WO2023282539A1

WO2023282539A1 - Education content recommendation method, education content recommendation device, and education content recommendation system

Info

Publication number: WO2023282539A1
Application number: PCT/KR2022/009429
Authority: WO
Inventors: 노현빈
Original assignee: (주)뤼이드
Priority date: 2021-07-09
Filing date: 2022-06-30
Publication date: 2023-01-12
Also published as: KR102626442B1; KR20230009817A

Abstract

An education content recommendation method according to an embodiment of the present application comprises the steps of: acquiring learning data of a target user, wherein the learning data includes log data including problem data related to a problem solved by the target user before and response data related to the target user's response to the problem; acquiring a problem database including at least one candidate problem; calculating expected correct answer rate information of the target user for the candidate problem on the basis of the candidate problem and the learning data; acquiring the target user's ability information associated with at least a part of the log data, on the basis of the expected correct answer rate information; and determining a recommended content on the basis of the ability information of the target user.

Description

Educational contents recommendation method, educational contents recommendation device, and educational contents recommendation system

This application relates to a method for recommending educational contents, an apparatus for recommending educational contents, and a system for recommending educational contents. Specifically, the present application relates to a method for recommending educational contents, an apparatus for recommending educational contents, and a system for recommending educational contents for calculating user's ability information from user's learning data and recommending educational contents based on the user's ability information.

As artificial intelligence technology develops, the field of educational technology that diagnoses a user's learning ability and recommends educational contents based on the diagnosis result is drawing attention.

However, conventional techniques are simply aimed at increasing a user's expected score for a specific test. For example, research is being conducted focusing on improving the accuracy of a technology that recommends educational contents that increase the score of a specific official test. However, in actual learning, training on basic abilities required for learning, such as how much concentration can be maintained during the test time in order to increase the score of an official test and how to increase a user's reasoning ability, is required in actual learning.

However, there is a lack of research on educational content recommendation systems that can train basic abilities that are the basis for learning. Accordingly, development of a method, apparatus, and system for recommending educational contents capable of training basic learning necessary for learning is required.

An object to be solved by the present invention is to provide a method for recommending educational contents, an apparatus for recommending educational contents, and a system for recommending educational contents for quantifying various types of ability information of a user.

An object to be solved by the present invention is to provide an educational content recommendation method, an educational content recommendation device, and an educational content recommendation system for recommending optimal educational content for each ability type.

The problem to be solved by the present invention is not limited to the above-mentioned problems, and problems not mentioned will be clearly understood by those skilled in the art from this specification and the accompanying drawings. .

A method for recommending educational contents according to an embodiment of the present application includes acquiring learning data of a target user - the learning data includes problem data related to a problem previously solved by the target user and information about the target user for the problem. contains log data containing response data related to the response; obtaining a problem database including at least one candidate problem; calculating expected correct answer information of the target user for the candidate problem based on the candidate problem and the learning data; Obtaining ability information of the target user associated with at least some data of the log data based on the expected correct answer rate information - the ability information is first information representing the maximum learning ability of the target user, the target user's ability information related to at least one of second information representing reasoning ability, third information representing logic ability of the target user, and fourth information representing application ability of the target user; and determining recommended content based on the ability information of the target user, wherein the acquiring of the capability information of the target user comprises: predicting the target user based on the learning data and the expected correct answer rate information; generating a growth curve related to learning ability; and obtaining the first information based on the growth curve.

An apparatus for determining recommended content by receiving learning data of a user from an external user terminal device according to an embodiment of the present application includes a transceiver unit communicating with the user terminal; A controller configured to obtain learning data of a target user through the transceiver and calculate ability information of the target user based on the learning data, wherein the controller includes: learning data of the target user-the learning data a problem database comprising at least one candidate problem; obtaining and calculating expected correct answer information of the target user for the candidate problem based on the candidate problem and the learning data; Capability information of the target user associated with at least some data of the log data based on the expected percent correct answer information - the capability information is first information indicating the maximum learning ability of the target user and inference ability of the target user Relating to at least one of second information, third information indicating the logical ability of the target user, and fourth information indicating the application ability of the target user; obtaining recommended content based on the ability information of the target user Wherein the controller is configured to generate a growth curve related to the expected learning ability level of the target user based on the learning data and the expected percent correct information, and obtain the first information based on the growth curve to obtain the ability information It can be configured to obtain.

The solutions to the problems of the present invention are not limited to the above-described solutions, and solutions not mentioned will be clearly understood by those skilled in the art from this specification and the accompanying drawings. You will be able to.

According to the method, apparatus, and system for recommending educational contents according to an embodiment of the present application, by determining educational contents to be recommended to users according to various types of abilities, basic abilities required for learning, such as learning ability, application ability, and logical ability, can be trained. Educational content may be provided to the user.

Effects of the present invention are not limited to the above-mentioned effects, and effects not mentioned will be clearly understood by those skilled in the art from this specification and the accompanying drawings.

1 is a schematic diagram of an educational content recommendation system according to an embodiment of the present application.

2 is a diagram illustrating operations of an educational content recommendation device according to an embodiment of the present application.

3 is a diagram illustrating an aspect of recommended content determined according to a type of capability information according to an embodiment of the present application.

4 is a flowchart of a method for recommending educational content according to an embodiment of the present application.

5 is a detailed flowchart of a method of obtaining first ability information related to a maximum learning ability of a target user according to an embodiment of the present application.

6 is a diagram illustrating an aspect of obtaining first capability information of a target user according to an embodiment of the present application.

7 is a detailed flowchart of a method of acquiring second capability information related to a reasoning capability of a target user according to an embodiment of the present application.

8 is a diagram illustrating an aspect of calculating second capability information of a target user according to an embodiment of the present application.

