KR20200056760A - System for evaluating educators and improving the educational achievement of the trainees using artificial intelligence and method thereof - Google Patents

Abstract

The present invention relates to a system and a method for evaluating an educator and improving an educatee′s educational achievement by using artificial intelligence. The system includes: a camera module imaging a predetermined offline space and generating video data; and a server evaluating the degree of guidance execution of the educator by performing image processing on the video data. The server includes: an educator detection unit detecting the educator′s feature based on the video data; an educatee detection unit defining a non-educator person on the predetermined offline space as the educatee based on the video data; a guidance execution detection unit detecting the feature of the guidance execution where the detected educator guides the defined educatee based on the video data; a question action detection unit detecting the feature of a question action of questioning the detected educator out of the defined educatee based on the video data; and a guidance execution evaluation unit evaluating the degree of the educator′s guidance execution based on at least one among the educator′s feature, the feature of the guidance execution, and the feature of the question action.

Description

A system and method for evaluating educators using artificial intelligence and improving the educator's educational achievements.

The present invention relates to a system and method for evaluating an educator using artificial intelligence and improving the educational achievement of an educated person.

Due to the rapid development and spread of Internet technology, in recent years, in addition to the education method in an offline space, there is an increasing demand for online video lectures where students who are educated can learn in an online space. Students taking online video lessons usually learn in situations where educators are not at their side. Therefore, online video lecture students had a problem in maintaining their concentration more than a certain level during the lecture. In addition, online video lecture students were unable to ask the educator even if they did not know anything during the lecture, so it was difficult to proceed with the lecture, or there was a problem that they could not understand the contents of the lecture.

In addition, the number of students who want to learn the programming language is increasing. Program language is an area in which the quality of instruction varies depending on the ability of educators and the degree of students' perception of learning can also vary. However, an educator with excellent ability has a problem in that the cost of taking a class is high or difficult to hire.

Publication Patent Publication No. 10-2000-0054038, 2000.09.14

The problem to be solved by the present invention is to provide a system and method for evaluating an educator using artificial intelligence and improving the educational achievement of an educated person.

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

The system for evaluating an educator using artificial intelligence according to an aspect of the present invention for solving the above-mentioned problems is a camera module that generates video data by photographing a predetermined offline space, and an educator's guidance by image processing the video data And a server for evaluating the degree of performance, wherein the server comprises an educator detection unit that detects a feature point of the educator based on the video data, and a person other than the educator on the predetermined offline space based on the video data. An educator detection unit defined as an educator, based on the video data, an instructional performance detection unit configured to detect a feature point of an instruction performed by the detected educator to guide the defined educator, based on the video data, the Based on at least one of the feature of the educator, the feature of the educator, the feature of the instruction, and the feature of the questioning behavior, the evaluating degree of the educator is evaluated based on at least one of the feature of the educator asking the detected educator to question the feature. It includes the guidance performance evaluation department.

In some embodiments, the server further includes an educator database for managing and storing one or more feature points of the educator, and the educator detection unit determines a person object standing and moving among the object objects on the predetermined offline space as the educator. Alternatively, the similarity between the characteristic point of the educator on the educator database and the characteristic point of the human object on the predetermined offline space is determined to determine the human object as the educator, and the characteristic point of the educator is detected.

In some embodiments, the instruction performing detection unit determines that the educator performs the instruction for the educator when the educator and the educator are within a predetermined distance or more for a predetermined period of time, and determines a feature point of the instruction execution. To detect.

In some embodiments, the instruction performing detection unit detects the center point of the educator and the center point of the trainee, and for a predetermined time or more, when the center point of the educator and the center point of the trainee are less than a predetermined distance, the educator It judges that the trainee performs the instruction, and detects a feature point of the instruction.

In some embodiments, when the question behavior detecting unit maintains a question attitude of one of the predetermined one or more question attitudes among the trainees for a predetermined time or more, it is determined that the trainee acts as a question, and the question behavior Feature points are detected.

In some embodiments, the server further includes an educator evaluation unit, and the educator evaluation unit evaluates the educator based on one or more of the teaching performance, and transmits the evaluation result of the educator to the terminal of the educator manager do.

A method for evaluating an educator using artificial intelligence according to another aspect of the present invention for solving the above-described problem is a method realized by a computer, generating video data by photographing a predetermined offline space, and the video data Based on the step of detecting a feature point of the educator, based on the video data, defining a person other than the educator as the trainee on the predetermined offline space, based on the video data, the detected educator Detecting a feature point of the instruction performance for guiding the defined educator, detecting a feature point of a question act of asking the detected educator among the defined educators based on the video data, and the feature point of the educator, And evaluating the degree of the teaching performance of the educator based on at least one of the characteristic points of the teaching performance and the characteristic points of the questioning behavior.

