Personalized learning recommendation method and system based on learning attitude of learner

Abstract

Disclosed are a personalized learning recommendation method and system based on a learning attitude of a learner. The personalized learning recommendation method comprises the steps of: determining learning attitude information indicating a learning attitude of a learner by inputting, to a learner analysis model, real-time information collected by measuring the learner who performs learning in real time; collecting student information and education-related information about the learner to match the collected pieces of information with the learning attitude information; determining a comprehensive score of the learner on the basis of the student information about the learner, the education-related information about the learner, and the learning attitude information; and determining a recommended learning strategy for the learner on the basis of the comprehensive score, wherein the student information may include subject scores of the learner determined by a teacher according to answers of the learner and an evaluation of the learner created by the teacher, and the education-related information may include history of learner's educational content and learning activity information about the learner.

Description

A personalized learning recommendation method and system based on the learner's learning attitude.

[0001] The present invention relates to a method and device for recommending personalized learning, and more specifically, to a method and device for recommending personalized learning by recognizing and analyzing learners' learning attitudes in real time.

[0002] In traditional education, teachers typically directly observe and manage students' attitudes during class. This approach relies on the teacher's subjective judgment and presents challenges in simultaneously monitoring the learning attitudes of all students.

[0003] Furthermore, conventional personalized learning recommendation methods assess a student's learning level based on their responses to questions and then recommend courses based on that level. However, because students may have different learning levels and attitudes toward a given subject based on their interests, conventional personalized learning recommendation methods that solely assess learning level are unable to identify and recommend courses that are of current interest to the student.

[0004] In addition, it is difficult to detect changes in learners' attitudes during the learning process using conventional personalized learning recommendation methods.

[0005] Therefore, there is a need for a method that analyzes learners' interests and aptitudes to recommend optimal learning strategies and content, and to immediately identify students' unfocused states.

[0006] The present invention provides a method and device for recognizing and analyzing students' learning attitudes in real time, analyzing the learner's interests and aptitudes based on the learner's current learning attitudes and past learning information, and recommending learning strategies and content optimized for the learner.

[0007] The present invention provides a method and device for providing an online / offline learning data analysis function through a dashboard and a communication window between educational entities (teachers, learners, parents).

[0008] The present invention provides a method and device for supporting more accurate and rich personalized learning by analyzing learners' interests and aptitudes based on offline class attitude records and life records and recommending career paths and fields of study.

[0009] The present invention provides a method and device that outputs a notification to a teacher when a student becomes unfocused during class, thereby enabling the teacher to immediately identify and take action on the student's unfocused state.

[0010] The present invention provides a method and device for guiding learners' learning attitudes in a more accurate and detailed direction than the current level by reflecting the learning attitudes in an offline environment where actual education takes place.

[0011] A personalized learning recommendation method according to one embodiment of the present invention includes the steps of: determining learning attitude information representing a learning attitude of a learner by inputting real-time information collected by measuring a learner performing learning in real time into a learner analysis model; collecting student information and education-related information of the learner and matching the collected information with the learning attitude information; determining a comprehensive score of the learner based on the student information, the education-related information of the learner, and the learning attitude information; and determining a recommended learning strategy of the learner based on the comprehensive score. The student information may include the subject grades of the learner determined by a teacher based on the learner's answers and an evaluation of the learner written by a teacher, and the education-related information may include the learner's educational content history and learning activity information.

[0012] A personalized learning recommendation method according to one embodiment of the present invention may further include a step of accumulating learning attitude information for a preset period of time; and a step of analyzing the time and situation in which the learner's learning concentration changes based on the accumulated learning attitude information to determine the learner's learning tendency.

[0013] A personalized learning recommendation method according to one embodiment of the present invention includes a step of analyzing the learning motivation, concentration, and stress cause of the learner based on the learning attitude information and proposing a learning strategy corresponding to the learner, wherein the real-time information includes at least one of image information photographing the learner, audio information recording the voice of the learner, text information input by the learner, and biometric information of the learner, and the learner analysis model may include at least one of an image analysis model analyzing the learning concentration and emotional state of the learner based on the face and posture of the learner included in the image information, an audio analysis model analyzing the learning concentration of the learner based on the voice tone and speed of the learner included in the audio information, a text analysis model analyzing the learning concentration and learning performance of the learner based on the handwritten content of the learner included in the text information or an answer written by the learner, and a biometric analysis model analyzing the learning concentration and fatigue of the learner based on the biometric information.

[0014] The step of determining the comprehensive score of the personalized learning recommendation method according to one embodiment of the present invention may include the step of applying preset weights for each type of information included in the real-time information to the video information, the audio information, the text information, and the biometric information and then adding them to determine an online score; the step of defining the learner's academic performance as an offline score; and the step of applying weights to each of the online score and the offline score to determine the comprehensive score.

[0015] The step of determining the recommended learning strategy of the personalized learning recommendation method according to one embodiment of the present invention may include the step of determining the learning effect of the learner for online classes and the learning effect of the learner for offline classes based on the online score and the offline score; and the step of determining a learning strategy that adjusts the proportion of online classes and the proportion of offline classes for the learner based on the determined learning effects of the learner.

[0016] A personalized learning recommendation method according to one embodiment of the present invention may further include a step of determining whether the learning attitude information is below a first threshold, which is a criterion for determining that the learning attitude requires improvement; and a step of determining feedback for increasing the learning concentration of the learner based on the learning attitude information if the learning attitude information is below the first threshold.

[0017] The step of determining the feedback of the personalized learning recommendation method according to one embodiment of the present invention may estimate the emotional state and fatigue of the learner based on the learning attitude information, and may adjust the learning schedule including the interval or length of the break time according to the emotional state and fatigue of the learner, or determine feedback suggesting a change in the learning environment.

[0018] A personalized learning recommendation method according to one embodiment of the present invention may further include a step of determining whether the learning attitude information exceeds a second threshold, which is a criterion for determining that the learning attitude is superior to that of other learners; and a step of determining feedback requesting the teacher to praise the learner if the learning attitude information exceeds the second threshold.

[0019] The step of determining the recommended learning strategy of the personalized learning recommendation method according to one embodiment of the present invention may include the step of recommending an in-depth learning path or a basic review path for each subject based on the comprehensive score; the step of determining the learning style of the learner based on the comprehensive score, the evaluation of the learner, and the education-related information; and the step of selecting learning content based on the learning style and the recommended path and recommending the learning content to the learner.