According to the educational content recommendation method according to an embodiment of the present application, the obtaining of the first information based on the growth curve may include obtaining change rate information of the growth curve; obtaining target change rate information having a change rate equal to or smaller than a predetermined threshold change rate based on the change rate information; and obtaining the first information based on the predicted learning capability value corresponding to the target change rate information.

According to the method for recommending educational content according to an embodiment of the present application, the problem database includes information on average percent correct answers for the candidate questions, and the acquiring of ability information of the target user includes average percent correct answers for the candidate questions. and obtaining the second information of the target user by comparing information and the expected correct answer rate information of the target user for the candidate problem.

According to the method for recommending educational contents according to an embodiment of the present application, the obtaining of the second information of the target user may include: obtaining a reference candidate problem having an average correct answer rate equal to or lower than a predetermined standard correct answer rate; and calculating the second information of the target user by comparing the expected percent correct information of the target user with respect to the reference candidate problem and the average percent correct answer information with respect to the reference candidate problem.

According to the method for recommending educational contents according to an embodiment of the present application, the determining of the recommended contents may include obtaining a set of educational contents; calculating an expected change in the ability information of the target user when the content included in the educational content set is provided to the target user; and determining, as the recommended content, content having a maximum expected change in the capability information.

According to the educational content recommendation method according to an embodiment of the present application, the step of determining the recommended content; determining a neural network model based on the capability information; distributing resources corresponding to the determined model; and obtaining the recommended content through the determined neural network model.

According to one embodiment of the present application, a computer-readable recording medium on which a program for executing the educational content recommendation method may be provided.

According to the apparatus for recommending educational contents according to an embodiment of the present application, the controller obtains change rate information of the growth curve, and has a change rate equal to or smaller than a predetermined threshold change rate based on the change rate information. It may be configured to obtain target change rate information and obtain the first information based on the predicted learning capability corresponding to the target change rate information.

According to the apparatus for recommending educational contents according to an embodiment of the present application, the problem database includes information on average percent correct answers for the candidate problems, and the controller includes information about average percent correct answers for the candidate problems and information about the average percent correct answers for the candidate problems. and acquire the ability information by obtaining the second information of the target user by comparing the expected correct answer rate information of the target user.

According to the apparatus for recommending educational contents according to an embodiment of the present application, the controller obtains a standard candidate problem having an average correct rate equal to or lower than a predetermined standard correct rate, and the target user's prediction for the standard candidate problem. It may be configured to calculate the second information of the target user by comparing the correct answer information with the average correct answer information for the reference candidate problem.

According to the apparatus for recommending educational contents according to an embodiment of the present application, the controller obtains a set of educational contents, and when the contents included in the set of educational contents are provided to the target user, the ability information of the target user It may be configured to calculate an expected change, and determine content having a maximum expected change of the capability information as the recommended content.

According to the apparatus for recommending educational contents according to an embodiment of the present application, the controller determines a neural network model based on the capability information, distributes resources corresponding to the determined model, and recommends the recommendation through the determined neural network model. It can be configured to obtain content.

The foregoing objects, features and advantages of the present application will become more apparent from the following detailed description taken in conjunction with the accompanying drawings. However, the present application can apply various changes and can have various embodiments. Hereinafter, specific embodiments will be illustrated in the drawings and described in detail.

Like reference numerals designate essentially like elements throughout the specification. In addition, components having the same function within the scope of the same idea appearing in the drawings of each embodiment will be described using the same reference numerals, and overlapping descriptions thereof will be omitted.

If it is determined that a detailed description of a known function or configuration related to the present application may unnecessarily obscure the subject matter of the present application, the detailed description thereof will be omitted. In addition, numbers (eg, first, second, etc.) used in the description process of this specification are only identifiers for distinguishing one component from another component.

In addition, the suffixes "module" and "unit" for components used in the following embodiments are given or used interchangeably in consideration of ease of writing the specification, and do not have meanings or roles that are distinguished from each other by themselves.

In the following examples, expressions in the singular number include plural expressions unless the context clearly dictates otherwise.

In the following embodiments, terms such as include or have mean that features or components described in the specification exist, and do not preclude the possibility that one or more other features or components may be added.

In the drawings, the size of components may be exaggerated or reduced for convenience of description. For example, the size and thickness of each component shown in the drawings are arbitrarily shown for convenience of explanation, and the present invention is not necessarily limited to those shown.

If an embodiment is otherwise implementable, the order of specific processes may be performed differently from the order described. For example, two processes that are described in succession may be performed substantially concurrently, or may proceed in an order reverse to that described.

In the following embodiments, when components are connected, a case in which the components are directly connected as well as a case in which components are interposed between the components and connected indirectly is included.

For example, when it is said that components are electrically connected in this specification, not only the case where the components are directly electrically connected, but also the case where the components are interposed and electrically connected indirectly is included.

Hereinafter, a method for recommending educational contents, an apparatus for recommending educational contents, and a system for recommending educational contents according to the present application will be described with reference to FIGS. 1 to 8 .

The educational content recommendation system 10 according to an embodiment of the present application may include a user terminal 100 and an educational content recommendation device 1000 .

The user terminal 100 may obtain educational content from the educational content recommendation device 1000 or any external device. For example, the user terminal 100 may receive recommended content determined from the educational content recommendation device 1000 and display the received recommended content to the user through an arbitrary output unit. Subsequently, the user may input a response to the suggested content into the user terminal 100 through an arbitrary input unit.

The user terminal 100 may acquire learning data based on the user's response and transmit the user's learning data to the educational content recommendation device 1000 . Here, the learning data may include problem identification information solved by the user, user response information and/or incorrect answer information, log data, and the like. Meanwhile, the user terminal 100 may transmit user information to the educational content recommendation device 1000 .

The apparatus 1000 for recommending educational contents according to an embodiment of the present application may include a transceiver 1100 , a memory 1200 and a controller 1300 .

