KR102650585B1 - Method for a learning management system based on virtual lecture space using learner-digital twin generation - Google Patents

Abstract

The present invention relates to a learning management system and management method in a virtual lecture space utilizing the creation of a learner-digital twin. More specifically, the present invention relates to a learning management system in a virtual lecture space that utilizes the creation of a learner's avatar and provides a learning service in a virtual space reflecting the learner's avatar. It relates to systems and methods for increasing efficiency.
The learning management system of the virtual lecture space utilizing learner-digital twin creation according to the present invention includes an image acquisition unit that captures the learner's video and transmits the video data, and receives virtual lecture space information created using the video data. It includes a learner terminal that displays information, a learning management server that generates virtual lecture space information using video data and analyzes the learner's concentration, and an instructor terminal that is connected to the learning management server and delivers lecture content.

Description

Learning management system and management method of virtual lecture space using learner-digital twin creation {METHOD FOR A LEARNING MANAGEMENT SYSTEM BASED ON VIRTUAL LECTURE SPACE USING LEARNER-DIGITAL TWIN GENERATION}

The present invention relates to a learning management system and management method in a virtual lecture space utilizing the creation of a learner-digital twin. More specifically, the present invention relates to a learning management system in a virtual lecture space that utilizes the creation of a learner's avatar and provides a learning service in a virtual space reflecting the learner's avatar. It relates to systems and methods for increasing efficiency.

Due to COVID-19, the mode of class has shifted to non-face-to-face classes (remote, online classes), and as time goes by, both instructors and learners are adapting to a certain extent to the non-face-to-face class method. However, regardless of adaptation to the non-face-to-face class method, there are problems such as students' lack of concentration in non-face-to-face classes, decreased learning effectiveness due to excessive use of the Internet or SNS during study, and worsening polarization of students' grades. Problems and other disadvantages are still being raised as unresolved issues.

Due to COVID-19, the improvement of self-directed learning ability in a non-face-to-face learning environment has inevitably become an important issue, and according to the prior art (Domestic Registered Patent No. 10-1766347), the viewing behavior of the evaluator watching video content A technology is being developed to evaluate the degree of concentration on video content using captured video data.

The present invention was proposed to solve the above-mentioned problems, and it analyzes the learner's gaze, face angle and direction, and upper body posture in a complex way to determine whether the learner is concentrating on the online class (virtual lecture) and uses a 3D avatar to determine the learner's learning status. Through this, it is reflected in the 3D virtual space in the video lecture, and through concentration analysis, it encourages learners to focus more on online classes, and by sharing concentration analysis results, the instructor can clearly see the status of multiple learners in online classes. The purpose is to provide a learning management system and management method for a virtual lecture space utilizing learner-digital twin creation that can support verification.

The learning management system of the virtual lecture space utilizing learner-digital twin creation according to the present invention includes an image acquisition unit that captures the learner's video and transmits the video data, and receives virtual lecture space information created using the video data. It includes a learner terminal that displays information, a learning management server that generates virtual lecture space information using video data and analyzes the learner's concentration, and an instructor terminal that is connected to the learning management server and delivers lecture content.

The learning management server receives the image data from the image acquisition unit, extracts facial feature points and upper body feature points, and measures the learner's concentration by comprehensively considering the face angle, gaze direction, and upper body posture.

The learning management server provides behavior correction guidelines to the learner terminal based on the concentration level.

The learning management server provides a quest by adjusting quest settings differently based on the concentration, and verifies the concentration by receiving the quest performance result of the learner terminal.

The learning management server updates the concentration analysis model using the results of the concentration verification.

The learning management method in a virtual lecture space using learner-digital twin creation according to the present invention includes the steps of (a) acquiring the learner's image, (b) creating a 3D avatar of the learner using the learner's image, and creating a 3D avatar. (c) analyzing the learner's concentration on watching the virtual lecture content and reflecting the learner's behavior in the behavior of the 3D avatar; (d) according to the analysis result of the concentration Providing a quest, performing analysis on the quest performance results, and (e) using the analysis of the quest performance results to verify the concentration analyzed in step (c) and updating the concentration analysis model. It includes steps to perform.

In step (b), feature points of the face and upper body are extracted from the learner's image, and the 3D avatar is created based on the feature points.

In step (c), the learner's concentration is analyzed considering the face angle, gaze direction, and upper body posture, and the learner's actual behavior is reflected in the behavior of the 3D avatar included in the virtual lecture content.

