a process of continuously filtering the answer data suitable for the question data, to train the AI engine.

Assignments (8)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Apr 12, 2024
From: Z INTERMEDIATE GLOBAL CORPORATION
To: LY CORPORATION
Reel/Frame 067096/0431 →
CHANGE OF NAME Recorded Apr 10, 2024
From: LINE CORPORATION
To: Z INTERMEDIATE GLOBAL CORPORATION
Reel/Frame 067069/0467 →
CORRECTIVE ASSIGNMENT TO CORRECT THE THE CITY SHOULD BE SPELLED AS TOKYO PREVIOUSLY RECORDED AT REEL: 058597 FRAME: 0141. ASSIGNOR(S) HEREBY CONFIRMS THE ASSIGNMENT. Recorded Jan 17, 2023
From: LINE CORPORATION
To: A HOLDINGS CORPORATION
Reel/Frame 062401/0328 →
CORRECTIVE ASSIGNMENT TO CORRECT THE SPELLING OF THE ASSIGNEES CITY IN THE ADDRESS SHOULD BE TOKYO, JAPAN PREVIOUSLY RECORDED AT REEL: 058597 FRAME: 0303. ASSIGNOR(S) HEREBY CONFIRMS THE ASSIGNMENT. Recorded Jan 17, 2023
From: A HOLDINGS CORPORATION
To: LINE CORPORATION
Reel/Frame 062401/0490 →
CHANGE OF ADDRESS Recorded Dec 28, 2021
From: LINE CORPORATION
To: LINE CORPORATION
Reel/Frame 059511/0374 →
CHANGE OF NAME Recorded Dec 28, 2021
From: LINE CORPORATION
To: A HOLDINGS CORPORATION
Reel/Frame 058597/0141 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Dec 28, 2021
From: A HOLDINGS CORPORATION
To: LINE CORPORATION
Reel/Frame 058597/0303 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Feb 17, 2017
From: HO KANG, SEOK; LEE, JAE GWANG; JUNG, JEA SEUNG; SEO, HEE-CHEOL; LEE, JOONGJAE; LEE, JEEHYUN; LIM, YOUNG-SIK; LEE, INJAE
To: LINE CORPORATION
Reel/Frame 041282/0929 →
Priority Claims (2)
KR 10-2016-0021895 · Feb 24, 2016 · national
KR 10-2016-0042494 · Apr 6, 2016 · national
Continuity (1)
Related Publication 20170243136A1 · Aug 24, 2017
Cited By (1)
US 12,437,160