The transceiver 1100 may communicate with any external device including the user terminal 100 . For example, the educational content recommendation device 1000 may receive user learning data and/or user information from the user terminal 100 or transmit recommended content to the user terminal 100 through the transceiver 1100. .

The apparatus 1000 for recommending educational contents may transmit and receive various types of data by accessing a network through the transceiver 1100 . The transceiver 1200 may largely include a wired type and a wireless type. Since the wired type and the wireless type each have advantages and disadvantages, the wired type and the wireless type may be simultaneously provided in the apparatus 1000 for recommending educational contents according to circumstances. Here, in the case of the wireless type, a wireless local area network (WLAN)-based communication method such as Wi-Fi may be mainly used. Alternatively, in the case of a wireless type, a cellular communication, eg, LTE, 5G-based communication method may be used. However, the wireless communication protocol is not limited to the above example, and any suitable wireless type communication method may be used.

In the case of a wired type, LAN (Local Area Network) or USB (Universal Serial Bus) communication is a representative example, and other methods are also possible.

The memory 1200 may store various types of information. Various types of data may be temporarily or semi-permanently stored in the memory 1200 . Examples of the memory 1200 include a hard disk drive (HDD), a solid state drive (SSD), a flash memory, a read-only memory (ROM), a random access memory (RAM), and the like. This can be. The memory 1200 may be provided in a form embedded in the educational content recommendation device 1000 or in a detachable form. The memory 1200 includes an operating system (OS) for driving the educational content recommendation device 1000 or a program for operating each component of the educational content recommendation device 1000, as well as the contents of the educational content recommendation device 1000. Various data required for operation may be stored.

The controller 1300 may control overall operations of the educational content recommendation device 1000 . For example, the controller 1300 obtains candidate problems and average percent correct information from a problem database, which will be described later, and calculates expected percent correct information for the user's candidate problems based on the candidate problems and the user's learning data. and overall operations of the apparatus 1000 for recommending educational contents, such as obtaining user ability information based on the average percent correct answer information or acquiring recommended contents based on the user's ability information. In detail, the controller 1300 may load and execute a program for overall operation of the educational content recommendation device 1000 from the memory 1200 . The controller 1300 may be implemented as an application processor (AP), a central processing unit (CPU), or a similar device according to hardware, software, or a combination thereof. In this case, in terms of hardware, it may be provided in the form of an electronic circuit that processes electrical signals to perform a control function, and in terms of software, it may be provided in the form of a program or code that drives a hardware circuit.

Hereinafter, the operation of the educational content recommendation device 1000 according to an embodiment of the present application will be described in detail with reference to FIGS. 2 to 8 .

The apparatus 1000 for recommending educational contents according to an embodiment of the present application may perform an operation of recommending educational contents based on user learning data.

Conventional technologies simply aimed at increasing a user's expected score for a specific test. For example, research is being conducted focusing on improving the accuracy of a technology that recommends educational contents that increase the score of a specific official test. However, in actual learning, training on basic abilities required for learning, such as how much concentration can be maintained during the test time in order to increase the score of an official test and how to increase a user's reasoning ability, is required in actual learning. However, there is a lack of research on educational content recommendation systems that can train basic abilities that are the basis for learning.

The apparatus 1000 for recommending educational contents according to an embodiment of the present application includes basic abilities (eg, learning capacity, reasoning power, logic, Application ability, basic strength, concentration, grit, etc.) are evaluated and, based on this, recommended contents are determined for each ability type, so that optimal training for various abilities related to learning can be provided to the user.

Hereinafter, with reference to FIG. 2 , an operation of the educational content recommendation device 1000 according to an embodiment of the present application for achieving the above-described objects and effects will be described in detail. 2 is a diagram illustrating operations of the apparatus 1000 for recommending educational contents according to an embodiment of the present application.

The apparatus 1000 for recommending educational content according to an embodiment of the present application may obtain a problem database from an arbitrary database. Here, the problem database may include information on at least one candidate problem and average correct answer information on the candidate problem.

Also, the apparatus 1000 for recommending educational content according to an embodiment of the present application may obtain user learning data from a database. Here, the learning data, as described above, may mean any data related to the user's learning, such as problem identification information previously solved by the user, user response information and/or incorrect answer information to the problem. . For example, the learning data may include log data including problem data related to a problem previously solved by the user and response data related to the user's response to the problem.

Alternatively, the apparatus 1000 for recommending educational contents according to an embodiment of the present application may include learning data including user response information obtained through the user terminal 100 .

The apparatus 1000 for recommending educational contents according to an embodiment of the present application may calculate an expected correct answer rate for a user's candidate problem based on information on candidate problems obtained from a problem database and learning data of the user. Various artificial neural network models, including RNN, LSTM, bidirectional LSTM, or transformer-structured artificial neural networks, can be used to calculate the expected percent correct. Alternatively, any suitable algorithm may be used, as well as artificial neural network models.

The apparatus 1000 for recommending educational content according to an embodiment of the present application may perform an operation of evaluating or calculating a user's ability. For example, the apparatus 1000 for recommending educational content may calculate ability information by evaluating the user's ability based on the user's learning data and problem database. Here, the ability refers to the user's ability related to learning that can be diagnosed using any method such as the user's current score on various official tests, prediction score, reasoning ability, logic power, application ability, basic strength, grit, concentration, potential ability, etc. It can mean inclusive. In addition, the ability information may include any type of information that quantifies or can quantify a user's ability related to the aforementioned diagnosable learning.

The apparatus 1000 for recommending educational content according to an embodiment of the present application may obtain user ability information based on the user's learning data and a problem database. A method of calculating the user's capability information will be described later in detail with reference to FIGS. 4 to 8 .

The apparatus 1000 for recommending educational content according to an embodiment of the present application may perform an operation of determining recommended content based on capability information.

See Figure 3. 3 is a diagram illustrating an aspect of recommended content determined according to a type of capability information according to an embodiment of the present application.