In step (d), the setting information of the quest is adjusted differently based on the analysis result of the concentration and the purpose of provision.

Step (e) uses the results of verifying the concentration to update at least one of a concentration analysis model built for each learner and a general concentration analysis model that does not reflect individual characteristics.

According to the present invention, the problem of sexual polarization due to differences in educational infrastructure is solved by encouraging learners to participate in virtual lectures with interest and providing learning-related information such as grades and curriculum and information on learning attitudes such as concentration. It is expected that this will be possible.

According to the present invention, by improving the learner's immersion and concentration in lectures and providing academic-related information such as lecture concentration and grades, the effect of solving problems in online lectures and cultivating self-directed learning ability in a non-face-to-face class environment is achieved. there is.

According to the present invention, it is possible to provide guidelines (feedback) using a 3D avatar, and it has the effect of increasing the learner's usability of the system by providing information in the form of a chatbot about problems that may be encountered during learning.

The effects of the present invention are not limited to those mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art from the description below.

Figure 1 shows a learning management system in a virtual lecture space utilizing learner-digital twin creation according to an embodiment of the present invention.
Figure 2 is an example diagram showing an algorithm applied to the concentration analysis unit in the learning management system of the virtual lecture space utilizing learner-digital twin creation according to an embodiment of the present invention.
Figure 3 shows a learning management method in a virtual lecture space using learner-digital twin creation according to an embodiment of the present invention.
Figure 4 is a block diagram showing a computer system for implementing a method according to an embodiment of the present invention.
Figure 5 shows learning management using learner back data according to an embodiment of the present invention.

The above-mentioned object and other objects, advantages and features of the present invention, and methods for achieving them will become clear by referring to the embodiments described in detail below along with the accompanying drawings.

However, the present invention is not limited to the embodiments disclosed below and can be implemented in various different forms. The following embodiments are merely intended to convey to those skilled in the art the purpose of the invention, It is only provided to easily inform the configuration and effect, and the scope of rights of the present invention is defined by the description of the claims.

Meanwhile, the terms used in this specification are for describing embodiments and are not intended to limit the present invention. As used herein, singular forms also include plural forms, unless specifically stated otherwise in the context. As used in the specification, “comprises” and/or “comprising” means that the mentioned element, step, operation and/or element precludes the presence of one or more other elements, steps, operations and/or elements. Or it is not excluded that it is added.

Metaverse, an extended virtual world, refers to a three-dimensional virtual world. Specifically, it is widely used to mean a living or game-type virtual world where reality and unreality coexist in general aspects of politics, economy, society, and culture.

According to an embodiment of the present invention, a system and management method are proposed to provide a learning service in a virtual space to increase learning efficiency, create a learner's avatar and reflect it in the virtual space, and perform learning management. According to an embodiment of the present invention, the learner's 3D avatar is reflected in a video lecture in a 3D virtual space, and the real-time 3D rendering work results for the learner's face are linked so that they appear in the form of a 3D avatar, so that the learner's behavior and Changes in gaze are reflected in the virtual space in real time. For example, when a learner moves out of the camera's viewing angle, the learner's 3D avatar also disappears from the screen or an indication of the learner's absence is reflected on the 3D avatar, allowing the instructor or administrator to intuitively monitor the learner's attendance and class participation attitude. You can check it.

According to an embodiment of the present invention, feature points and upper body feature points for each facial element are extracted from the learner image obtained as an RGB image, and the learner's concentration is measured by complexly extracting and analyzing the face angle, gaze direction, and upper body posture. . Based on the measured concentration of the learner, after the theory class in the virtual space (or at a preset point during the class), the learner can review the lecture content by performing quests in various formats (for example, it may be provided in the form of a quiz). can be performed. At this time, the quest setting information (e.g., difficulty of the problem, number of problems, classification of the problem) may be adjusted and provided based on the learner's concentration, and the adjustment of the quest setting information may be set differently depending on the purpose. there is. For example, if the purpose of the evaluation is to enable students who concentrated in class to receive a better evaluation, the quest should include not only the basic learning content but also a quiz that includes questions that students who focused on the class are likely to answer correctly. can be set. As another example, even if a student is unable to concentrate in class, if the purpose is to improve understanding of the content of the lesson rather than evaluation, the concept explained at a time when the student was unable to concentrate is provided with additional explanatory materials. Quests can be set up to include questions (quizzes) to check whether one understands or not. As another example, it is possible to provide a quest containing more advanced problems than the basic questions to a student who was highly focused in class, and to grant a higher reward for successfully completing the quest. As another example, even if the question matches the same answer depending on class concentration, the quest settings are adjusted differently depending on whether the quest is provided in a multiple-choice, subjective, or narrative format. possible.