[0020] According to one embodiment of the present invention, a personalized learning recommendation system includes: a learner analysis device that measures a learner performing learning in real time, collects real-time information, and inputs the collected real-time information into a learner analysis model to determine learning attitude information representing the learner's learning attitude; and an integrated education device that collects student information and education-related information of the learner, matches the collected information with the learning attitude information, determines a comprehensive score of the learner based on the student information, the education-related information of the learner, and the learning attitude information, and determines a recommended learning strategy of the learner based on the comprehensive score, wherein the student information may include the learner's subject grades determined by a teacher based on the learner's answers and an evaluation of the learner written by a teacher, and the education-related information may include the learner's educational content history and learning activity information.

[0021] The real-time information of the personalized learning recommendation system according to one embodiment of the present invention includes at least one of image information photographing the learner, audio information recording the learner's voice, text information input by the learner, and biometric information of the learner, and the learner analysis model may include at least one of an image analysis model analyzing the learner's learning concentration and emotional state based on the learner's face and posture included in the image information, an audio analysis model analyzing the learner's learning concentration based on the learner's voice tone and speed included in the audio information, a text analysis model analyzing the learner's learning concentration and learning performance based on the learner's handwriting included in the text information or an answer written by the learner, and a biometric analysis model analyzing the learner's learning concentration and fatigue based on the biometric information.

[0022] The integrated education device of the personalized learning recommendation system according to one embodiment of the present invention may apply preset weights for each type of information included in the real-time information to the video information, the audio information, the text information, and the biometric information, and then add them up to determine an online score, define the learner's academic performance as an offline score, and apply weights to each of the online score and the offline score to determine the comprehensive score.

[0023] The integrated education device of the personalized learning recommendation system according to one embodiment of the present invention can determine the learning effect of the learner for online classes and the learning effect of the learner for offline classes based on the online score and the offline score, and can determine a learning strategy that adjusts the proportion of online classes and the proportion of offline classes for the learner based on the determined learning effects of the learner.

[0024] The learner analysis device of the personalized learning recommendation system according to one embodiment of the present invention may determine whether the learning attitude information is less than a first threshold value, which is a criterion for determining that the learning attitude requires improvement, and, if the learning attitude information is less than the first threshold value, estimate the emotional state and fatigue of the learner based on the learning attitude information, and adjust the learning schedule including the interval or length of the break time according to the emotional state and fatigue of the learner, or determine feedback suggesting a change in the learning environment.

[0025] The learner analysis device of the personalized learning recommendation system according to one embodiment of the present invention can determine whether the learning attitude information exceeds a second threshold, which is a standard for determining that the learning attitude is superior to that of other learners, and, if the learning attitude information exceeds the second threshold, determine feedback requesting the teacher to praise the learner.

[0026] The integrated education device of the personalized learning recommendation system according to one embodiment of the present invention may recommend an in-depth learning path or a basic review path for each subject based on the comprehensive score, determine the learning style of the learner based on the comprehensive score, an evaluation of the learner, and the education-related information, and select and recommend learning content to the learner based on the learning style and the recommended path.

[0027] According to one embodiment of the present invention, the learning attitudes of students can be recognized and analyzed in real time, and the interests and aptitudes of the learners can be analyzed based on the learners' current learning attitudes and past learning information, thereby recommending learning strategies and content optimized for the learners.

[0028] According to one embodiment of the present invention, an online / offline learning data analysis function through a dashboard and a communication channel between educational entities (teachers, learners, parents) can be provided.

[0029] According to one embodiment of the present invention, by analyzing learners' interests and aptitudes based on offline class attitude records and life records and recommending career paths and fields of study, more accurate and rich personalized learning can be supported.

[0030] According to one embodiment of the present invention, when a learner becomes unfocused during class, a notification is output to the teacher, allowing the teacher to immediately identify the student's unfocused state and take action.

[0031] According to one embodiment of the present invention, by reflecting the teaching attitude in an offline environment where actual education takes place, it is possible to induce the learning attitude of the learner in a more accurate and detailed direction than the current level.

[0032] FIG. 1 is a diagram illustrating a personalized learning recommendation system according to one embodiment of the present invention.

[0033] Figure 2 is an example of devices for collecting real-time information in one embodiment of the present invention.

[0034] FIG. 3 is a flowchart illustrating a personalized learning recommendation method of a learner analysis device according to one embodiment of the present invention.

[0035] FIG. 4 is a flowchart illustrating a personalized learning recommendation method of an integrated education device according to one embodiment of the present invention.

[0036] Hereinafter, embodiments are described in detail with reference to the attached drawings. However, the embodiments may be modified in various ways, and the scope of the patent application is not limited or restricted by these embodiments. It should be understood that all modifications, equivalents, or alternatives to the embodiments are included within the scope of the patent application.

[0037] The terms used in the examples are for illustrative purposes only and should not be construed as limiting. Singular expressions include plural expressions unless the context clearly dictates otherwise. In this specification, terms such as "comprise" or "have" are intended to indicate the presence of a feature, number, step, operation, component, part, or combination thereof described in the specification, but should be understood to not preclude the presence or addition of one or more other features, numbers, steps, operations, components, parts, or combinations thereof.

[0038] In addition, when describing with reference to the attached drawings, identical components will be assigned the same reference numerals regardless of the drawing numbers, and redundant descriptions thereof will be omitted. When describing embodiments, if a detailed description of a related known technology is judged to unnecessarily obscure the gist of the embodiment, the detailed description will be omitted.

[0039] Hereinafter, embodiments of the present invention will be described in detail with reference to the attached drawings.

[0040] FIG. 1 is a diagram illustrating a personalized learning recommendation system according to one embodiment of the present invention.

[0041] A personalized learning recommendation system according to an embodiment of the present invention may include a learner analysis model learning device (120), a learner analysis device (130), and an integrated education device (140), as illustrated in FIG. 1. The learner analysis model learning device (120), the learner analysis device (130), and the integrated education device (140) may be PCs or servers configured as separate hardware, as illustrated in FIG. 2. In addition, according to an embodiment, the learner analysis model learning device (120), the learner analysis device (130), and the integrated education device (140) may be included in one PC or server, and the learner analysis model learning device (120), the learner analysis device (130), and the integrated education device (140) may be different processors, or may be respective modules included in a program executed in one processor. In addition, the learner analysis device (130) and the integrated education device (140) may be included in one PC or server, and the learner analysis model learning device (120) may be configured as a separate PC or server.

