Granted Patent B2

US 12,725,530 · App. 17/853,015 · Granted Sep 1, 2026

Inventor: Hyun Bin Loh (Seoul, KR)

Assignee: Socra AI Inc.

G09B7/04G06F17/18

Abstract

Claims (67)

1 . A method of recommending educational content by a device for analyzing learning data of a user, the device including a transceiver configured to communicate with a user terminal, a memory storing computer-readable instructions and a processor configured to execute the instructions to perform the method comprising:

acquiring the learning data of the user received from the user terminal through the transceiver, wherein the learning data includes first learning ability information of the user at a first time point, second learning ability information of the user at a second time point, and question answering information of the user during a time period between the first time point and the second time point;

acquiring target learning ability information of the user using a trained neural network model based on the learning data;

monitoring computing resources of the device, wherein the computing resources include a computation amount, memory or a network of the device, and wherein the monitoring comprises:

(i) continuously monitoring available computing resources of the device,

(ii) acquiring resource information on computing resources to be distributed based on the continuously monitored available computing resources, and

(iii) determining the computing resources to be distributed based on the resource information;

determining a neural network model based on the resource information and the target learning ability information;

distributing the computing resources of the device to correspond to the determined neural network model, wherein the distributing of the computing resources comprises adjusting the computing resources to be distributed to computing resources corresponding to the determined neural network model; and

acquiring educational content to be recommended to the user through the determined neural network model and transmitting the educational content to the user terminal through the transceiver,

wherein the determining of the neural network model comprises:

determining a first neural network model which demands first computing resources when the target learning ability information of the user includes a first target learning ability value and determining a second neural network model which demands second computing resources greater than the first computing resources when the target learning ability information of the user includes a second target learning ability value lower than the first target learning ability value, and

wherein the distributing of the computing resources comprises:

distributing the first computing resources to the first neural network model based on the first neural network model being determined and distributing the second computing resources to the second neural network model based on the second neural network model being determined, and

wherein the acquiring of the educational content comprises:

acquiring a first educational content through the first neural network model based on the first neural network model being determined and acquiring a second educational content through the second neural network model based on the second neural network model being determined, at least a part of the second educational content being different from the first educational content.

2 . The method of claim 1 , wherein the acquiring of the target learning ability information comprises:

calculating maximum learning ability information based on the learning data; and

acquiring the target learning ability information based on the maximum learning ability information,

wherein the target learning ability information is determined to be a predetermined ratio of a maximum learning ability value included in the maximum learning ability information.

3 . The method of claim 2 , wherein the calculating of the maximum learning ability information comprises:

generating a probability distribution graph related to a predicted learning ability of the user based on the first learning ability information, the second learning ability information, and the question answering information; and

calculating the maximum learning ability information based on the probability distribution graph.

4 . The method of claim 3 , wherein the calculating of the maximum learning ability information based on the probability distribution graph comprises:

acquiring rate-of-change information of the probability distribution graph;

acquiring first rate-of-change information including a smaller value than a predetermined rate of change in the rate-of-change information; and

determining a predicted learning ability of the user at a time point corresponding to the first rate-of-change information as the maximum learning ability information.

5 . A non-transitory computer-readable recording medium in which a computer program executed by a computer is recorded, the computer program comprising:

acquiring learning data of a user received from a user terminal through a transceiver of the computer, wherein the learning data includes first learning ability information of the user at a first time point, second learning ability information of the user at a second time point, and question answering information of the user during a time period between the first time point and the second time point;

acquiring target learning ability information of the user using a trained neural network model based on the learning data;

monitoring computing resources of the computer, wherein the computing resources include a computation amount, memory or a network of the computer, and wherein the monitoring comprises:

(i) continuously monitoring available computing resources of the computer,

(ii) acquiring resource information on computing resources to be distributed based on the continuously monitored available computing resources, and

