KR102546207B1 - Method and device for providing ai-based online learning using wearable devices - Google Patents

Abstract

An AI-based online learning providing device and method using a wearable device are disclosed. An AI-based online learning providing apparatus using a wearable device includes: an information management unit that receives user information from a user and registers the user based on the received user information; a bio-signal detector for sensing bio-information including a PPG signal and a GSR signal; a motion detector for sensing motion information including motion acceleration and direction; an EEG prediction unit that predicts an EEG state of the user based on the biometric information and the motion information; a learning state determining unit for deriving a degree of concentration and fatigue from the predicted brain wave state, and determining a learning state of the user with respect to learning content based on the derived degree of concentration and the degree of fatigue; and a customized environment and customized content providing unit that determines a customized learning environment and customized learning contents according to the determined learning state and the user information, and provides the determined customized learning environment and customized learning contents to the user.

Description

Apparatus and method for providing AI-based online learning using wearable devices {METHOD AND DEVICE FOR PROVIDING AI-BASED ONLINE LEARNING USING WEARABLE DEVICES}

The present invention relates to a technology for providing AI-based online learning using a wearable device.

Unless otherwise indicated herein, material described in this section is not prior art to the claims in this application, and inclusion in this section is not an admission that it is prior art.

Online learning solutions and platforms using existing artificial intelligence are services that analyze learners' learning results using artificial intelligence technology and provide personalized and customized curriculum and contents necessary for learners. Only the information that can be collected in the online learning process, such as 'problem solving time', 'types of incorrect answers', and 'common points of incorrect answers', is used for analysis.

The need for a biosignal-based artificial intelligence online learning solution service that accurately predicts the learner's actual learning ability by using not only the information that can be collected in the online learning process but also the learner's biometric information and wrist movement information has emerged.

An object of the present invention to solve the above problems is to provide an AI-based online learning providing apparatus and method using a wearable device.

One aspect of the present invention for achieving the above object provides an AI-based online learning providing device using a wearable device.

An AI-based online learning providing apparatus using a wearable device includes: an information management unit that receives user information from a user and registers the user based on the received user information; a bio-signal detector for sensing bio-information including a PPG signal and a GSR signal; a motion detector for sensing motion information including motion acceleration and direction; an EEG prediction unit that predicts an EEG state of the user based on the biometric information and the motion information; a learning state determining unit for deriving a degree of concentration and fatigue from the predicted brain wave state, and determining a learning state of the user with respect to learning content based on the derived degree of concentration and the degree of fatigue; and a customized environment and customized content providing unit that determines a customized learning environment and customized learning contents according to the determined learning state and the user information, and provides the determined customized learning environment and customized learning contents to the user.

The EEG prediction unit may predict the EEG state of the user by inputting the biometric information and the motion information to a pre-supervised EEG prediction model.

The brain wave state may include magnitudes for each of the SMR wave, mid-beta wave, theta wave, and alpha wave, which are obtained as numerical integration values for each of predefined frequency bands in an electroencephalogram (EEG) signal.

The brain wave prediction model may be trained in advance using learning data having a first feature vector obtained by converting biometric information and motion information as an input value and a second feature vector obtained by converting the brain wave state as an output value. .

The learning state determiner may determine the user's learning state for the learning content as one of four learning states of concentration, distraction, cognitive stress, and immersion based on the degree of concentration and fatigue.

The present invention has the effect of predicting the brain wave state through the learner's biometric information and wrist movement information, and calculating the degree of concentration and fatigue of the learner according to the predicted brain wave state.

In addition, the present invention can maximize the learning effectiveness and efficiency of the learner by providing a customized learning environment and customized learning contents suitable for the learner according to the calculated degree of concentration and fatigue of the learner.

Effects obtainable from the embodiments are not limited to the effects mentioned above, and other effects not mentioned are clearly derived and understood by those skilled in the art based on the detailed description below. It can be.

BRIEF DESCRIPTION OF THE DRAWINGS Included as part of the detailed description to aid understanding of the embodiments, the accompanying drawings provide various embodiments and, together with the detailed description, describe technical features of the various embodiments.
1 is a schematic diagram of an AI-based online providing method and apparatus using a wearable device according to an embodiment of the present invention.
FIG. 2 is a block diagram showing functional modules of the wearable device according to FIG. 1 as an example.
3 is a diagram for explaining an EEG prediction model according to an embodiment.
FIG. 4 is a conceptual diagram for explaining the structure and operation of an artificial neural network used in the wearable device (eg, brain wave prediction model) according to FIG. 1 .
5 is a diagram showing a hardware configuration of a wearable device according to an embodiment of the present invention by way of example.