[0042] A learner analysis model learning device (120) can receive learning information from a learning database (110) and learn a learner analysis model. The learning information may include at least one of image information created by photographing the faces and postures of each of a plurality of learners while they are participating in learning, audio information including voices spoken by each of a plurality of learners while they are participating in learning, handwritten content written by each of a plurality of learners while they are participating in learning, text information including answers written by each of a plurality of learners to questions included in learning, biometric information of each of a plurality of users participating in learning, and at least one of the learners' learning concentration, fatigue, and learning performance.

[0043] For example, learning information may include video information collected by filming various learning attitudes of multiple students during class at school.

[0044] Additionally, learning information may include the results of labeling collected video data by learning attitude. For example, each video data can be labeled as focused (seeteacher), unfocused (sleep, somethingelse), etc., and included in the learning data. If the learning falls into the lecture-style class category, a higher percentage of focused states may be judged as a higher level of learning concentration. For example, a focused state proportion of 80% or more may be judged as excellent, 70-79% as good, 49-69% as average, and 48% or less as poor (needs improvement).

[0045] The learner analysis model learning device (120) can learn a learner analysis model by applying augmentation techniques such as blurring and color change to learning information and then using the augmented learning information.

[0046] In addition, the learner analysis model may include at least one of a video analysis model, an audio analysis model, a text analysis model, and a biometric analysis model. For example, the video analysis model may be a CNN (Convolutional Neural Network) model, and the audio analysis model may be a RNN (Recurrent Neural Network) model. In addition, the text analysis model may be a Transformer model, and the biometric analysis model may be a DNN (Deep Neural Network) model.

[0047] The learner analysis model learning device (120) can learn to output the learner's learning concentration and emotional state corresponding to the learner's face and posture included in the input image information using the image information.

[0048] In addition, the learner analysis model learning device (120) can learn to output the learner's learning concentration corresponding to the learner's voice tone and speed included in the input audio information using audio information.

[0049] In addition, the learner analysis model learning device (120) can learn to output the learner's learning concentration and learning performance corresponding to the learner's handwriting or the answer written by the learner included in the input text information by using the text analysis model.

[0050] In addition, the learner analysis model learning device (120) can learn to output the learner's learning concentration and fatigue level corresponding to the biometric information received by the biometric analysis model using biometric information.

[0051] The learner analysis model learning device (120) can evaluate the performance of the learned learner analysis model, and if the performance of the learner analysis model exceeds a threshold, transmit the learner analysis model (101) to the learner analysis device (130). At this time, the learner analysis model learning device (120) can also transmit the learner analysis model (101) to the learner analysis device (130) using wired communication or wireless communication.

[0052] The learner analysis device (130) can collect real-time information (102) of a learner in real time using devices such as those illustrated in FIG. 2. For example, the real-time information (102) may include at least one of video information capturing the facial expression or posture of a learner during learning, audio information recording the learner's voice, text information input by the learner, and biometric information of the learner. In this case, the text information may include the learner's handwritten content, a question included in the learning, or an answer written by the learner to a problem.

[0053] Next, the learner analysis device (130) can normalize real-time information (102), remove noise, and convert it into a form that can be processed by CNN, RNN, and transformer models.

[0054] Next, the learner analysis device (130) can input real-time information (102) into the learner analysis model (101) to determine learning attitude information indicating the learner's learning attitude. Specifically, the learner analysis device (130) can diagnose the learner's immersion, concentration, emotional state, etc. by synthesizing the output of the learner analysis model (101) that has received real-time information (102) and transmit learning attitude information (103) indicating the result of comprehensively judging the learner's current learning state to the integrated education device (140).

[0055] The learner analysis device (130) can determine whether learning attitude information falls below a first threshold, which is a criterion for determining that learning attitude improvement is necessary. For example, the first threshold may be 48 points.

[0056] If the learning attitude information is below the first threshold, the learner analysis device (130) can determine feedback to increase the learner's learning concentration based on the learning attitude information. Specifically, the learner analysis device (130) can estimate the learner's emotional state and level of fatigue based on the learning attitude information. Furthermore, the learner analysis device (130) can adjust the learning schedule, including the interval or length of break times, or determine feedback suggesting changes to the learning environment based on the estimated learner's emotional state and level of fatigue.

[0057] The learner analysis device (130) determines learning attitude information representing the learner's current learning attitude based on real-time information, and if the learning attitude information is below a first threshold, provides feedback to improve the learner's learning attitude, thereby correcting the learner's learning attitude in real time. For example, the learner analysis device (130) may provide feedback to the teacher requesting that the learner say something like, "Hey, 000, can you focus a little more?" using 000 (the learner's name).

[0058] Additionally, the learner analysis device (130) can determine whether the learning attitude information exceeds a second threshold, which is a standard for determining that the learner has a superior learning attitude compared to other learners. For example, the second threshold may be 90 points. At this time, the learner analysis device (130) can determine feedback that requests the teacher to praise the learner. The learner analysis device (130) can provide feedback to the teacher by using 000 (the learner's name) to say something like, "Hey 000, you did a great job on the science problem! Keep up the good work!"

[0059] The learner analysis device (130) determines learning attitude information indicating the learner's current learning attitude based on real-time information, and when the learning attitude information exceeds a second threshold, provides feedback praising the learner, thereby inducing the learner to maintain his / her current good learning attitude.

[0060] In addition, the learner analysis device (130) can analyze the learner's learning motivation, concentration, and stress causes based on learning attitude information and suggest a learning strategy corresponding to the learner.

[0061] The integrated education device (140) can collect the learner's student information (104) and education-related information (105) and match them to learning attitude information (103).

[0062] At this time, the student information (104) includes the learner's subject grades determined by the teacher based on the learner's answers and the evaluation of the learner written by the teacher, and the integrated education device (140) can collect the student information (104) from the education information server (150).