As an example, when the apparatus 1000 for recommending educational contents acquires first ability information of a first type indicating the learning ability of the user, the apparatus for recommending educational contents 1000 converts the first ability information to the user's learning ability. The first content set may be determined as recommended content by linking the data to a specific point of log data included in the data. The first set of contents may be educational contents useful for improving the maximum learning ability related to the user's first ability information.

As another example, when the apparatus 1000 for recommending educational contents obtains the second ability information of the second type representing reasoning power of the user, the apparatus for recommending educational contents 1000 converts the second ability information to the user's learning The second content set may be determined as the recommended content by linking the data to a specific point of the log data included in the data. The second content set may be educational content that is advantageous for improving reasoning ability related to the user's second ability information.

As another example, when the apparatus 1000 for recommending educational contents obtains the third ability information of the third type representing the logical power of the user, the apparatus for recommending educational contents 1000 converts the third ability information to the user's learning data. A third content set may be determined as recommended content by linking data to a specific point of log data. The third content set may be educational content that is advantageous for improving logic related to the user's third ability information.

As another example, when the apparatus 1000 for recommending educational contents obtains the fourth ability information of the fourth type representing the application ability of the user, the apparatus for recommending educational contents 1000 converts the fourth ability information to the user's learning data. A fourth content set may be determined as recommended content by linking data to a specific point of log data. The fourth content set may be educational content that is advantageous for improving the application ability related to the user's fourth capability information.

However, the above-mentioned type of ability information is only an example, and the educational content recommendation device 1000 obtains information on abilities related to arbitrary learning, such as Grit (ability related to fighting spirit or courage) and concentration, and maximizes the corresponding ability. A set of content that can be played may be determined as recommended content.

Meanwhile, the apparatus 1000 for recommending educational content according to an embodiment of the present application may determine recommended content based on expected correct answer information. The recommended content may be educational content having the highest expected score calculated based on the expected correct answer rate information.

According to an embodiment, the apparatus 1000 for recommending educational content may use a learning degree when calculating an expected score that is a basis for determining recommended content. The degree of learning may include information about an educational effect generated when learning educational content, such as reading a commentary on a candidate problem or taking a related lecture. For example, the expected score reflecting the learning degree may be calculated by adding “the value obtained by multiplying the learning degree by the maximum expected score” and “the value obtained by multiplying the non-learning degree by the expected score without the learning degree reflected”.

Referring back to FIG. 2 , the apparatus 1000 for recommending educational content according to an embodiment of the present application may transmit recommended content to the user terminal 100 through the transceiver 1100 .

The user terminal 100 may display recommended content to the user through an arbitrary output unit. The user may input a response to the recommended content through an arbitrary input unit. The user's input may be updated to the learning data and may be stored in a database or any storage device.

Meanwhile, although not shown in FIG. 2 , the apparatus 1000 for recommending educational contents according to an embodiment of the present application calculates the user's expected score for each case where a specific problem is correct and incorrect based on the user's learning data. action can be performed. In this case, the expected score when a specific problem is correct may be the maximum expected score, and the expected score when the problem is incorrect may be the minimum expected score. The user's learning data may be updated in real time whenever the user solves a problem. In this case, the calculated expected score of the user may be used to calculate the expected score.

4 is a flowchart of a method for recommending educational content according to an embodiment of the present application. The educational content recommendation method according to an embodiment of the present application includes acquiring learning data of a target user (S1000), acquiring a problem database (S2000), calculating expected correct answer information (S3000), and It may include acquiring capability information (S4000) and acquiring recommended content (S5000).

In the step of acquiring learning data of the target user ( S1000 ), the apparatus 1000 for recommending educational content may obtain learning data from a database. Specifically, the apparatus 1000 for recommending educational contents may be implemented to recognize user information of a target user and acquire learning data related to the target user from a database. Alternatively, the educational content recommendation apparatus 1000 may obtain learning data of a target user from a response received from a user terminal. Here, the learning data, as described above, may include problem identification information solved by the user, response information and/or incorrect answer information, and the like.

Alternatively, the learning data may include score information of a target user over time. For example, the learning data may include first score information at a first point in time and/or second point information at a second point in time. For a specific example, the first score information may be a score for an official test (eg, TOEIC, SAT, CSAT, etc.) of the target user at the first time point. The second score information may be score information on an official test (eg, TOEIC, SAT, CSAT, etc.) of the target user at the second time point. However, this is only an example, and the learning data can be diagnosed using any method such as the user's current score on various official tests, prediction score, reasoning ability, logic power, application ability, basic strength, concentration, grit, potential ability, etc. It may mean to encompass arbitrary data of a user related to learning.

In the step of obtaining a problem database ( S2000 ), the apparatus 1000 for recommending educational content may acquire a problem database from the database. The problem database may include at least one candidate problem and average correct answer information of reference users for the candidate problem.

In the step of calculating expected percent correct information (S3000), the educational content recommendation apparatus 1000 may calculate the expected percent correct answer for the target user's candidate problem based on the target user's learning data and the candidate problem obtained from the problem database. Specifically, the apparatus 1000 for recommending educational content may predict a correct answer rate for candidate questions based on the learning data of the target user. Various artificial neural network models, including RNN, LSTM, bidirectional LSTM, or transformer-structured artificial neural networks, can be used for predicting the percent correct. For example, when the predicted correct answer rate for a candidate problem is obtained using a transformer-structured artificial neural network, information about the candidate problem is input to the encoder side and learning data (eg, response information) of the target user is input to the decoder side to determine the target for the candidate problem. An expected correct answer rate of the user may be obtained.

In the step of acquiring ability information of the target user (S4000), the educational content recommendation apparatus 1000 determines the ability of the user based on the expected percent correct information for the target user's candidate problem and the average percent correct answer information of reference users for the candidate problem. can be evaluated or quantified.