According to an embodiment of the present invention, the difficulty of the quest is provided differently according to the learner's concentration analysis data, and by reflecting the quest result data, it is possible to more accurately analyze the learner's concentration. By linking with the academic management platform and learning information accumulated during lectures in real time, it is possible to provide learners with various types of guidelines (feedback) related to lectures and attitudes during learning.

Figure 1 shows a learning management system in a virtual lecture space utilizing learner-digital twin creation according to an embodiment of the present invention.

The learner terminal 100 is a terminal of a learner accessing the Internet in a virtual lecture space, and includes a plurality of first learner terminals 100a to nth learner terminals 100n. Each learner terminal includes an image acquisition unit (110a to 110n), for example, the image acquisition unit is equipped with a camera including a depth sensor. The image acquisition unit may be a camera installed on the learner's terminal, or it may be a separate camera. The learning management server includes a user recognition unit 210, a virtual lecture space creation unit 220, a question and answer management unit 230, a concentration analysis unit 240, and an analysis standard update unit 250. The learning management server 200 is connected to the instructor terminal 300 and the academic database 400 to share information in the virtual lecture space.

The learning management server 200 transmits concentration and attendance data to the academic database 400 and stores them, and receives grades, course information, and course satisfaction data from the academic database 400 and shares them with the learner terminal 100.

The user recognition unit 210 receives images captured through the image acquisition units 110a to 110n and extracts learner characteristic information and depth information.

The virtual lecture space creation unit 220 generates a 3D avatar capable of expressing movements in real time using the extracted feature information and depth information, and creates a virtual lecture space in which the 3D avatar operates. The virtual lecture space creation unit 220 reflects changes in the learner's behavior over time on the 3D avatar in real time. The virtual lecture space creation unit 220 receives learner recognition information and motion recognition results, and analyzes facial expression, face angle, and face direction information to create a 3D avatar.

The concentration analysis unit 240 analyzes the learner's concentration by analyzing the learner's facial and upper body posture. The concentration analysis results are transmitted to the instructor terminal 300 and the academic database 400, so that the instructor can check the concentration information of each of the plurality of learners in real time through the instructor terminal 300.

As described above, the learning management server 200 is linked with the instructor terminal 300 and the academic database 400 to provide learning-related information such as concentration analysis results, academic-related data, and subject curriculum. Learners can participate in virtual lectures and obtain academic-related information using a learning management system according to an embodiment of the present invention on a personal terminal (smart device, learner terminal).

The learning management server 200 provides information about the virtual lecture space by reflecting the instructor's 3D avatar as well as the learner's 3D avatar in the virtual space.

The virtual lecture space creation unit 220 includes a module that creates a 3D avatar of the instructor and a module that creates a 3D avatar of the learner, and displays an image including the 3D avatar from a first-person or third-person perspective according to user settings. .

The background for the virtual lecture space may be applied differently depending on user settings. The learning management server 200 can reflect the instructor's video acquired by the instructor terminal 300 in creating the instructor's 3D avatar in real time. As another example, the learning management server 200 may use the instructor's voice received through the instructor terminal 300 to automatically reflect the pose or movement on the instructor's 3D avatar. As another example, the learning management server 200 analyzes the instructor's voice received through the instructor terminal 300 and changes the content, location, etc. of the learning content currently being provided in the virtual lecture space, or changes the current explanation. Highlighting processing, etc. for content within the virtual lecture space can be performed automatically. A virtual lecture space is displayed on the learner terminal 100, and multiple learners can share the virtual lecture space in real time and communicate with the instructor through voice, chat, etc. In addition, it transmits the voice received through a smart device or a separate microphone, recognizes the voice, and displays it as text in chat.

The virtual lecture space creation unit 220 according to an embodiment of the present invention collects motion data about the learner's movements, infers behavior using the motion data, and changes the movement of the learner's 3D avatar. According to an embodiment of the present invention, when a learner falls asleep, the motion of the learner's 3D avatar changes to a dozing motion, and when the learner leaves the classroom, the learner's 3D avatar also leaves the classroom. In addition to class content, learners can check their own academic information and information about other students taking the course together, and it is possible to communicate any questions they have in the form of a chatbot. For example, if a student asks a question about some area of the class content, the student can review by providing the exact page and content of the relevant textbook.