[0063] For example, the education information server (150) is the National Education Information System (NEIS), and the student information (104) may be subject-by-subject grade information at the school the learner attends, attendance information such as absences, tardiness, and early departures, and comprehensive opinions and observations of the homeroom teacher in charge of the learner at the school the learner attends, evaluations, student specifics, and student behavioral characteristics, including student record information. At this time, the integrated education device (140) can periodically load the student information (104), which includes the student's record information, grade information, and attendance records, through the NEIS application programming interface (API). Next, the integrated education device (140) can convert the student information (104), which is in the NEIS data format, into a database format of the integrated education device (140) and store it. At this time, the NEIS can protect the student's personal information by using an encryption protocol when transmitting the student information (104).

[0064] In addition, the education-related information (105) includes the learner's education content history and learning activity information, and the integrated education device (140) can collect the education-related information (105) from the content server (160).

[0065] For example, the content server (160) may be EBS (Educational Broadcasting System), and the education-related information (105) may be online lecture viewing history and learning activity information. In this case, the integrated education device (140) may be linked to the online learning content of EBS through the API of EBS, thereby enabling the learner to access various learning materials. In addition, the integrated education device (140) may define the learning activity information provided by EBS as education-related information (105) and load it, and may comprehensively analyze the learner's learning history and achievement level based on the loaded education-related information (105).

[0066] In addition, the content server (160) is Google Classroom, and the education-related information (105) may include at least one of assignment submission, learning material sharing, class progress information, online learning activities, assignment submission, and evaluation information described in Google Classroom. At this time, the integrated education device (140) may use the API of Google Classroom to define the learner's assignment submission, learning material sharing, and class progress information as the education-related information (105) and link them. In addition, the integrated education device (140) may synchronize learning activity information occurring in Google Classroom in real time to accurately grasp the learner's current status.

[0067] In addition, the content server (160) is a Padlet, and the education-related information (105) may include at least one of the Padlet's collaborative learning activity information and bulletin board information. At this time, the integrated education device (140) may use the Padlet's API to link the collaborative learning activity information and idea sharing information. In addition, the integrated education device (140) may integrate the information collected from the Padlet with the learner's other education-related information (105) and comprehensively analyze it.

[0068] Next, the integrated education device (140) can determine the learner's comprehensive score based on the learner's student information (104), the learner's education-related information (105), and the learner's learning attitude information (103). At this time, the integrated education device (140) can determine the online score by applying preset weights for each type of information included in the real-time information (102) to video information, audio information, text information, and biometric information, and then adding them up. In addition, the integrated education device (140) can define the learner's academic performance included in the student information (104) as an offline score. At this time, the integrated education device (140) can find a method to maximize the learning effect by cross-analyzing the learner's online score and offline score. Specifically, the integrated education device (140) can determine the comprehensive score by applying weights to each of the online score and the offline score.

[0069] Next, the integrated education device (140) can determine a recommended learning strategy for the learner based on the comprehensive score. At this time, the integrated education device (140) can recommend an in-depth learning path or a basic review path for each subject based on the comprehensive score. Furthermore, the integrated education device (140) can determine the learner's learning style based on the comprehensive score, the learner's evaluation included in the student information (104), and education-related information (105).

[0070] At this time, the integrated education device (140) can determine the learner's learning effectiveness for online classes and offline classes based on the online and offline scores. Furthermore, the integrated education device (140) can determine a learning strategy that adjusts the learner's proportion of online and offline classes based on the determined learning effectiveness. For example, if the learner's online score is a certain percentage higher than the learner's offline score, the integrated education device (140) can determine a recommended learning strategy for the learner so that the proportion of online classes is higher than the proportion of offline classes.

[0071] Additionally, the integrated education device (140) can accumulate learning attitude information (103) for a preset period of time. Furthermore, the integrated education device (140) can analyze the time and situation in which the learner's learning concentration changes based on the accumulated learning attitude information to determine the learner's learning tendency. At this time, the integrated education device (140) can determine the learner's recommended learning strategy by considering the learner's learning tendency.

[0072] In addition, the integrated education device (140) can analyze each of the learner's student information (104), the learner's education-related information (105), and the learner's learning attitude information (103). For example, the integrated education device (140) can analyze the learner's long-term learning pattern and tendency by combining the academic performance information for the past six months, the learning development status, behavioral characteristics, and comprehensive opinion information included in the student information (104), and the learner's learning attitude information (103).

[0073] Specifically, the integrated education device (140) can evaluate learning achievement by analyzing subject-specific grade information included in the student record information. The integrated education device (140) can analyze attendance records included in the student record information to chronologically analyze classes and extracurricular activities participated in during the corresponding period. Furthermore, the integrated education device (140) can analyze behavioral characteristics and overall opinions included in the student record information to identify learning attitudes and characteristics.

[0074] The integrated education device (140) can recommend personalized learning (106) to learners based on recommended learning strategies. For example, the integrated education device (140) can select and recommend learning content to learners based on a determined learning style and a recommended path (an in-depth learning path or a basic review path).

[0075] A personalized learning recommendation system according to one embodiment of the present invention recognizes and analyzes students' learning attitudes in real time, and analyzes the learner's interests and aptitudes based on the learner's current learning attitude (learning attitude information (103)) and past learning information (student information (104), education-related information (105)), thereby recommending learning strategies and contents optimized for the learner.

[0076] For example, a learner analysis model of a personalized learning recommendation system can analyze facial expressions included in video information to detect that a learner is actively participating in class and has a positive expression, such as smiling, and determine the learner's concentration level as 90%. Furthermore, the learner analysis model can detect high immersion from the learner's voice tone included in audio information to determine the learner's immersion level as 85%. Furthermore, the learner analysis model can analyze biometric information to determine that the learner's heart rate is stable and stress level is low to determine the learner's stress level as 10%. The integrated education device (140) of a personalized learning recommendation system can determine that the learner has excellent science grades and has consistently shown high achievement in the past based on the student record information included in the student information to determine the learner's achievement level as 95%. At this time, the integrated education device (140) can determine the weights of the video information, audio information, biometric information, and life record as 30%, 20%, 20%, and 30%, respectively. And, the integrated education device (140) can determine the comprehensive evaluation score according to mathematical formula 1.

[0077] [Mathematical Formula 1]

[0078] Comprehensive evaluation score: (90*0.3)+(85*0.2)+(90*0.2)+(95*0.3)=90.5

[0079] At this time, the integrated education device (140) can determine the evaluation result for the learner as very excellent.