Since the present invention can make various changes and have various embodiments, specific embodiments will be illustrated in the drawings and described in detail in the detailed description. However, this is not intended to limit the present invention to specific embodiments, and should be understood to include all modifications, equivalents, and substitutes included in the spirit and scope of the present invention. Like reference numerals have been used for like elements throughout the description of each figure.

Terms such as first, second, A, and B may be used to describe various components, but the components should not be limited by the terms. These terms are only used for the purpose of distinguishing one component from another. For example, a first element may be termed a second element, and similarly, a second element may be termed a first element, without departing from the scope of the present invention. The terms and/or include any combination of a plurality of related recited items or any of a plurality of related recited items.

It is understood that when an element is referred to as being "connected" or "connected" to another element, it may be directly connected or connected to the other element, but other elements may exist in the middle. It should be. On the other hand, when an element is referred to as “directly connected” or “directly connected” to another element, it should be understood that no other element exists in the middle.

Terms used in this application are only used to describe specific embodiments, and are not intended to limit the present invention. Singular expressions include plural expressions unless the context clearly dictates otherwise. In this application, the terms "include" or "have" are intended to designate that there is a feature, number, step, operation, component, part, or combination thereof described in the specification, but one or more other features It should be understood that it does not preclude the possibility of addition or existence of numbers, steps, operations, components, parts, or combinations thereof.

Unless defined otherwise, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which the present invention belongs. Terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with the meaning in the context of the related art, and unless explicitly defined in the present application, they should not be interpreted in an ideal or excessively formal meaning. don't

Hereinafter, preferred embodiments according to the present invention will be described in detail with reference to the accompanying drawings.

1 is a schematic diagram of an AI-based online providing method and apparatus using a wearable device according to an embodiment of the present invention.

Referring to FIG. 1, an AI-based online providing device using a wearable device (hereinafter, 'wearable device', 100) detects user's biometric information and motion information, and provides information to the user based on the detected biometric information and motion information. Customized learning content and customized learning environment can be provided.

The wearable device 100 can store the artificial neural network 10 in internal storage and drive the artificial neural network 10, but communicates with an independent server that drives the artificial neural network 10, thereby providing information to the artificial neural network 10. Biometric information and motion information may be input, and a prediction result of an brain wave state may be received as an output of the artificial neural network 10 .

In addition, the wearable device 100 may perform supervised learning on the artificial neural network 10 in advance by using pre-obtained learning data including biometric information, motion information, and brain wave states.

FIG. 2 is a block diagram showing functional modules of the wearable device according to FIG. 1 as an example.

Referring to FIG. 2 , the wearable device 100 includes an information management unit 101, a bio-signal detection unit 102, a motion detection unit 103, an EEG prediction unit 104, a learning state determination unit 105, an environment and a content providing unit 106 .

The information management unit 101 may receive user information from a user and register the user based on the received user information. At this time, the user is a user (learner) who wants to use the wearable device 100 .

The biosignal detector 102 performs a function of detecting user's biometric information. For example, biometric information includes PPG (pulse wave) and GSR (skin electrical conductivity).

In addition, the biosignal detector 102 may detect biometric information in a user's learning situation. At this time, the biosignal detector 102 may start to detect biometric information according to the user's input signal (signal to start learning), and to detect biometric information according to the user's input signal (signal to end learning). detection can be terminated.

The bio-signal detector 102 may include a light emitting unit (a device that emits light) and a light receiving unit (for example, a photodiode or photosensor) in order to detect a PPG (pulse wave). The light reflected from blood streams in the body is detected by the light receiver and converted into an electrical signal, and the electrical signal can be output as a pulse wave. In particular, the PPI (Peak to Peak Interval) of the PPG signal may indicate an interval that changes according to the degree of stress. In a state of stress, the heart rate increases, so the PPI decreases, and in a state of relaxation, the heart rate decreases, so the PPI increases. . That is, the degree of stress can be evaluated by changing the PPI that changes according to the stress stimulus.

The biosignal detector 102 may include a current electrode for flowing a microcurrent through the skin to detect GSR (skin electrical conductivity) and a voltage electrode for detecting a skin voltage in response to the microcurrent, wherein the current electrode Skin electrical conductivity can be detected by flowing a microcurrent to the skin and measuring the skin voltage at which the voltage electrode reacts to the microcurrent.