For example, the apparatus 1000 for recommending educational contents may quantify first ability information indicating the maximum learning ability of the target user based on the expected percentage correct information for the candidate problem of the target user and the average correct percentage information for the candidate problem of the reference user. can Specifically, the educational content recommendation device 1000 generates a growth curve related to the expected learning ability of the target user based on the expected correct rate information and/or the average correct answer rate information, and based on the growth curve, the target user's maximum learning ability. 1 Ability information can be calculated. Hereinafter, a method of calculating first capability information will be described in detail with reference to FIGS. 5 and 6 .

5 is a detailed flowchart of a method of obtaining first ability information related to a maximum learning ability of a target user according to an embodiment of the present application. 6 is a diagram illustrating an aspect of obtaining first capability information of a target user according to an embodiment of the present application.

The step of obtaining ability information of the target user (S4000) includes generating a growth curve related to the expected learning ability level of the target user (S4110), obtaining change rate information of the growth curve (S4120), and 1 may include acquiring capability information (S4130).

In the step of generating a growth curve related to the expected learning ability level of the target user (S4110), the educational contents recommendation apparatus 1000 is based on the expected correct answer information of the target user for the candidate problem or the average correct answer information of reference users for the candidate problem. Thus, a growth curve (f) related to the expected learning ability of the target user may be obtained. More specifically, the apparatus 1000 for recommending educational content may calculate the expected learning ability value of the target user for each of the case of correct candidate problem and the case of incorrect, based on the information on the expected percentage of correct answers. Also, the apparatus 1000 for recommending educational content may generate a growth curve f related to the expected learning capability of the target user based on the calculated predicted learning capability.

As another example, the apparatus 1000 for recommending educational contents may generate a growth curve related to the expected learning capability of the target user based on the target user's learning data. For example, the apparatus 1000 for recommending educational contents may be implemented to generate a growth curve f based on response information about problems previously solved by the target user, correct answer information, or score information of the target user. More specifically, the apparatus 1000 for recommending educational contents includes first point information of a target user at a first time point, second point information at a second time point, and learning of a target user at a time point between the first time point and the second time point. It may be implemented to predict a probability distribution related to an expected learning ability level of a target user using data. Also, the apparatus 1000 for recommending educational content may generate a growth curve f related to the user's expected learning ability based on the predicted probability distribution.

The apparatus 1000 for recommending educational content may estimate an expected learning ability of the user using an arbitrary algorithm and/or a trained neural network model, and may generate a growth curve f related to the expected learning ability.

In the step of acquiring change rate information of the growth curve ( S4120 ), the educational content recommendation apparatus 1000 may calculate change rate information from the growth curve f. For example, the apparatus 1000 for recommending educational content may obtain a first derivative f' of the growth curve f and calculate change rate information y' from the first derivative f'. As another example, the educational content recommendation apparatus 1000 may obtain the second derivative f″ of the growth curve f and calculate the change rate information y″ from the first derivative f″.

In the step of acquiring first capability information based on the rate of change information ( S4130 ), the apparatus 1000 for recommending educational content may calculate first capability information indicating the maximum learning capability of the target user based on the rate of change information.

For example, the apparatus 1000 for recommending educational contents may obtain first change rate information y'1 including a value equal to or smaller than a predetermined rate of change of the change rate information y' of the growth curve f of the target user. can Here, the apparatus 1000 for recommending educational contents may determine the expected learning ability value of the target user at time point t1 corresponding to the first rate of change information y'1 as the first ability information representing the maximum learning ability.

As another example, the apparatus 1000 for recommending educational contents may include change rate information y'' indicating how much the change rate y' of the growth curve f is slowing down from the second derivative f'' of the growth curve f. ) may be calculated, and second change rate information including a value equal to or smaller than a predetermined value of the change rate information y″ may be obtained. Here, the apparatus 1000 for recommending educational contents may determine the expected learning ability of the target user at the time point corresponding to the second rate of change information as the first ability information representing the maximum learning ability.

However, the content of calculating the first ability information based on the change rate information is only an example, and the educational content recommendation apparatus 1000 can calculate the first ability information representing the maximum learning ability level of the target user using any appropriate method. can be implemented so that For example, the apparatus 1000 for recommending educational content may calculate first capability information based on area information A of the growth curve f. For example, the educational content recommendation apparatus 1000 calculates first capability information representing the maximum learning capability of the target user based on the area information (A) of the generated growth curve (f) and the expected learning capability (y). can do. For example, when the area information (A/y) for the expected learning capability has a first value, the educational content recommendation apparatus 1000 may allocate first capability information including the first maximum learning capability to the target user, When the area information (A/y) of the expected learning capability value has the second value, the educational content device 1000 may allocate first capability information including the second maximum learning capability value to the target user.

In FIGS. 5 and 6 , a growth curve related to an expected learning ability level of a target user is generated and first ability information is acquired based on the growth curve. However, this is merely an example for convenience of description, and may be implemented to obtain first ability information related to the maximum learning ability of the target user based on any suitable method.

Referring back to FIG. 4 , in the step of acquiring capability information of the target user (S4000), the educational contents recommendation apparatus 1000 includes information on expected percent correct answers for candidate problems of the target user and an average of reference users for the corresponding candidate problems. Based on the correct answer rate information, the ability of the user may be evaluated or quantified.

For example, the apparatus 1000 for recommending educational contents may quantify second capability information indicating the reasoning ability of the target user based on expected percent correct information for candidate problems of the target user and/or average percent correct answer information for candidate problems of the reference user. can In detail, the apparatus 1000 for recommending educational contents may sort the candidate questions in the problem database in order of average correct answer based on the average correct answer information. In addition, the educational contents recommendation apparatus 1000 obtains a standard candidate problem based on a predetermined standard correct answer rate, compares the expected correct answer rate information of the target user for the standard candidate problem with the average correct answer rate information of standard users for the standard candidate problem Second ability information related to the reasoning ability of the target user may be calculated. Hereinafter, a method of calculating second capability information will be described in detail with reference to FIGS. 7 and 8 .