[0080] In another embodiment, a learner analysis model of a personalized learning recommendation system may have a learner's concentration level determined by facial expressions included in video information as 50%, a learner's immersion level determined by voice tone included in audio information as 45%, and a learner's stress level determined by biometric information as 70%. In addition, the integrated education device (140) of the personalized learning recommendation system may determine that the learner's English grades are low and that the learner has shown low achievement in the past based on the student record information included in the student information, and may determine the learner's achievement level as 60%. At this time, the integrated education device (140) may determine a comprehensive evaluation score according to Mathematical Formula 2.

[0081] [Equation 2]

[0082] Comprehensive evaluation score: (50×0.3)+(45×0.2)+(30×0.2)+(60×0.3)=48

[0083] Additionally, the integrated education device (140) can determine that the evaluation results for the learner require improvement.

[0084] A personalized learning recommendation system according to one embodiment of the present invention can provide online and offline learning data analysis functions through a dashboard and a communication channel between educational stakeholders (teachers, learners, and parents). Furthermore, the personalized learning recommendation system according to one embodiment of the present invention analyzes learners' interests and aptitudes based on offline class attitude records and life records, and recommends career paths and areas of study, thereby supporting more accurate and enriched personalized learning.

[0085] A personalized learning recommendation system according to one embodiment of the present invention can output a notification to a teacher when a learner becomes unfocused during class, thereby enabling the teacher to immediately identify the student's unfocused state and take action.

[0086] A personalized learning recommendation system according to one embodiment of the present invention can guide learners' learning attitudes in a more accurate and detailed direction than the current level by reflecting their learning attitudes in an offline environment where actual education takes place.

[0087] Figure 2 is an example of devices for collecting real-time information in one embodiment of the present invention.

[0088] As illustrated in FIG. 2, the camera (210) can capture a learner (200) while learning and generate image information including the learner's (200) facial expression and posture. In addition, the camera (210) can transmit the image information to the learner analysis device (130).

[0089] A microphone (220) is placed within a certain distance from a learner (200) and can record the voice of the learner (200) while learning to generate audio information containing the content of the learner's (200) speech. In addition, the microphone (220) can transmit the audio information to a learner analysis device (130).

[0090] The terminal (230) can receive input from the learner (200) of handwritten content written by the learner (200) during learning, or answers written to questions included in learning. In addition, the terminal (230) can generate text information including the input handwritten content or answers and transmit it to the learner analysis device (130). In FIG. 2, the terminal (230) is illustrated as a tablet terminal, but a PC or smart phone including an interface for receiving handwritten content or answers and a communication device for transmitting text information can also be used as the terminal (230).

[0091] The wearable device (240) can generate biometric information by measuring biometric signals such as heart rate and skin conductance of the learner (200). In addition, the wearable device (240) can transmit the generated biometric information to the learner analysis device (130). For example, the wearable device (240) can be one of a smart watch, a smart ring, and a portable heart rate monitor.

[0092] FIG. 3 is a flowchart illustrating a personalized learning recommendation method of a learner analysis device according to one embodiment of the present invention.

[0093] In step (310), the learner analysis device (130) can receive a learner analysis model (101) from the learner analysis model learning device (120).

[0094] In step (320), the learner analysis device (130) can collect real-time information (102) of the learner in real time using devices such as those illustrated in FIG. 2. For example, the real-time information (102) may include at least one of video information capturing the facial expression or posture of the learner during learning, audio information recording the learner's voice, text information input by the learner, and biometric information of the learner. In this case, the text information may include one of the learner's handwritten content, a question included in the learning, or an answer written by the learner to a problem.

[0095] In step (330), the learner analysis device (130) can input the real-time information (102) collected in step (320) into the learner analysis model (101) received in step (310) to determine learning attitude information indicating the learner's learning attitude. At this time, the image analysis model included in the learner analysis model (101) can analyze the learner's face and posture included in the image information to evaluate the learner's emotional state, such as happiness, sadness, and anger, and the learner's learning concentration. In addition, the audio analysis model included in the learner analysis model (101) can evaluate the learner's emotional state, such as tension and comfort, and learning immersion based on the learner's voice tone and speed included in the audio information, and determine the learner's learning concentration based on the evaluation result. In addition, the biometric analysis model included in the learner analysis model (101) can analyze the learner's heart rate and skin conductance information included in the biometric information to evaluate the learner's stress level and fatigue level.

[0096] Specifically, the learner analysis device (130) can diagnose the learner's immersion, concentration, emotional state, etc. by synthesizing the output of the learner analysis model (101) that receives real-time information (102) and determine learning attitude information that represents the result of comprehensively judging the learner's current learning state.

[0097] In step (340), the learner analysis device (130) can determine whether the learning attitude information falls below a first threshold. For example, the first threshold is 48 points, which is considered to be a learning attitude that requires improvement. A learning attitude information below the first threshold may indicate that the learner's concentration level is below a certain level or that the learner is experiencing a persistent negative emotional state.

[0098] If the learning attitude information is less than the first threshold, the learner analysis device (130) may perform step (350). If the learning attitude information is greater than or equal to the first threshold, the learner analysis device (130) may perform step (370).

[0099] In step (350), the learner analysis device (130) may output a notification to the teacher. For example, if a learner is not concentrating, the learner analysis device (130) may output a warning message to the teacher's monitor about the learner not paying attention. At this time, the teacher who acknowledges the warning message can immediately respond to the learner by referring to the warning message.

[0100] In step (360), the learner analysis device (130) may determine feedback to increase the learner's learning concentration based on the learning attitude information. Specifically, the learner analysis device (130) may estimate the learner's emotional state and fatigue level based on the learning attitude information. Furthermore, the learner analysis device (130) may determine feedback based on the estimated learner's emotional state and fatigue level. For example, if the learner's current concentration level is lower than the learner's average concentration level or the concentration levels of other learners, the learner analysis device (130) may determine feedback to the learner on methods to increase concentration, such as deep breathing or simple exercise.

[0101] Additionally, if the learner's current concentration is lower than the average concentration of the learner or the concentration of other learners, and the emotional state is negative, the learner analysis device (130) can determine feedback to change the learning method or recommend a break time.