The motion detection unit 103 performs a function of detecting user motion information. For example, the motion information may include motion acceleration and direction of the wrist when the wearable device is worn on the user's wrist. The motion sensor 103 may include a gyro sensor and an acceleration sensor to detect user motion information.

The motion sensor 103 may include a gyro sensor that measures an angular velocity, and may obtain a value of an angle rotated by the wearable device 100 for a unit time based on one axis. The angular velocity means an angle that rotates per unit time, and can be measured by a conventional method. In addition, since the gyro sensor may have different data values depending on a change in temperature, a temperature sensor may be further included to check the change in temperature and correct the data value, if necessary.

The motion sensor 103 includes an acceleration sensor that measures linear acceleration, and when the wearable device 100 moves in 3D, it can measure acceleration in the x-axis, y-axis, and z-axis directions.

The motion detection unit 103 may obtain motion information including motion acceleration and direction by analyzing data values detected by the gyro sensor and the acceleration sensor.

Also, the motion detection unit 103 may detect motion information in a user's learning situation. At this time, the motion detection unit 103 may start sensing motion information according to the user's input signal (signal to start learning), and detect motion information according to the user's input signal (signal to end learning). can be terminated. At this time, the user's input signal may be a signal sent from the server in response to the user's act of logging in to the online AI math education solution or platform operated by the server.

The brain wave prediction unit 104 performs a function of predicting a user's brain wave state based on biometric information and motion information.

The brain wave prediction unit 104 predicts the brain wave state by using the bio information received from the bio signal sensor 102 and the motion information received from the motion sensor 103. At this time, the brain wave state is the magnitude (or intensity) of each of the SMR wave, Mid-beta wave, theta wave, and alpha wave, which is obtained as a numerical integration value for each of the predefined frequency bands in the electroencephalogram (EEG). include

In relation to this, SMR waves (Sensory Motor Rhythm) have a frequency band of 12 to 15 Hz and are brain waves that appear when concentration on external information is strengthened in a state of attention. In addition, Mid beta waves have a frequency band of 16 to 20 Hz and are brain waves that appear when performing tasks in an awake state. In other words, mid beta waves become dominant when performing thinking activities with relatively high mental load while concentrating on one subject, such as calculation or mental arithmetic. In addition, theta waves have a frequency band of 4 to 8 Hz and are brain waves that appear when you are sleepy or unconscious. In addition, alpha waves have a frequency band of 8 to 13 Hz and have a very close relationship with the visual area, and an increase in the size of alpha waves can be interpreted as visual attention function and cognitive decline.

In this regard, in order to quantitatively determine how much weight each frequency component occupies in the predicted EEG state, power spectrum analysis may be used. A power spectrum is a Fourier transform of an autocorrelation function, and in general, an EEG signal is subjected to short-term Fourier transform (STFT) or wavelet transform. It refers to the power expressed in log scale (unit: dB).

In one embodiment, the brain wave prediction unit 104 may predict the state of the brain wave using the pre-trained brain wave prediction model 104b. That is, the brain wave prediction unit 104 may predict the user's brain wave state by inputting biometric information and motion information to the pre-supervised brain wave prediction model 104b.

3 is a diagram for explaining an EEG prediction model according to an embodiment.

Referring to FIG. 3 , the EEG prediction unit 104 may include an EEG prediction model learning unit 104a and an EEG prediction model 104b. The EEG prediction model learning unit 104a and the EEG prediction model 104b are configured according to the functions of the EEG prediction unit 104, and it is obvious that the EEG prediction unit 104 can implement all of the corresponding functions.

The EEG prediction model learning unit 104a may pre-train the EEG prediction model 104b by using pre-obtained biometric information, motion information, and EEG states detected during learning by a plurality of users as learning data. . An artificial neural network may be used as an EEG prediction model 104b according to an embodiment. An artificial neural network is a predictive model implemented in software or hardware that mimics the computational power of a biological system using a large number of artificial neurons (or nodes).

In this case, the biometric information, motion information, and brainwave state may be obtained in advance. They are obtained by input from a manager, or the wearable device 100 further includes an brainwave detector to transmit the brainwave state according to the biometric information and the motion information to the wearable device ( 100) can also be obtained directly.

The EEG prediction model learning unit 104a uses a first feature vector obtained by converting the biometric information and motion information as an input value so that the EEG prediction model 104b can receive the biometric information and motion information and predict the EEG state. , the EEG prediction model 104b may be supervised using learning data having the second feature vector obtained by converting the EEG state as an output value.