7 is a detailed flowchart of a method of acquiring second capability information related to a reasoning capability of a target user according to an embodiment of the present application. 8 is a diagram illustrating an aspect of calculating second capability information of a target user according to an embodiment of the present application.

Acquiring capability information of the target user, in particular, second capability information (S4000) includes obtaining standard candidate problems (S4210), and the target user's expected percent correct information and the standard user's average percent correct information for the standard candidate problems. It may include comparing and calculating second capability information of the target user (S4220).

In the step of obtaining a standard candidate problem (S4210), the apparatus 1000 for recommending educational content according to an embodiment of the present application may obtain at least one candidate problem obtained from a problem database and information on the average correct answer rate of standard users for the candidate problem. there is. Also, the apparatus 1000 for recommending educational contents may obtain information on an expected correct answer rate of a target user corresponding to at least one candidate problem. The apparatus 1000 for recommending educational contents may sort at least one candidate question in order of correct answer rate based on average correct answer information. In this case, the apparatus 1000 for recommending educational content may acquire a standard candidate problem based on a predetermined standard correct answer rate. For example, the apparatus 1000 for recommending educational content may obtain candidate problems having information on an average correct answer rate lower than a predetermined reference correct answer rate as standard candidate problems.

In step S4220 of calculating the second ability information of the target user by comparing the expected percent correct information and the average percent correct information for the standard candidate problem, the educational content recommendation apparatus 1000 provides average percent correct information of standard users for the standard candidate problem. Second ability information representing the reasoning ability of the target user may be calculated based on the target user's predicted correct answer rate information for the reference candidate problem. For example, referring to FIG. 8 , the first target user may indicate an expected correct answer rate that is relatively higher than the average correct answer rate of standard users with respect to standard candidate problems. In this case, the apparatus 1000 for recommending educational contents may obtain second capability information including a first reasoning capability value for the first target user. On the other hand, the second target user of FIG. 8 may show a relatively low expected correct answer rate for the standard candidate problem compared to the first target user. In this case, the educational content recommendation apparatus 1000 may obtain second capability information including a second reasoning capability value relatively lower than the first reasoning capability value for the second target user.

Meanwhile, the educational content recommendation apparatus 1000 may calculate second ability information related to the reasoning ability of the target user based on the expected correct answer information of the target user and the average correct answer information of the reference users. For example, the apparatus 1000 for recommending educational content calculates the second capability information of the target user based on the integral value of the average percent correct answer for the standard candidate problem of the reference users and the expected percent correct answer for the standard candidate problem of the target user. can For example, when the integral value of the expected rate of correct answers for the standard candidate problem of the target user has a first value and the integral value of the average correct rate of correct answer for the standard candidate problem of the reference users has a second value, the first value and Second capability information of the first target user may be calculated based on the second value. In detail, the apparatus 1000 for recommending educational contents may quantify the second capability information of the target user based on the difference between the first value and the second value. Specifically, since the difference between the first value and the second value is greater in the case of the first target user of FIG. 8 than in the case of the second target user of FIG. The capability information may be calculated to be relatively higher than the second capability information of the second target user.

7 and 8 focus on obtaining candidate problems having average percent correct information lower than the standard percent correct as standard candidate problems, and quantifying second ability information by comparing expected percent correct information and average percent correct information for the standard candidate problems. described as However, this is just for convenience of explanation, and candidate problems having average percent correct information higher than the standard percent correct are obtained as standard candidate problems, and based on the expected percent correct information and the average percent correct information, the second ability related to the reasoning ability of the target user is obtained. Of course, it can be configured to quantify information. In addition, as another example, a first weight is assigned to a candidate problem exhibiting an average correct rate lower than the standard correct rate, and a second weight is assigned to a candidate problem exhibiting an average correct rate higher than the standard correct rate, so that the second capability information of the target user is obtained. It can be implemented to compute. Also, a standard correct answer rate may be set in advance for this operation.

As another example, the apparatus 1000 for recommending educational contents may quantify third ability information representing the logical power of the target user based on the information about the expected rate of correct answers for the candidate problems of the target user and the average rate of correct answers for the candidate problems of the reference user. there is. The third capability information representing the logical ability of the target user may be quantified in a similar manner to the above-described second capability information representing the reasoning capability. For example, the apparatus 1000 for recommending educational contents may quantify the third capability information by comparing the average correct rate of reference users to the standard candidate problem with the expected correct rate of the target user. Specifically, when the educational content recommendation device 1000 acquires information that the expected correct answer rate for the target user's standard candidate problem is relatively higher than the average correct answer rate, the educational content recommendation device 1000 quantifies the logical power of the target user relatively high. can do. At this time, similar to the above-described method, the third ability information of the target user may be calculated based on the integral value of the average percent correct answers for the standard candidate problems of the reference users and the expected percent correct answers for the standard candidate problems of the target user. .

As another example, the apparatus 1000 for recommending educational contents may quantify fourth capability information indicating the application ability of the target user based on the expected percent correct information for the target user's candidate problem and the average percent correct answer information for the candidate problem of the reference user. there is. The fourth capability information representing the application ability of the target user may be quantified in a similar way to the above-described second capability information representing the reasoning capability. For example, the apparatus 1000 for recommending educational contents may quantify the fourth capability information by comparing the average correct rate of reference users to the standard candidate problem with the expected correct rate of the target user. Specifically, when the educational content recommendation device 1000 acquires information that the expected correct answer rate for the standard candidate problem of the target user is relatively higher than the average correct answer rate, the educational content recommendation device 1000 quantifies the target user's application power relatively high. can do. At this time, similar to the above-described method, the fourth ability information of the target user may be calculated based on the integral value of the average percent correct answers for the standard candidate problems of the reference users and the expected percent correct answers for the standard candidate problems of the target user. .