[0102] Additionally, the learner analysis device (130) can provide feedback to the learner in a visual (message on the monitor), auditory (voice message) or tactile (smart watch vibration) manner.

[0103] Depending on the embodiment, step (360) may be performed in parallel with step (350), or the order of performance may be switched. In addition, only one of step (350) and step (360) may be performed.

[0104] In step (370), the learner analysis device (130) can analyze the learner's learning motivation, concentration, and stress causes based on learning attitude information and suggest a learning strategy corresponding to the learner.

[0105] For example, the learner analysis device (130) can identify the number of times the learner's fatigue exceeds a fatigue threshold during a class period. If the number of times the learner's fatigue exceeds the fatigue threshold exceeds a preset threshold, the learner analysis device (130) can determine that the learner frequently becomes fatigued during class and can suggest a learning strategy to the learner or teacher, such as, "To improve the learner's learning attitude, we recommend taking a 5-minute break after each class. Using a meditation app to reduce stress is also recommended."

[0106] In step (380), the learner analysis device (130) may transmit the learning attitude information determined in step (330) to the integrated education device (140). Depending on the embodiment, step (380) may be performed in parallel with step (340) after step (330) is performed, or steps (340) to (370) may be omitted.

[0107] FIG. 4 is a flowchart illustrating a personalized learning recommendation method of an integrated education device according to one embodiment of the present invention.

[0108] In step (410), the integrated education device (140) can collect the learner's student information (104) and education-related information (105). At this time, the student information (104) includes the learner's subject grades determined by the teacher based on the learner's answers and the teacher's evaluation of the learner, and the integrated education device (140) can collect the student information (104) from the education information server (150). In addition, the education-related information (105) includes the learner's education content history and learning activity information, and the integrated education device (140) can collect the education-related information (105) from the content server (160).

[0109] In step (420), the integrated education device (140) can receive learning attitude information determined in step (330) from the learner analysis device (130).

[0110] In step (430), the integrated education device (140) can accumulate learning attitude information (103) for a preset period of time. Then, the integrated education device (140) can analyze the time and situation in which the learner's learning concentration changes based on the accumulated learning attitude information to determine the learner's learning tendency.

[0111] Specifically, the integrated education device (140) can store the learner's learning attitude information (103) on a daily, weekly, and monthly basis and utilize it for long-term analysis. At this time, the integrated education device (140) can analyze the stored learning attitude information (103) to identify when and under what circumstances the learner's concentration increases or decreases. For example, the integrated education device (140) can analyze the learner's learning attitude information (103) for the past three months and identify a pattern in which the learner's concentration is always lower in Monday morning classes compared to classes on other days and at other times. At this time, based on the identification result, the integrated education device (140) can suggest to the teacher in step (460) to provide the learner with more frequent breaks in the learner's Monday morning class or to start the class with an activity that can attract the learner's interest.

[0112] In step (440), the integrated education device (140) can match the learner's student information (104) and education-related information (105) received in step (420) with the learning attitude information (103) collected in step (410). At this time, the integrated education device (140) can align the learning attitude information (103), the learner's student information (104), and the education-related information (105) based on the time axis, and can match and combine the learner's student information (104) and the education-related information (105) with the learning attitude information (103) according to the learner's unique identification information. For example, the integrated education device (140) can combine the record that the learner excelled in the xx activity of the xx unit in terms of the detailed abilities and special features of the learning development situation of the first grade science subject with the learner's learning attitude information (103).

[0113] In step (450), the integrated education device (140) can determine the learner's comprehensive score based on the learner's student information (104), the learner's education-related information (105), and the learner's learning attitude information (103). At this time, the integrated education device (140) can apply preset weights for each type of information included in the real-time information (102) to video information, audio information, text information, and biometric information, and then add them up to determine the online score. In addition, the integrated education device (140) can define the learner's academic performance included in the student information (104) as an offline score. The integrated education device (140) can determine the comprehensive score by applying weights to each of the online score and the offline score.

[0114] In addition, the integrated education device (140) may apply preset weights to video information, audio information, text information, and biometric information according to the type of information included in the real-time information (102), and then add them up to determine a comprehensive score. For example, the integrated education device (140) may set a weight of 50% for video information, 20% for audio information, 20% for text information, and 10% for biometric information. In addition, the integrated education device (140) may classify the current status of the learner based on the comprehensive score. For example, the integrated education device (140) may classify a learner whose comprehensive score is 90 points or higher as having a current status of “very good,” and may classify a learner whose comprehensive score is 70 to 89 points as having a current status of “excellent.” Additionally, the integrated education device (140) can classify learners with a comprehensive score of 49 to 69 points as having a current status of “average” and classify learners with a comprehensive score of 48 points or less as having a current status of “insufficient (needs improvement).”

[0115] In step (460), the integrated education device (140) may determine a recommended learning strategy for the learner based on the comprehensive score determined in step (450). At this time, the integrated education device (140) may recommend an in-depth learning path or a basic review path for each subject based on the comprehensive score. For example, the integrated education device (140) may recommend an in-depth learning path to a learner whose comprehensive score exceeds an upper threshold. Additionally, the integrated education device (140) may recommend a basic review path to a learner whose comprehensive score falls below a lower threshold.

[0116] Additionally, the integrated educational device (140) can recommend learning content based on the learner's current learning status and past information. For example, if a learner is having difficulty in a science class, the integrated educational device (140) can recommend an explanatory video and interactive quiz on the topic. Furthermore, the integrated educational device (140) can provide gamified content that can stimulate interest in learners whose emotional state is below a threshold, thereby motivating them to learn.

[0117] The integrated educational device (140) can identify a learner's learning style and recommend the most effective learning method for the learner. For example, the integrated educational device (140) can identify which learning style best suits the learner, such as visual, auditory, or kinesthetic. Furthermore, the integrated educational device (140) can determine the content to recommend to the learner based on the identified learning style. For example, the integrated educational device (140) can recommend infographics and videos to visual learners, and audio lectures and podcasts to auditory learners.

[0118] Additionally, the integrated education device (140) can determine the learner's learning style based on the comprehensive score and the evaluation of the learner included in the student information (104), and the education-related information (105).