At this time, supervised learning means learning to find an output value according to a given input value by using data with input values and corresponding output values as learning data, and means learning that takes place in a state where the correct answer is known. The set of input values and output values given in supervised learning is called training data. That is, the first feature vector obtained by converting the aforementioned biometric information and motion information and the second feature vector obtained by converting the EEG state are input values and output values, respectively, and can be used as learning data for supervised learning of the EEG prediction model 104b. there is.

FIG. 4 is a conceptual diagram for explaining the structure and operation of an artificial neural network used in the wearable device (eg, brain wave prediction model) according to FIG. 1 .

Referring to FIG. 4 , the artificial neural network 10 may include an input layer 11, a hidden layer 12, and an output layer 13.

In one embodiment, the brain wave prediction model 104b receives an input value, and the input layer 10 having nodes corresponding to the number of components of the first feature vector and the connection strength (for each output value of the input layer 10) or weights), one or more hidden layers (12) for outputting after adding a bias; and an output layer 13 that multiplies each output value of the hidden layer 12 by a connection strength (or weight) and outputs the result using an activation function. For example, the activation function may be a LeRU function or a Softmax function, but is not limited thereto. Connection strength and bias can be continuously updated by supervised learning.

Specifically, the EEG prediction model 104b may be supervised so that an output value of a loss function according to a given input value (first feature vector) and output value (second feature vector) is minimized. For example, the loss function H(Y,Y′) may be defined as in Equation 1 below.

In Equation 1, Ym may be the m-th component of the second feature vector, and Y'm may be the m-th component of the output vector that is output by receiving the first feature vector from the EEG prediction model 104b.

In summary, the brain wave prediction unit 104 converts the biometric information and motion information of the user into input vectors, inputs them to the pre-supervised brain wave prediction model 104b, converts the output vector obtained as the output of the brain wave prediction model, An EEG state of the user may be predicted. At this time, the brain wave state is the magnitude (or intensity) of each of the SMR wave, Mid-beta wave, theta wave, and alpha wave, which is obtained as a numerical integration value for each of the predefined frequency bands in the electroencephalogram (EEG). include

The learning state determination unit 105 performs a function of determining a learning state including concentration and fatigue based on the predicted EEG state. In this case, the degree of concentration means the degree to which the learner maintains the learning ability, and the degree of fatigue means the degree to which the learner's learning ability is reduced.

The learning state determining unit 105 may obtain learning content provided to the user when sensing biometric information and motion information, and determine the learning state of the provided learning content based on the acquired learning content and the predicted EEG state. . In this case, the learning content provided to the user may be content provided to the user by the wearable device, or content learned by the user and input to the wearable device. In addition, the learning content provided to the user may be content provided to the user terminal by providing the learning content to the user terminal in an online AI learning solution or platform operated by the server.

The learning state determination unit 105 may derive concentration and fatigue from the predicted brain wave state, and determine the user's learning state for the learning content based on the derived concentration and fatigue.

In this regard, the learning state determiner 105 may determine the degree of concentration for the learning content based on the predicted EEG state including the magnitude of the SMR wave, the magnitude of the mid beta wave, and the magnitude of the theta wave. At this time, the degree of concentration is determined by Equation 2 below.

In addition, the learning state determining unit 105 may determine the degree of fatigue with respect to the learning content based on the predicted EEG state including the magnitude of theta wave and the magnitude of the alpha wave. At this time, it is known that the degree of fatigue is highly related to theta and alpha waves, but the higher the degree of fatigue, the higher the possibility that the degree of concentration is measured relatively low. Therefore, in the present invention, the correlation coefficient representing the correlation between the degree of concentration and the alpha and theta waves is calculated. and define the degree of fatigue based on the calculated correlation coefficient.

A correlation coefficient representing a correlation between alpha waves and theta waves and concentration according to Equation 2 is calculated as shown in Table 1 below.

number of measurements Alpha (8-13 Hz) Theta (4-8Hz) One 0.84 2.89 2 0.82 2.76 3 0.84 2.87 4 0.83 2.90 5 0.86 3.26

Referring to Table 1, the correlation coefficient between alpha waves and concentration can be determined as 0.84, which is the average value for 5 measurement results, and the correlation coefficient between theta waves and concentration can be determined as 2.94, which is the average value for 5 measurement results. Based on the correlation coefficient determined by the above table, the degree of fatigue can be calculated according to Equation 3 below.