When a virtual lecture space is created by the virtual lecture space creation unit 220, information on the virtual lecture space (lecture name, professor name, classroom environment) is provided, and the question and answer management unit 230 provides information on the content of the lecture, Provides learning services (quizzes, teaching materials, etc.). At this time, the settings of the provided quest (e.g. difficulty level, number of questions, type of question) may be adjusted differently depending on the concentration analysis results.

The question and answer management unit 230 adjusts the quest setting information (e.g., difficulty level of the problem, number of questions, classification of the problem) based on the learner's concentration, and the adjustment of the quest setting information can be set differently depending on the purpose. there is. If the purpose in terms of evaluation is to enable students who relatively focused more on the class to receive a better evaluation, the question and answer management unit 230 not only provides the basic learning content, but also ensures that students who focused on the class have a high probability of getting the correct answer. Set up the quest to include a quiz with questions. As another example, even if the student is unable to concentrate in class, if the purpose is to improve understanding of the class content rather than evaluation, the question and answer management unit 230 determines the timing when the student was unable to concentrate and responds to that timing. For the main concepts established, the quest is set up to include additional explanatory materials and questions (quizzes) to check whether the concepts are understood. The question and answer management unit 230 provides quests containing more advanced problems than the basic questions to students who were highly focused in class, and also grants higher rewards for successfully completing the quest. Even if the question matches the same answer, the question and answer management unit 230 determines whether to provide a quest by setting the answer method for the question in a multiple-choice format, providing a quest by setting it in a subjective format, or providing a quest in a narrative format. Adjust the quest settings differently.

The virtual lecture space can be set according to the content of the educational content. For example, if the lecture content is related to Chinese history, the instructor can set a place related to China as the background of the virtual lecture space. As another example, learners can set their favorite background (e.g., a photo of an amusement park) as the background of the virtual lecture space.

When the educational content is a team project, the virtual lecture space creation unit 220 requires collaboration between multiple learner 3D avatars, so it provides a service for transmitting materials necessary for team members' collaboration and a chat for communication between team members. Additionally, voice exchange services and screen sharing services for collaboration are provided.

According to another embodiment of the present invention, the learning management server 200 receives data sources (contribution, average concentration, etc.) from the instructor terminal and uses this to allow the 3D avatar to proceed with learning on its own.

As learning progresses using data sources, lecture and academic-related information is accumulated, and based on the accumulated data, learners are provided with academic/lecture/assignment-related information and learning-related guidelines in the form of a chatbot through a 3D avatar. You can receive it.

According to another embodiment of the present invention, it is possible to perform a precise concentration analysis based on academic data sources and provide learner guidance based on data utilized and acquired within the system, including input academic data. For example, for users who have difficulty performing team assignments, they can ask questions such as 'Are you worried about studying for the test?'/'Is the group assignment difficult?'/'Is the assignment difficult?' through the chatbot chat room. If the second option among the chat boxes is selected, a service is provided so that the learner can receive the desired information through a preset question process.

As another example, in the case of users who have difficulty recruiting team members when performing team tasks, they can request help from a 3D avatar in the form of a chatbot (chat) and receive information about the lecture and other participants (average team task score, contribution level, etc.) Recommend team members based on data. As another example, if a learner feels difficulty while studying for a test, he or she can request help in the form of a chatbot (chat) from his/her avatar, and can request the instructor's voice file and class materials recorded during the unit or lecture. It is possible to provide

According to an embodiment of the present invention, analysis results (eye tracking result data, upper body posture result data, concentration data, emotional change data by time period) and academic information (professor evaluation data, course satisfaction, grade data, academic data, course curriculum) ) to create and provide an analysis report.

The analysis report includes daily and monthly concentration analysis results, increase compared to the previous day or previous month, posture verification, information on concentration time compared to total learning time, concentration change during today's study, and graphs measuring posture change. The analysis report provides a service that recommends curricular and extra-curricular activities related to lectures in which learners showed high concentration, informs them of core competencies related to lectures in which learners showed high concentration, and recommends appropriate job groups.

Figure 2 is an exemplary diagram showing an algorithm applied to the concentration analysis unit in the learning management system through provision of virtual lecture space according to an embodiment of the present invention.

As a first input, a user recognition result is received. The user recognition result is a recognition result of video data captured about the learner, and includes the learner's gaze, head direction, and upper body posture recognition results.

Image-based concentration analysis is performed using a general concentration analysis model, which is a general model for analyzing the learner's average concentration, and an individual concentration analysis model for analyzing the characteristic concentration of each individual learner.