In order to solve the above-mentioned problems, an auxiliary system for improving educational achievement using artificial intelligence according to another aspect of the present invention includes a lecture analysis unit that analyzes a result of a video lecture taken by a trainee in a predetermined online space, the predetermined Based on at least one of the coding analysis unit for analyzing the result of the execution of the coding problem corresponding to the video lecture of the trainee in the online space of the, at least one of the performance of the analyzed video lecture and the result of the analysis of the coding problem A map based on at least one of a hint list generation unit that calculates a need to provide and generates a hint list when the need to provide a hint exceeds a threshold, a log file of the analyzed video lecture, and the analyzed coding Calculate the necessity of performance, and if the calculated necessity of performing the map exceeds a threshold, transmit a closeness map need message generation unit that generates a closeness map need message, and the hint list to the terminal of the trainee, And a communication unit transmitting the instruction execution notification message to the educator's terminal.

In some embodiments, the coding analysis unit, the number of compilations of the coding performed on the predetermined online space, the number of compilation errors, the algorithm design time, the algorithm design evaluation result, the coding execution time, the number of questions / answers for a given problem, a given Based on at least one of the correct answer to the problem, the result of coding, the size of the memory to which the code is allocated, and the presence or absence of the correct answer of the code, a result of performing the coding problem corresponding to the video lecture is analyzed.

In some embodiments, the lecture analysis unit, based on at least one of the playback date, playback number, average playback completion rate, full screen button clicks, playback bar clicks, and loading time of the video lecture, the result of performing the video lecture Analyzes.

In some embodiments, the lecture analysis unit analyzes a result of the video lecture based on a quiz result performed during the video lecture.

An auxiliary method for improving educational achievement using artificial intelligence according to another aspect of the present invention for solving the above-described problems is a method realized by a computer, and is a result of performing a video lecture taken by an educator in a predetermined online space. Analyzing the result of performing a coding problem corresponding to the video lecture of the trainee in the predetermined online space, at least one of a result of performing the analyzed video lecture and a result of performing the analyzed coding problem Based on the, calculating the need to provide a hint and generating a list of hints when the need to provide a hint exceeds a threshold, based on at least one of the analyzed video lecture log file and the analyzed coding , Calculating the necessity of performing the instruction, generating a close instructional message when the calculated necessity of performing the instruction exceeds a threshold value, and transmitting the hint list to the terminal of the trainee, and performing the instruction And transmitting a notification message to the educator's terminal.

Other specific matters of the present invention are included in the detailed description and drawings.

According to a system and method for evaluating an educator using artificial intelligence of the present invention, the video data of a predetermined offline space is image-processed to evaluate the educator's instructional performance, and corresponds to the educator's administrator's instructional performance. Can provide the results of an educator's evaluation.

In addition, the instructional performance evaluation unit may determine the degree of instructional performance of a plurality of items, and provide an educator or an educator's manager with what item of instructional performance should be improved for an educator to perform high-level instruction.

In addition, according to the auxiliary system and method for improving the educational achievement using the artificial intelligence of the present invention, the video lectures taken by the trainees are analyzed, and the coding results created by the trainees are analyzed to automatically calculate whether the trainees need to provide hints. can do. When the calculated need for providing a hint exceeds a threshold, a list of hints can be automatically generated and provided to the trainee.

In addition, the video lectures taken by the trainees may be analyzed, and the coding results created by the trainees may be analyzed to automatically calculate the necessity of teaching the trainees. When the calculated necessity of conducting the instruction exceeds a threshold, an educated instructional necessity message is automatically generated and provided to the educator so that the educator can instruct the educated person closely.