In one embodiment, the learning state determiner 105 determines the user's learning state for the learning content according to the degree of concentration and fatigue for the learning content derived from the predicted brain wave state into four learning states: concentration, distraction, cognitive stress, and immersion. One of the states can be determined as a learning state. At this time, immersion means a state of high concentration, a state of being deeply immersed in learning.

Specifically, the learning state determining unit 105 may determine the user's learning state for the learning content as 'immersion' when the degree of concentration is greater than a preset first value and the degree of fatigue is greater than a preset second value, When the degree of concentration is greater than a preset first value and the degree of fatigue is less than a preset third value, the user's learning state for the learning content may be determined as 'concentration', and the degree of concentration is less than a preset fourth value and , When the degree of fatigue is greater than a preset second value, the user's learning state for the learning content may be determined as 'cognitive stress', the degree of concentration is less than the preset fourth value, and the fatigue degree is a preset third value. If the value is less than , the user's learning state for the learning content may be determined as 'distracted'. At this time, it is preferable that the first value is set higher than the fourth value, and the second value is set higher than the third value.

That is, in relation to the learning state of the user with respect to the learning content, the learning state determining unit 105 determines distraction when both concentration and fatigue are low, concentration when concentration is high and fatigue is low, and cognitive stress when concentration is low and fatigue is high. Therefore, if the degree of concentration is high and the degree of fatigue is high, it can be determined as immersion.

The customized environment and customized content providing unit 106 performs a function of determining a customized learning environment and customized learning contents according to the determined learning state and user information, and providing at least one of the determined customized learning environment and customized learning contents to the user. .

The customized environment and customized content providing unit 106 provides various learning contents to a plurality of users, predicts the brain wave state of each of the plurality of users for each of the provided learning contents, and the learning state derived from the predicted brain wave state. You can create a database in advance that collects them.

That is, since there are various learning states derived from learning content for each user, the customized environment and customized content providing unit 106 may store the learning content by mapping the learning state for each learning content based on user information in the database.

The customized environment and customized content provider 106 may derive users similar to the user from a pre-generated database, determine customized learning content based on the similar user's learning history, and provide the determined customized learning content to the user. . For example, the learning history is information related to the learning contents that the user has learned in the past, and includes learning process, learning questions, question difficulty, solving time, and the like.

For example, when the user's learning state is determined to be 'distraction', 'cognitive stress', or 'immersion', the customized environment and customized content providing unit 106 performs the learning that a user similar to the user has learned in a pre-generated database. Among contents, learning contents mapped to a learning state of 'concentration' may be determined as customized learning contents.

In this case, a method of deriving a user similar to the user may be performed by calculating a degree of similarity between vectors expressing user information.

For example, the user information is information indicating the characteristics of the user in order to distinguish the user from other users, and may include various information such as age, gender, previous learning history, and intelligence quotient (IQ).

The degree of similarity can be determined by Equation 4.

Here, S _A,B denote similarities between user A and user B, A denotes a vector representing information about user A, and B denotes a vector representing information about user B. When the calculated S _{A and B} are higher than preset values, it can be determined that A and B are similar.

In addition, the customized learning environment and customized learning contents provided to the user by the customized environment and customized content providing unit 106 may exist in various ways.

In one embodiment, the customized environment and customized content providing unit 106 may provide visual learning content that can increase concentration to the user when the user's learning state for the learning content is determined to be 'distracted'. .

In another embodiment, the customized environment and customized content providing unit 106 may provide the user with learning content having a higher level of difficulty when it is determined that the user's learning state for the learning content is 'concentrated'.

In another embodiment, the customized environment and customized content providing unit 106, when the user's learning status for the learning content is determined to be 'cognitive stress', provides learning content with a level of difficulty lowered by one level to reduce fatigue to the user. can provide

In another embodiment, the customized environment and customized content providing unit 106, when the user's learning status for the learning content is determined to be 'immersion', information instructing the user to readjust the learning environment so that the user can take a break. can provide. For example, the information about the learning environment suitable for the user may include information about adjusting the brightness of lighting to a specific brightness in order to improve concentration or information about generating white noise.

5 is a diagram showing a hardware configuration of a wearable device according to an embodiment of the present invention by way of example.

Referring to FIG. 5 , the wearable device 100 includes at least one processor 110; and a memory for storing instructions instructing the at least one processor 110 to perform at least one operation.

The at least one operation may include the operation of the wearable device 100 described with reference to FIGS. 1 to 4 .