As another example, the apparatus 1000 for recommending educational content quantifies fifth capability information representing the basic strength of the target user based on the expected percent correct information for candidate problems of the target user and the average percent correct answer information for candidate problems of the reference user. can do. The fifth capability information representing the basic strength of the target user may be quantified in a similar manner to the above-described second capability information representing the reasoning capability. For example, the apparatus 1000 for recommending educational contents may quantify the fifth capability information by comparing the average correct rate of standard users to the standard candidate problem with the expected correct rate of the target user. However, the reference candidate problems used to quantify the fifth capability information may be candidate problems having average correct answer information higher than a predetermined standard correct answer rate among candidate problems. Specifically, when the educational content recommendation device 1000 acquires information that the expected correct answer rate for the target user's standard candidate problem is relatively higher than the average correct answer rate, the educational content recommendation device 1000 determines the target user's basic strength relatively. highly quantifiable. At this time, similar to the above-described method, the fifth ability information of the target user may be calculated based on the integral value of the average percent correct answers for the standard candidate problems of the reference users and the expected percent correct answers for the standard candidate problems of the target user. .

Meanwhile, the apparatus 1000 for recommending educational contents according to another embodiment of the present application may evaluate or quantify grit of a target user based on learning data. Specifically, the learning data may further include information about a learning time based on user login information. Here, the educational content recommendation apparatus 1000 may train a model for quantifying how long to study, ie, grit, by using information about the learning time of users. For example, the apparatus 1000 for recommending educational contents may train a model for quantifying the user's grit based on information about the learning time and label information to which a level of grit is assigned to the information about the learning time. Also, the apparatus 1000 for recommending educational contents may be configured to calculate grit information of the target user by using the learned model and information about the learning time of the target user.

Also, the apparatus 1000 for recommending educational content according to another embodiment of the present application may evaluate or quantify the concentration of a target user based on learning data. Specifically, the learning data may further include information about a learning time based on user login information. In particular, the apparatus 1000 for recommending educational content may calculate information about a problem solving time based on log data (eg, user login information) included in learning data. For example, the apparatus 1000 for recommending educational contents trains a model for obtaining information on problem solving time based on users' log data (eg, user's login information), and based on the information on problem solving time, the target user concentration can be quantified. Specifically, the apparatus 1000 for recommending educational contents may train a model for predicting information on the user's problem solving time based on the log data and the user's actual problem solving time. In addition, the educational content recommendation apparatus 1000 may obtain problem solving time information from log data of the target user using the learned model, and may be implemented to quantify the target user's concentration based on the obtained problem solving time information. can For example, the educational content recommendation apparatus 1000 may be implemented to quantify the level of concentration of the target user with a higher level value as the problem solving time information becomes longer.

Referring back to FIGS. 3 and 4 , the method for recommending educational content according to an embodiment of the present application may include acquiring recommended content (S5000).

In the step of acquiring recommended content (S5000), the apparatus 1000 for recommending educational content may determine recommended content based on capability information of the target user. For example, the apparatus 1000 for recommending educational content may determine recommended content based on expected correct answer information and/or capability information of the target user. The recommended content may be a content set having the highest expected score related to capability information calculated based on expected percent correct information and/or capability information. Here, the expected score may be calculated by associating the target user's ability information with the log data of a specific point included in the target user's learning data.

For example, when first ability information related to the target user's maximum learning ability is acquired, the educational contents recommendation apparatus 1000 determines the target user's expected percent correct answer information or the first ability information of learning data associated with a specific point. An expected score related to the first capability information of the target user may be calculated using log data, and the first content set having the highest expected score may be determined as the recommended content.

As another example, when the second ability information related to the reasoning ability of the target user is obtained, the educational content recommendation apparatus 1000 may include log data of learning data linked to a specific point in which the target user's expected percent correct information or the second ability information is linked to a specific point. An expected score related to the second capability information of the target user may be calculated, and a second content set having the highest expected score for reasoning capability may be determined as the recommended content.

As another example, when Nth ability information (eg, logic, application ability, etc.) of the target user is obtained, the educational content recommendation apparatus 1000 determines whether the expected correct answer rate information or the Nth ability information of the target user is linked to a specific point. An expected score related to the Nth capability information of the target user may be calculated using log data of the learning data, and an Nth content set corresponding to the highest calculated expected score may be determined as recommended content.

In order to implement the above, the step of acquiring recommended content according to an embodiment (S5000) is the step of acquiring an educational content set, when the content included in the educational content set is provided to the target user. It may include calculating a change in the capability information of and determining content having a maximum change in the capability information as the recommended content.

Meanwhile, the educational content recommendation apparatus 1000 according to an embodiment of the present application determines a neural network model based on capability information, distributes resources corresponding to the determined model, and obtains recommended content through the determined neural network model. It can be. Through these operations, the apparatus 1000 for recommending educational content can provide the user with educational content that maximizes the improvement of the target user's ability by properly distributing resources required for selecting the educational content according to the ability information of the target user. . Alternatively, the apparatus 1000 for recommending educational contents may ensure fairness of education by properly distributing resources required for selecting educational contents according to ability information of a target user.

Referring back to FIG. 4 , although not shown, the method for recommending educational content according to an embodiment of the present application may further include transmitting recommended content. For example, the educational content recommendation device 1000 may transmit recommended content to the user terminal 100 through the transceiver 1100 .

The apparatus 1000 for recommending educational contents according to an embodiment of the present application quantifies basic abilities related to various types of learning (eg, learning ability, reasoning ability, logical ability, application ability, basic strength, concentration, grit, etc.) Educational content that maximizes expected scores calculated for various types of abilities when the content is provided to the user may be determined as the recommended content. Through this, the apparatus 1000 for recommending educational contents according to an embodiment of the present application may provide the user with educational contents capable of training basic abilities related to learning, not simply test scores.