[0119] At this time, the integrated education device (140) can determine the learner's learning effect for online classes and the learner's learning effect for offline classes based on the online and offline scores. In addition, the integrated education device (140) can determine a learning strategy that adjusts the learner's proportion of online classes and offline classes based on the determined learner's learning effects. For example, if a learner's online score for the same subject and at the same time is higher than his / her offline score, the integrated education device (140) can determine that the learner shows high achievement in online learning but tends to have low concentration in offline classes. At this time, the integrated education device (140) can determine a learning strategy so that the proportion of online learning is higher than that of offline learning. In addition, the integrated education device (140) can also determine a learning strategy that changes the learning environment of the offline class so that the learner's concentration is increased.

[0120] At this time, the integrated education device (140) may determine a recommended learning strategy for the learner by considering the learner's learning tendency determined in step (430).

[0121] At step (460), the integrated education device (140) can recommend personalized learning (106) to the learner based on the recommended learning strategy. The integrated education device (140) can select and recommend learning content to the learner based on the determined learning style and the recommended path (advanced learning path or basic review path).

[0122] For example, if a learner shows interest in science and excels in academics, the integrated education device (140) may encourage the learner to prepare for a science competition and provide in-depth learning materials and competition preparation questions. If the learner is struggling with English, the integrated education device (140) may provide the learner with basic supplementary learning materials and recommend customized tutoring sessions for learning basic English grammar and vocabulary.

[0123] Additionally, if a learner has recently shown low concentration and poor grades in math classes, the integrated education device (140) can use learning attitude information to confirm that the learner has low concentration and, through the life record information included in the student information (104), can identify that a similar pattern has existed in the past. At this time, the integrated education device (140) can recommend a basic concept review path for the learner and recommend an interactive quiz that can stimulate interest as a personalized learning method.

[0124] The integrated education device (140) may also recommend team projects and group activities based on collaborative learning activities participated in on Padlet.

[0125] In step (470), the integrated education device (140) may periodically or irregularly transmit a report to at least one of the learner, the parent, and the teacher based on the learner's learning tendency determined in step (430) and the recommended learning strategy determined in step (460). At this time, the report provided by the integrated education device (140) may include graphs, charts, etc. that visually represent the learner's learning pattern, concentration change, and emotional state analysis results. For example, the integrated education device (140) may transmit a report containing the following content to at least one of the learner, the parent, and the teacher: "The learner showed very high concentration in science class this week and low concentration in English class. We will provide additional learning materials to improve English grades."

[0126] A personalized learning recommendation system according to one embodiment of the present invention recognizes and analyzes students' learning attitudes in real time, and analyzes the learner's interests and aptitudes based on the learner's current learning attitude (learning attitude information (103)) and past learning information (student information (104), education-related information (105)), thereby recommending learning strategies and contents optimized for the learner.

[0127] Meanwhile, the personalized learning recommendation system or personalized learning recommendation method according to the present invention can be written as a program that can be executed on a computer and can be implemented in various recording media such as a magnetic storage medium, an optical reading medium, and a digital storage medium.

[0128] Implementations of the various technologies described herein may be implemented as digital electronic circuitry, or as computer hardware, firmware, software, or combinations thereof. Implementations may be implemented as a computer program product, for example, a computer program tangibly embodied in a machine-readable storage device (computer-readable medium), for processing by the operation of a data processing device, for example, a programmable processor, a computer, or multiple computers, or for controlling the operation thereof. A computer program, such as the computer program(s) described above, may be written in any form of programming language, including compiled or interpreted languages, and may be deployed in any form, including as a standalone program or as a module, component, subroutine, or other unit suitable for use in a computing environment. A computer program may be deployed to be processed on one computer or multiple computers at a single site, or to be distributed across multiple sites and interconnected by a communications network.

[0129] Processors suitable for processing a computer program include, for example, both general-purpose and special-purpose microprocessors, and any one or more processors of any type of digital computer. Typically, a processor will receive instructions and data from read-only memory or random-access memory, or both. Components of a computer may include at least one processor for executing instructions and one or more memory devices for storing instructions and data. Typically, a computer may include, or be coupled to receive data from, transmit data to, or both, one or more mass storage devices, such as magnetic, magneto-optical, or optical disks, for storing data. Information carriers suitable for embodying computer program instructions and data include, for example, semiconductor memory devices, magnetic media such as hard disks, floppy disks, and magnetic tapes, optical media such as compact disk read only memory (CD-ROM), digital video disks (DVD), magneto-optical media such as floptical disks, read only memory (ROM), random access memory (RAM), flash memory, erasable programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), etc. The processor and memory may be supplemented by, or included in, special purpose logic circuitry.

[0130] Additionally, the computer-readable medium may be any available medium that can be accessed by a computer, and may include all computer storage media.

[0131] While this specification contains details of a number of specific implementations, these should not be construed as limitations on the scope of any invention or what may be claimed, but rather as descriptions of features that may be unique to particular embodiments of particular inventions. Certain features described herein in the context of individual embodiments may also be implemented in combination in a single embodiment. Conversely, various features described in the context of a single embodiment may also be implemented in multiple embodiments, either individually or in any suitable subcombination. Furthermore, although features may operate in a particular combination and may initially be described as being claimed as such, one or more features from a claimed combination may in some cases be excluded from that combination, and the claimed combination may be modified into a subcombination or variation of a subcombination.

[0132] Likewise, while operations are depicted in the drawings in a particular order, this should not be construed as requiring that those operations be performed in the particular or sequential order depicted to achieve desired results, or that all depicted operations be performed. In certain instances, multitasking and parallel processing may be advantageous. Furthermore, the separation of the various device components of the embodiments described above should not be construed as requiring such separation in all embodiments, and it should be understood that the program components and devices described may generally be integrated together in a single software product or packaged into multiple software products.

[0133] Meanwhile, the embodiments of the present invention disclosed in this specification and drawings are merely specific examples presented to aid understanding and are not intended to limit the scope of the present invention. It will be apparent to those skilled in the art that other modifications based on the technical concepts of the present invention are possible in addition to the embodiments disclosed herein.

Claims

1. A step of determining learning attitude information representing the learning attitude of the learner by inputting real-time information collected by measuring the learner performing learning in real time into a learner analysis model; A step of collecting the student information and education-related information of the above learner and matching it to the learning attitude information; A step of determining a comprehensive score of a learner based on the learner's student information, the learner's education-related information, and the learning attitude information; and A step of determining a recommended learning strategy for the learner based on the above comprehensive score. Including, The above student information is, Includes the learner's course grades determined by the teacher based on the learner's answers and the evaluation of the learner written by the teacher. The above education-related information is: A personalized learning recommendation method including the learner's educational content history and learning activity information.