At this time, using the general concentration analysis model, the learner terminal is provided with a learning service using a virtual lecture space or does not stare at the screen during the explanation of the concept, rests its chin, lowers its head, or dozes off while nodding its head. If you focus on an area other than the area of , the analysis is performed as if your concentration is low.

Learning habits may differ depending on the learner. For example, if learner A has a habit of closing his eyes at the end of a concept explanation and repeating the content while learning, he should not be judged to be dozing based on the video analysis results. will be. Therefore, it is possible to increase the reliability of the concentration analysis, which is the result of analysis of the acquired video, by using not only the general concentration analysis model but also the individual concentration analysis model built for each individual student.

In addition, according to an embodiment of the present invention, the concentration level is not simply analyzed based on video analysis, but the result of the question and answer about the class content is received as a second input to determine whether the class content was actually attended and understood well by concentrating. Check . In other words, concentration verification is performed based on question and answer results.

Based on the concentration level, which is the first output, the type, form (subjective, multiple choice, descriptive), and difficulty level of the query (quest) can be adjusted differently.

For the quest provided based on concentration, which is the first output, the learner's response is received as a question and answer result to determine whether the quest is performed incorrectly despite being judged to be high concentration or whether the quest is performed well despite being judged to be low concentration. The performance results are confirmed and it is confirmed whether the level of concentration was actually high.

For example, it was determined that Learner B's gaze was off the screen, that is, the screen of the learner's terminal where the virtual lecture space is provided, and it was judged that his concentration was low. However, in reality, Learner B did not stare at the screen for a while and was studying the subject spread out on his desk. Assume that you were looking at a reference book. In the case of video-based concentration analysis results, it is determined that the learner's gaze is simply not focused on the content, so concentration will be judged to be low. However, if learner B attempted to understand the concept more accurately by briefly referring to other reference books during class through self-directed learning, the probability of success in the quest would be higher. In other words, as a result of the analysis of learner B's gaze, it was recognized that his concentration was low, but when verifying his concentration based on the question and answer results, he could be recognized as having a high concentration. In this case, by reflecting it in the individual concentration analysis model, the model can be updated to ensure that the learning concentration does not decrease even if the learner's gaze deviates. This model update is performed by the analysis standard update unit 250.

Even if it is determined that a specific learner's behavior was low in concentration as a result of video analysis, or that he was actually well focused as a result of completing the quest, there are cases where the specific learner's behavior cannot be considered correct behavior. For example, if a specific learner has the habit of biting his or her nails when concentrating, even if he or she concentrated on taking the lecture and performed the quest well, this habit should not be recommended as a correct habit. Therefore, according to an embodiment of the present invention, the behavioral pattern in the concentration state is confirmed for each learner, and whether to provide behavior correction guidelines is determined based on the positive index classified for the behavioral pattern. For example, the habit of biting nails has a positive index of 0 points (standard point: 10 points), so behavior correction guidelines are immediately provided to learners.

Behavior correction guidelines may be provided based on video analysis results, and the timing of provision may vary depending on the monitoring results of whether it is a one-time behavior or a habitual behavior. For example, in the case of fanning one's hands in hot weather, even if the video analysis results show that concentration is low and behavioral correction guidelines need to be provided accordingly, the behavior was performed temporarily in a hot environment and is performed habitually. If it is monitored that it is not being done (for example, while the 1 hour and 30 minute lecture was conducted three times, only the first lecture was conducted for a total of 25 minutes), the behavior correction guideline is controlled so that it is not provided.

According to an embodiment of the present invention, the analysis standard update unit 250 analyzes the update results for each learner's individual concentration analysis model to determine whether the update will be limited to an individual concentration analysis model update or a general concentration analysis model update. Decide whether to reflect it or not. For example, the concentration analysis standard of the general concentration analysis model can be updated based on the behavioral patterns of learners exceeding a certain standard number compared to the total number of people connected to the virtual lecture space. For example, let's assume that this is a subject in which concepts are explained in a virtual lecture space and practice using supplementary materials is performed offline. In this case, learners will stare at the screen when explaining the concept, but after the concept explanation, they will stare at the supplementary material when practicing using the supplementary material offline. Therefore, during this time, concentration cannot be determined to be low simply using the general concentration analysis model. Instead, it is desirable to adjust the analysis standards of the general concentration analysis model to ensure that the reliability of the concentration analysis does not decrease.

Figure 3 shows a learning management method in a virtual lecture space using learner-digital twin creation according to an embodiment of the present invention.