Various operations of the above-described educational content recommendation device 1000 may be stored in the memory 12000 of the educational content recommendation device 1000, and the controller 1300 of the educational content recommendation device 1000 may be stored in the memory 1200. Can be provided to perform actions.

The features, structures, effects, etc. described in the embodiments above are included in at least one embodiment of the present invention, and are not necessarily limited to only one embodiment. Furthermore, the features, structures, effects, etc. illustrated in each embodiment can be combined or modified with respect to other embodiments by those skilled in the art in the field to which the embodiments belong. Therefore, contents related to these combinations and variations should be construed as being included in the scope of the present invention.

In addition, although the embodiment has been described above, this is only an example and does not limit the present invention, and those skilled in the art to the present invention pertain to the above to the extent that does not deviate from the essential characteristics of the present embodiment. It will be appreciated that various modifications and applications not exemplified are possible. That is, each component specifically shown in the embodiment can be implemented by modifying it. And differences related to these modifications and applications should be construed as being included in the scope of the present invention as defined in the appended claims.

Claims

A method for determining recommended content by an apparatus for analyzing user learning data, the method comprising:

Obtaining learning data of a target user, wherein the learning data includes log data including problem data related to a problem previously solved by the target user and response data related to the target user's response to the problem- ;

obtaining a problem database including at least one candidate problem;

calculating expected correct answer information of the target user for the candidate problem based on the candidate problem and the learning data;

Obtaining ability information of the target user associated with at least some data of the log data based on the expected correct answer rate information - the ability information is first information representing the maximum learning ability of the target user, the target user's ability information related to at least one of second information representing reasoning ability, third information representing logic ability of the target user, and fourth information representing application ability of the target user; and

Determining recommended content based on the capability information of the target user; including,

Obtaining the ability information of the target user,

generating a growth curve related to an expected learning ability value of the target user based on the learning data and the expected percent correct information; and

Acquiring the first information based on the growth curve; including,

How to recommend educational content.
According to claim 1,

Obtaining the first information based on the growth curve,

obtaining change rate information of the growth curve;

obtaining target change rate information having a change rate equal to or smaller than a predetermined threshold change rate based on the change rate information; and

Acquiring the first information based on the expected learning capability corresponding to the target change rate information;

How to recommend educational content.
According to claim 1,

The problem database includes average correct answer information for the candidate problems,

Obtaining the ability information of the target user,

Comprising the step of obtaining the second information of the target user by comparing the average correct answer information for the candidate problem and the expected correct answer information of the target user for the candidate problem,

How to recommend educational content.
According to claim 3,

Obtaining the second information of the target user,

obtaining a standard candidate problem having an average correct rate equal to or lower than a predetermined standard correct rate; and

Comprising: calculating the second information of the target user by comparing the expected percent correct information of the target user for the reference candidate problem with the average percent correct answer information for the standard candidate problem;

How to recommend educational content.
According to claim 1,

The step of determining the recommended content,

obtaining a set of educational content;

calculating an expected change in the ability information of the target user when the content included in the educational content set is provided to the target user; and

Determining content having the maximum expected change in the capability information as the recommended content;

How to recommend educational content.
According to claim 1,

The step of determining the recommended content,

determining a neural network model based on the capability information;

distributing resources corresponding to the determined model; and

Acquiring the recommended content through the determined neural network model; including,

How to recommend educational content.
A computer-readable recording medium recording a program for executing the method according to any one of claims 1 to 6 in a computer.
An apparatus for determining recommended content by receiving user learning data from an external user terminal device,

a transceiver for communicating with the user terminal;

A controller configured to acquire learning data of a target user through the transceiver and calculate ability information of the target user based on the learning data;

The controller,

Acquiring learning data of a target user, the learning data including problem data related to a problem previously solved by the target user and log data including response data related to the target user's response to the problem; obtaining a problem database including at least one candidate problem, and calculating expected correct answer information of the target user for the candidate problem based on the candidate problem and the learning data; Capability information of the target user associated with at least some data of the log data based on the expected percent correct answer information - the capability information is first information indicating the maximum learning ability of the target user and inference ability of the target user Relating to at least one of second information, third information indicating the logical ability of the target user, and fourth information indicating the application ability of the target user; obtaining recommended content based on the ability information of the target user It is configured to

The controller is configured to generate a growth curve related to an expected learning ability value of the target user based on the learning data and the expected percent correct answer information, and obtain the first information based on the growth curve to obtain the ability information. ,

Educational content recommendation device.
According to claim 8,

The controller,

Obtain change rate information of the growth curve, obtain target change rate information having a change rate equal to or smaller than a predetermined threshold change rate based on the change rate information, and obtain the expected learning corresponding to the target change rate information configured to obtain the first information based on a capability value;

Educational content recommendation device.
According to claim 8,

The problem database includes average correct answer information for the candidate problems,

The controller,

Acquiring the capability information by obtaining the second information of the target user by comparing the average percent correct information for the candidate problem and the expected percent correct answer information of the target user for the candidate problem,

Educational content recommendation device.
According to claim 10,

The controller,

A standard candidate problem having an average percent correct answer equal to or lower than a predetermined standard percent correct answer is obtained, and the expected percent correct information of the target user for the standard candidate problem is compared with the average percent correct answer information for the standard candidate problem, and the target user configured to calculate the second information of

Educational content recommendation device.
According to claim 8,

The controller,

An educational content set is acquired, and when the content included in the educational content set is provided to the target user, an expected change in the ability information of the target user is calculated, and the content for which the expected change in the capability information is the largest is recommended. configured to determine by content;

Educational content recommendation device.
According to claim 8,

The controller,

Determining a neural network model based on the capability information, distributing resources corresponding to the determined model, and obtaining the recommended content through the determined neural network model,

Educational content recommendation device.