2. In paragraph 1, A step of accumulating the learning attitude information for a preset period of time; and A step of determining the learner's learning tendency by analyzing the time and situation in which the learner's learning concentration changes based on accumulated learning attitude information; A personalized learning recommendation method that includes more.

3. In paragraph 1, A step of analyzing the learner's learning motivation, concentration, and stress causes based on the above learning attitude information and proposing a learning strategy corresponding to the learner. Including, The above real-time information is, It includes at least one of video information of the learner, audio information of the learner's voice, text information entered by the learner, and biometric information of the learner. The above learner analysis model is, An image analysis model that analyzes the learner's learning concentration and emotional state based on the learner's face and posture included in the image information; An audio analysis model that analyzes the learner's learning concentration based on the learner's voice tone and speed included in the audio information; A text analysis model that analyzes the learner's learning concentration and learning performance based on the learner's handwritten content or answers written by the learner included in the text information; and A biometric analysis model that analyzes the learner's learning concentration and fatigue level based on the above biometric information. A personalized learning recommendation method comprising at least one of:

4. In paragraph 3, The step of determining the above comprehensive score is: A step of applying preset weights for each type of information included in the real-time information to the video information, the audio information, the text information, and the biometric information, and then adding them to determine an online score; A step of defining the above learner's course grades as offline scores; and A step of determining the overall score by applying weights to each of the online score and the offline score. A personalized learning recommendation method that includes:

5. In paragraph 4, The steps for determining the above recommended learning strategy are: A step of determining the learning effect of the learner for the online class and the learning effect of the learner for the offline class based on the online score and the offline score; and A step of determining a learning strategy that adjusts the proportion of online and offline classes for the learner based on the learning effects of the learner determined above. A personalized learning recommendation method that includes:

6. In paragraph 1, A step of checking whether the above learning attitude information is below the first threshold, which is a criterion for determining that the learning attitude requires improvement; and A step of determining feedback to increase learning concentration for the learner based on the learning attitude information when the learning attitude information is less than a first threshold value; A personalized learning recommendation method that includes more.

7. In paragraph 6, The steps for determining the above feedback are: A personalized learning recommendation method that estimates the emotional state and fatigue level of the learner based on the above learning attitude information, and adjusts the learning schedule including the interval or length of break time according to the emotional state and fatigue level of the learner, or determines feedback suggesting a change in the learning environment.

8. In paragraph 1, A step of checking whether the above learning attitude information exceeds the second threshold, which is a standard for determining that the learning attitude is superior to that of other learners; and A step of determining feedback that requires the teacher to praise the learner when the above learning attitude information exceeds the second threshold. A personalized learning recommendation method that includes more.

9. In paragraph 1, The steps for determining the above recommended learning strategy are: A step of recommending an in-depth learning path or a basic review path for each subject based on the above comprehensive score; A step of determining the learning style of the learner based on the above comprehensive score, the evaluation of the learner, and the education-related information; and A step of selecting learning content and recommending it to the learner based on the above learning style and recommended path. A personalized learning recommendation method that includes:

10. A learner analysis device that measures learners performing learning in real time, collects real-time information, and inputs it into a learner analysis model to determine learning attitude information indicating the learner's learning attitude; and An integrated educational device that collects the student information and education-related information of the learner, matches them with the learning attitude information, determines the learner's comprehensive score based on the student information of the learner, the learner's education-related information, and the learning attitude information, and determines the learner's recommended learning strategy based on the comprehensive score. Including, The above student information is, Includes the learner's course grades determined by the teacher based on the learner's answers and the evaluation of the learner written by the teacher. The above education-related information is: A personalized learning recommendation system that includes the learner's educational content history and learning activity information.

11. In paragraph 10, The above real-time information is, It includes at least one of video information of the learner, audio information of the learner's voice, text information entered by the learner, and biometric information of the learner. The above learner analysis model is, An image analysis model that analyzes the learner's learning concentration and emotional state based on the learner's face and posture included in the image information; An audio analysis model that analyzes the learner's learning concentration based on the learner's voice tone and speed included in the audio information; A text analysis model that analyzes the learner's learning concentration and learning performance based on the learner's handwritten content or answers written by the learner included in the text information; and A biometric analysis model that analyzes the learner's learning concentration and fatigue level based on the above biometric information. A personalized learning recommendation system comprising at least one of:

12. In paragraph 11, The above integrated educational device, A personalized learning recommendation system that applies preset weights for each type of information included in the real-time information to the video information, the audio information, the text information, and the biometric information, and then adds them to determine an online score, defines the learner's academic performance as an offline score, and determines the comprehensive score by applying weights to each of the online score and the offline score.

13. In paragraph 12, The above integrated educational device, A personalized learning recommendation system that determines the learner's learning effect for online classes and the learner's learning effect for offline classes based on the online score and the offline score, and determines a learning strategy that adjusts the proportion of online classes and offline classes for the learner based on the determined learning effects of the learner.

14. In paragraph 10, The above learner analysis device, A personalized learning recommendation system that determines whether the above learning attitude information is below a first threshold, which is a criterion for determining that the learning attitude requires improvement, and if the above learning attitude information is below the first threshold, estimates the emotional state and fatigue of the learner based on the learning attitude information, and adjusts the learning schedule including the interval or length of break time according to the emotional state and fatigue of the learner, or determines feedback suggesting a change in the learning environment.

15. In paragraph 10, The above learner analysis device, A personalized learning recommendation system that determines whether the above learning attitude information exceeds a second threshold, which is a standard for determining that the learning attitude is superior to that of other learners, and, if the above learning attitude information exceeds the second threshold, determines feedback requesting the teacher to praise the learner.

16. In paragraph 10, The above integrated educational device, A personalized learning recommendation system that recommends an in-depth learning path or a basic review path for each subject based on the above comprehensive score, determines the learning style of the learner based on the above comprehensive score, an evaluation of the learner, and the above education-related information, and selects and recommends learning content to the learner based on the above learning style and the recommended path.

Images