The learning management method in a virtual lecture space using learner-digital twin creation according to an embodiment of the present invention includes the steps of acquiring a learner video (S310), creating virtual lecture content (S320), and maintaining the learner's concentration. A step of analyzing and reflecting the analysis results in the content (S330), a step of providing a quest to the learner and performing an analysis of the quest performance results (S340), and a step of verifying the concentration analysis results and updating the concentration analysis model ( S350).

In step S310, a learner video is acquired through a camera placed on the learner terminal or provided separately from the learner terminal. The learner video includes the learner's gaze information, the learner's behavior information, the learner's posture information, and the learner's facial information.

In step S320, a 3D avatar of the learner is created using the learner image obtained in step S310, and virtual lecture content reflecting the 3D avatar is created. A virtual lecture space containing a plurality of learners' 3D avatars and the instructor's 3D avatar is created, and the background can be set to a background related to the virtual lecture content.

In step S330, the learner's concentration is analyzed according to the results of analyzing the learner's video, and the learner's actual offline behavior is reflected in the behavior of the 3D avatar in the virtual space.

Step S340 provides a quest whose settings are adjusted differently according to the concentration information analyzed in step S330, and analyzes the learner's performance results for the quest.

Step S350 uses the quest performance results to verify the results of the image-based concentration analysis performed in step S330. Based on the verification results, updates to the individual concentration analysis model are performed and, if necessary, updates to the general concentration analysis model are performed.

According to another embodiment of the present invention, in order to collect data for providing a virtual lecture space, a proxy server is established and data crawling is performed. Securing a stable proxy is essential for data collection, and if a static IP is collected (calls) more than N times per second per site, the IP itself is often banned. In addition, because dynamic IPs are relatively advantageous in bypassing interference from the site for the purpose of data collection when collecting data such as crawling, proxy services generally charge a higher fee for dynamic IPs than for static IPs. It is set. According to another embodiment of the present invention, in performing automatic crawling, a proxy server formed by a group of mobile devices with attached mobile SIM cards is used, and the proxy server is managed and data collection is performed according to data collection requests. Crawling is performed using a management server that arranges mobile devices. The management server receives data collection requests from a crawler server or a web server that delivers requests from general users. The management server manages the registration status for the proxy server using information on mobile devices being incorporated into the group and mobile devices being expelled from the group. The management server controls the dynamic IP change cycle of mobile devices included in the group. The management server monitors whether the data collection request is a harmful attack on the site, and when the attack is confirmed, it blocks the user who sent the request and manages a blacklist. The mobile SIM is attached to a computing device (mobile device) to establish a stable dynamic IP proxy service. In addition to mobile devices, computing devices may be workstations, laptops, PDAs, etc. A separate app is installed on mobile devices, and grouped mobile devices are defined as proxy servers to build a dynamic IP proxy service. The management server receives a collection request from a crawler server or a web server and delivers the request to at least one device included in the connected proxy server. The management server manages the communication status of the grouped mobile devices and sub-groups at least one mobile device to perform data collection according to the request within the proxy server. The management server considers the expected task completion time of mobile devices that are currently performing other data collection requests, performs sub-grouping of mobile devices, and sends an alarm to the administrator when necessary to detect new mobile devices. request. When considering the behavior information of a 3D avatar to provide a virtual lecture space, the management server performs data crawling targeting the behavior information of other avatars. When considering facial expressions, the management server performs data crawling targeting the facial expression information of other avatars. Perform. Considering this category of crawl target information, server-grouping and in requests to mobile devices are performed. Considering the characteristics of the data collection request (urgency, time series characteristics of the information collection target), the management server collects a first sub-group that is already performing a task according to the first data collection request, and a new second data collection. By integrating the second server-group to perform the task according to the request, the task according to the second data collection request is intensively performed for a certain period of time, and after that, it is regrouped into the first and second sub-groups and each Control to perform tasks. For example, the data that the first sub-group wants to collect is when the need for real-time collection is below a preset value (e.g., not real-time collection, but "1-day cumulative data collection: accumulation for 1 day from 22:00 to 24:00 every day). Since "data aggregation" is a request (if the operating time required for data collection is 1/12 per day), it is possible to temporarily utilize the first sub-group to collect other data. At this time, the first sub-group is incorporated into the integrated group that will perform the task according to the new second data collection request, and performs the second data collection during the preset period, and collects the original first data during the other periods. Perform tasks according to request.