(iii) determining the computing resources to be distributed based on the resource information;

determining a neural network model based on the resource information and the target learning ability information;

distributing the computing resources of the computer to correspond to the determined neural network model, wherein the distributing of the computing resources comprises adjusting the computing resources to be distributed to computing resources corresponding to the determined neural network model; and

acquiring educational content to be recommended to the user through the determined neural network model and transmitting the educational content to the user terminal through the transceiver,

wherein the determining of the neural network model comprises:

determining a first neural network model which demands first computing resources when the target learning ability information of the user includes a first target learning ability value and determining a second neural network model which demands second computing resources greater than the first computing resources when the target learning ability information of the user includes a second target learning ability value lower than the first target learning ability value,

wherein the distributing of the computing resources comprises:

distributing the first computing resources to the first neural network model based on the first neural network model being determined and distributing the second computing resources to the second neural network model based on the second neural network model being determined, and

wherein the acquiring of the educational content comprises:

acquiring a first educational content through the first neural network model based on the first neural network model being determined and acquiring a second educational content through the second neural network model based on the second neural network model being determined, at least a part of the second educational content being different from the first educational content.

6 . A device for receiving learning data of a user from an external user terminal and recommending educational content, the device comprising:

a transceiver configured to communicate with the user terminal;

a memory storing computer-readable instructions; and

a controller configured to execute the instructions to acquire the learning data of the user received from the user terminal through the transceiver and determine educational content based on the learning data,

wherein the learning data includes first learning ability information of the user at a first time point, second learning ability information of the user at a second time point, and question answering information of the user during a time period between the first time point and the second time point,

wherein the controller is further configured to execute the instructions to:

acquire target learning ability information of the user using a trained neural network model based on the learning data,

monitor computing resources of the device, wherein the computing resources include a computation amount, memory or a network of the device, and wherein the monitoring comprises:

(i) continuously monitoring available computing resources of the device,

(ii) acquiring resource information on computing resources to be distributed based on the continuously monitored available computing resources, and

(iii) determining the computing resources to be distributed based on the resource information,

determine a neural network model based on the resource information and the target learning ability information,

distribute the computing resources of the device to correspond to the determined neural network model, wherein the distributing of the computing resources comprises adjusting the computing resources to be distributed to computing resources corresponding to the determined neural network model, and

acquire educational content to be recommended to the user through the determined neural network model and transmit the educational content to the user terminal through the transceiver,

wherein the controller is further configured to execute the instructions to:

determine a first neural network model which demands first computing resources when the target learning ability information of the user includes a first target learning ability value and determine a second neural network model which demands second computing resources greater than the first computing resources when the target learning ability information of the user includes a second target learning ability value lower than the first target learning ability value,

wherein the controller is further configured to execute the instructions to:

distribute the first computing resources to the first neural network model based on the first neural network model being determined and distribute the second computing resources to the second neural network model based on the second neural network model being determined, and

wherein the controller is further configured to execute the instructions to:

acquire a first educational content through the first neural network model based on the first neural network model being determined and acquire a second educational content through the second neural network model based on the second neural network model being determined, at least a part of the second educational content being different from the first educational content.

7 . The device of claim 6 , wherein the controller is further configured to execute the instructions to acquire maximum learning ability information based on the learning data and acquire the target learning ability information based on the maximum learning ability information,

wherein the target learning ability information is determined to be a predetermined ratio of a maximum learning ability value included in the maximum learning ability information.

8 . The device of claim 7 , wherein the controller is further configured to execute the instructions to generate a probability distribution graph related to a predicted learning ability of the user based on the first learning ability information, the second learning ability information, and the question answering information and calculate the maximum learning ability information based on the probability distribution graph.

9 . The device of claim 8 , wherein the controller is further configured to execute the instructions to acquire rate-of-change information of the probability distribution graph, acquire first rate-of-change information including a smaller value than a predetermined value in the rate-of-change information, and determine a predicted learning ability of the user at a time point corresponding to the first rate-of-change information as the maximum learning ability information.

Assignments (3)

Priority Claims (1)

Continuity (1)

References Cited (9)