Here, the at least one processor 110 may mean a central processing unit (CPU), a graphics processing unit (GPU), or a dedicated processor for performing methods according to embodiments of the present invention. can

The memory 120 may include at least one of volatile storage media and non-volatile storage media. For example, the memory 120 may include at least one of a read only memory (ROM) and a random access memory (RAM).

The storage device 160 may be, for example, a hard disk drive (HDD) or a solid state drive (SSD).

Also, the wearable device 100 may include a transceiver 130 that performs communication through a wireless network. In addition, the wearable device 100 may further include an input interface device 140 , an output interface device 150 , a storage device 160 , and the like. Each component included in the wearable device 100 may be connected by a bus 170 to communicate with each other.

For example, the wearable device 100 includes a communicable desktop computer, a laptop computer, a notebook, a smart phone, a tablet PC, and a mobile phone. ), smart watch, smart glass, e-book reader, portable multimedia player (PMP), portable game device, navigation device, digital camera, digital multimedia broadcasting (DMB) player, digital audio recorder, digital audio player, digital video recorder, digital video player, personal digital assistant (PDA), digital pen ) and the like.

The methods according to the present invention may be implemented in the form of program instructions that can be executed by various computer means and recorded on a computer readable medium. Computer readable media may include program instructions, data files, data structures, etc. alone or in combination. Program instructions recorded on a computer readable medium may be specially designed and configured for the present invention or may be known and usable to those skilled in computer software.

Examples of computer readable media may include hardware devices specially configured to store and execute program instructions, such as ROM, RAM, flash memory, and the like. Examples of program instructions may include not only machine language codes generated by a compiler but also high-level language codes that can be executed by a computer using an interpreter and the like. The hardware device described above may be configured to operate with at least one software module to perform the operations of the present invention, and vice versa.

In addition, the above-described method or device may be implemented by combining all or some of its components or functions, or may be implemented separately.

Although the above has been described with reference to preferred embodiments of the present invention, those skilled in the art will variously modify and change the present invention within the scope not departing from the spirit and scope of the present invention described in the claims below. You will understand that it can be done.

100: AI-based online learning providing device using a wearable device
101: information management unit 102: biosignal detection unit
103: motion sensor 104: EEG prediction unit
105: learning state determining unit 106: customized environment and customized content providing unit

Claims (5)

As an AI-based online learning providing device using a wearable device,
an information management unit that receives user information from a user and registers the user based on the received user information;
a bio-signal detector for sensing bio-information including a PPG signal and a GSR signal;
a motion detector for sensing motion information including motion acceleration and direction;
an EEG prediction unit that predicts an EEG state of the user based on the biometric information and the motion information;
a learning state determining unit for deriving a degree of concentration and fatigue from the predicted brain wave state, and determining a learning state of the user with respect to learning content based on the derived degree of concentration and the degree of fatigue; and
A customized environment and customized content providing unit determining a customized learning environment and customized learning contents according to the determined learning state and the user information, and providing the determined customized learning environment and customized learning contents to the user,
The brain wave prediction unit,
Predicting the user's brain wave state by inputting the biometric information and the motion information to a supervised learning EEG prediction model,
The brain wave state,
Including the magnitude of each of the SMR wave, Mid-beta wave, theta wave, and alpha wave, obtained as a numerical integral value for each of the predefined frequency bands in the electroencephalogram (EEG),
The brain wave prediction model,
Supervised learning in advance using learning data having a first feature vector obtained by converting biometric information and motion information as an input value and a second feature vector obtained by converting the brain wave state as an output value,
The learning state determination unit determines the concentration based on Equation 1 below,
[Equation 1]

In Equation 1, the magnitude of the SMR wave, the magnitude of the Mid beta wave, and the magnitude of the theta wave are values included in the EEG state predicted by the EEG prediction model,
The learning state determination unit determines the degree of fatigue based on Equation 2 below,
[Equation 2]

In Equation 2, the magnitude of the theta wave and the magnitude of the alpha wave are values included in the EEG state predicted by the EEG prediction model, and the correlation coefficient between the theta wave and the concentration and the correlation coefficient between the alpha wave and the concentration are An AI-based online learning providing device using a wearable device, which is determined by an experimentally calculated value of a correlation coefficient between the degree of concentration according to Equation 1 and theta and alpha waves included in the brain wave state. delete delete delete In claim 1,
The learning state determination unit,
Based on the degree of concentration and fatigue, the user's learning state for the learning content is determined as one of four learning states of concentration, distraction, cognitive stress, and immersion. An AI-based online learning providing device using a wearable device.

Images