The mobile device that receives the data collection request accesses target websites to retrieve information, and the information is delivered to the original requester (crawler server or general user). The management server manages the registration status of mobile devices entering and leaving the proxy server, and monitors the communication status of each mobile device as described above. When performing registration management of a proxy server, the management server operates and manages a reserve group when a new mobile device is incorporated into the group. At this time, the management server must monitor the communication status of each mobile device within the proxy server and transmit a data collection request to designate the mobile device that will perform the task, and calculates the computational load required for monitoring and designation. Monitor together. As the number of mobile devices subject to monitoring increases, the management server requires computational load for monitoring and designation. Therefore, when permitting new mobile devices to be incorporated into the group, the computational load required for monitoring and designation is considered. Decide when to grant permission. Based on the calculation load required for monitoring and specifying the communication status of mobile devices in the current proxy server as 100, it was determined that when n newly incorporated mobile devices are added, the computational load required for monitoring and specifying the communication status will be 120. In this case, the increase in computational load is 20%. At this time, considering the increase or decrease in data collection requests currently being received, if it is necessary to allow n newly incorporated mobile devices to be immediately incorporated into the group despite the increase in computational load, n mobile devices must be immediately incorporated into the group. incorporate. On the other hand, considering the increase/decrease in data collection requests currently being received, if it is determined that it is not necessary to incorporate n new mobile devices at this time, a reserve group can be operated/managed to meet the needs of the future. Prepare to be immediately incorporated into the proxy server group. The management server manages groups in multiple layers even within the proxy server, such as a first group that prioritizes the task of collecting data for the target website, and data collection request characteristics (amount of data to be collected, It is desirable to manage the proxy server efficiently by managing the group by stratifying it into a second group that is provisionally prepared in consideration of the collection period, etc., and the aforementioned preliminary group when incorporating a new device.

Figure 4 is a block diagram showing a computer system for implementing a method according to an embodiment of the present invention.

Referring to FIG. 4, the computer system 1000 includes a processor 1010, a memory 1030, an input interface device 1050, an output interface device 1060, and a storage device 1040 that communicate through a bus 1070. ) may include at least one of Computer system 1000 may also include a communication device 1020 coupled to a network. The processor 1010 may be a central processing unit (CPU) or a semiconductor device that executes instructions stored in the memory 1030 or the storage device 1040. Memory 1030 and storage device 1040 may include various types of volatile or non-volatile storage media. For example, memory may include read only memory (ROM) and random access memory (RAM). In embodiments of the present disclosure, the memory may be located inside or outside the processor, and the memory may be connected to the processor through various known means. Memory is various forms of volatile or non-volatile storage media, for example, memory may include read-only memory (ROM) or random access memory (RAM).

Accordingly, embodiments of the present invention may be implemented as a computer-implemented method or as a non-transitory computer-readable medium storing computer-executable instructions. In one embodiment, when executed by a processor, computer readable instructions may perform a method according to at least one aspect of the present disclosure.

The communication device 1020 can transmit or receive wired signals or wireless signals.

Additionally, the method according to an embodiment of the present invention may be implemented in the form of program instructions that can be executed through various computer means and recorded on a computer-readable medium.

The computer-readable medium may include program instructions, data files, data structures, etc., singly or in combination. Program instructions recorded on the computer-readable medium may be specially designed and configured for embodiments of the present invention, or may be known and usable by those skilled in the art of computer software. A computer-readable recording medium may include a hardware device configured to store and perform program instructions. For example, computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes, optical media such as CD-ROMs and DVDs, and floptical disks. It may be the same magneto-optical media, ROM, RAM, flash memory, etc. Program instructions may include not only machine language code such as that created by a compiler, but also high-level language code that can be executed by a computer through an interpreter, etc.

Although the embodiments of the present invention have been described in detail above, the scope of the present invention is not limited thereto, and various modifications and improvements made by those skilled in the art using the basic concept of the present invention defined in the following claims are also possible. It falls within the scope of rights.

Figure 5 shows learning management using learner back data according to an embodiment of the present invention.

According to an embodiment of the present invention, as learning progresses using learner back data, learning-related information is accumulated and provided as avatar and student information in virtual space.

In the case of Learner Back Data, it is information related to academic studies to which the present invention is applied, and data from academic information, extracurricular subjects, employment linkage, interviews, and counseling are reflected.

Academic information includes subjects completed by the learner, credits, curricular/non-curricular activities, and volunteer activity information. Extracurricular information includes diagnosed personal competencies, extracurricular program activities, career development status, and the contents of the learner's portfolio.

Employment-linked information includes information on companies the learner is interested in, the learner's resume/self-introduction, information on companies the learner hopes for or has applied to, and internship/practice information.

Interview and counseling information includes advisor counseling, individual counseling, group testing, and psychological testing.

Learner-based data is operated in connection with contents such as undergraduate DB/curricular and non-curricular/field training/volunteer activities/E-Class/career development/Legacy, etc. within the school, and external employment, start-up information/internship information/video lectures and It is operated in connection with the same contents.

According to an embodiment of the present invention, appropriate activities are recommended based on the learner's academic information and career direction data through linking the learner's back data and accumulated related information. It recommends appropriate lectures based on the learner's data, provides a system to exclude lectures that do not match from the list, and recommends curricular/non-curricular activities related to employment and entrepreneurship. According to an embodiment of the present invention, efficient learning is achieved by analyzing each learner's learning pattern based on the learner's back data, accumulated related information, and concentration analysis, and planning and providing a customized learning course. Based on the information, the learner's learning type (habits, behavior, etc.) is analyzed and an appropriate learning process is recommended and provided to the learner.

According to an embodiment of the present invention, the learner's type is calculated using the learner's back data, accumulated learner information, and analyzed concentration information, and the learning pattern analysis and learning-related information are recommended.

According to an embodiment of the present invention, information beneficial to the learner's future employment and start-up is provided in relation to employment-related content among the learner's back data. For example, learner A who wants to start a restaurant business is provided with cooking-related certification information (exam schedule, type of certification, etc.) or information about expert lectures, cooking-related competitions, and external activities.

According to an embodiment of the present invention, recommended lecture information is provided to the learner in relation to the learner's academic information, and the grades of learner B majoring in education are analyzed to select lectures with particularly superior grades. If the special education-related credits are high, lectures related to special education are recommended, and in the case of non-curricular subjects, lectures and books are recommended.

Claims (10)

An image acquisition unit that captures the learner's video and transmits video data;
A learner terminal that receives and displays virtual lecture space information created using the video data;
a learning management server that generates the virtual lecture space information using the video data and analyzes the learner's concentration; and
Includes an instructor terminal connected to the learning management server to deliver lecture content,
The learning management server provides behavior correction guidelines to the learner terminal based on the level of concentration. However, if the behavior of a specific learner is determined to be low in concentration as a result of video analysis but is actually well concentrated as a result of completing the quest, the specific learner's behavior is determined to be good concentration. It is decided whether or not to provide the behavior correction guidelines in consideration of the positive index related to the behavior, and the point at which the behavior correction guidelines are provided according to the results of monitoring whether the specific learner's behavior is a one-time behavior or a habitual behavior. control,
The learning management server provides a quest by adjusting the quest settings differently based on the concentration, receives the quest performance result of the learner terminal, and verifies the concentration, but adjusts the quest setting information according to the purpose. If the purpose of the evaluation is to enable students who concentrated on the class to receive a better evaluation, not only the basic learning content but also the probability of getting the correct answer by students who concentrated on the class will increase. Quests are set to include quizzes that include questions that are higher than the probability of students answering correctly, and even if students are unable to concentrate in class, if the purpose is to improve understanding of class content rather than evaluation, the student should focus. For concepts explained at a time that could not be explained, the quest is set to include questions to check whether the concept is understood along with additional explanatory materials, and for students who were highly focused in class, advanced questions are more difficult than the basic questions. It provides a quest that includes, and the reward for successfully performing the quest is also higher than when performing a quest that includes a problem with lower difficulty than the advanced problem, and even if the problem matches the same answer, Depending on the level of class concentration, the quest settings are adjusted differently to determine whether to provide quests in a multiple-choice, subjective, or narrative format.
The learning management server updates the concentration analysis model using the results of the verification of concentration, but analyzes the update results for each learner's individual concentration analysis model to ensure that the update is limited to an update of the individual concentration analysis model. or whether to reflect it in the general concentration analysis model update.
A learning management system in a virtual lecture space using learner-digital twin creation.
According to paragraph 1,
The learning management server receives the image data from the image acquisition unit, extracts facial feature points and upper body feature points, and measures the learner's concentration by comprehensively considering face angle, gaze direction, and upper body posture.
A learning management system in a virtual lecture space using learner-digital twin creation.
delete delete According to paragraph 1,
The learning management server performs an update on the concentration analysis model using the results of verification of the concentration.
A learning management system in a virtual lecture space using learner-digital twin creation. delete delete delete delete delete

Images