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

1 is a schematic configuration diagram of a system for evaluating an educator using artificial intelligence according to an embodiment of the present invention.
2 is a schematic configuration diagram of a server in a system for evaluating educators using artificial intelligence.
3 is an exemplary view of an educator and an educator image detected by an educator detection unit and an educator detection unit of a system for evaluating educators using artificial intelligence.
4 is an exemplary view of a map performance image detected by a map performance detection unit of a system for evaluating educators using artificial intelligence.
5 is an exemplary view of a question behavior image detected by a question behavior detection unit of a system for evaluating educators using artificial intelligence.
6 is a schematic flowchart of a method for evaluating an educator using artificial intelligence according to another embodiment of the present invention.
7 is a schematic configuration diagram of an auxiliary system for improving educational achievement using artificial intelligence according to another embodiment of the present invention.
8 is an exemplary view of a hint list generated by a hint list generation unit of an auxiliary system for improving educational achievement using artificial intelligence.
9 is an exemplary diagram of a message requiring a close map generated by a message generating unit requiring a close map of an auxiliary system for improving educational achievement using artificial intelligence.
10 is a schematic flowchart of an auxiliary method for improving educational achievement using artificial intelligence according to another embodiment of the present invention.

Advantages and features of the present invention, and methods for achieving them will be clarified with reference to embodiments described below in detail together with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, but may be implemented in various different forms, and only the present embodiments allow the disclosure of the present invention to be complete, and are common in the technical field to which the present invention pertains. It is provided to fully inform the skilled person of the scope of the present invention, and the present invention is only defined by the scope of the claims.

The terminology used herein is for describing the embodiments and is not intended to limit the present invention. In the present specification, the singular form also includes the plural form unless otherwise specified in the phrase. As used herein, “comprises” and / or “comprising” does not exclude the presence or addition of one or more other components other than the components mentioned. Throughout the specification, the same reference numerals refer to the same components, and “and / or” includes each and every combination of one or more of the components mentioned. Although "first", "second", etc. are used to describe various components, it goes without saying that these components are not limited by these terms. These terms are only used to distinguish one component from another component. Therefore, it goes without saying that the first component mentioned below may be the second component within the technical spirit of the present invention.

Unless otherwise defined, all terms (including technical and scientific terms) used in the present specification may be used as meanings commonly understood by those skilled in the art to which the present invention pertains. In addition, terms defined in the commonly used dictionary are not ideally or excessively interpreted unless explicitly defined.

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

1 is a schematic configuration diagram of a system for evaluating an educator using artificial intelligence according to an embodiment of the present invention.

Referring to FIG. 1, a system for evaluating an educator using artificial intelligence includes a camera module 100 and a server 200.

The camera module 100 photographs a predetermined offline space to generate video data. The camera module 100 is installed on top of a predetermined offline space (for example, a lecture hall, etc.) to photograph a person or object in the offline space. The camera module 100 transmits the captured video data to the server 200.

The server 200 evaluates the educator's teaching performance by processing video data. The server 200 image-processes the video data received from the camera module 100 to evaluate the educator's degree of teaching performance on a predetermined offline space.

2 is a schematic configuration diagram of a server in a system for evaluating educators using artificial intelligence.

Referring to FIG. 2, the server 200 of the system for evaluating educators using artificial intelligence includes an educator detection unit 210, an educator detection unit 220, an instruction execution detection unit 230, a question behavior detection unit 240, and an instruction execution It includes an evaluation unit 250, an educator database, and an educator evaluation unit 270.

The pose data detector (not shown) is based on the video data of a predetermined offline space, and the face (eye, nose, mouth, ear, face shape), neck, shoulder, elbow, wrist of a human object on the predetermined offline space , Detects joints such as fingers, pelvis, and knees, and creates connecting lines connecting each joint. The educator detection unit 210, the trainee detection unit 220, the instruction performance detection unit 230, the object center point detection unit (not shown), the question behavior detection unit 240, and the instruction performance evaluation unit 250 described below are the pose data detection unit ( On the basis of the connection line generated in (not shown), each feature point may be detected, or each object may be defined. The feature points may be the shape of the face (eyes, nose, mouth, face contours, etc.), the shape of the body (dry body, tall body, etc.), facial expressions (smiling faces, faces with frowns), body It can be an action (person standing, person sitting, etc.).

The educator detection unit 210 detects a feature point of the educator based on the video data. The educator detection unit 210 determines a person object standing and moving among the person objects on a predetermined offline space as an educator. Thereafter, the educator detection unit 210 detects a feature point for the determined educator.

In addition, the educator detection unit 210 may determine at least one human object as an educator among human objects on a predetermined offline space, and annotate the educator as an educator.

Annotation processing refers to the task of generating data describing the judged educator. Annotation processing includes labeling, coloring or layering the educator. In addition, any indication of who the educator is can be included as annotation processing. For example, the educator on the video data may be displayed by dividing the area into a box or the like. Annotation processing can take advantage of automated processing tools. For areas marked as educators, labeling can be written in one word, and can be written in detail rather than in one word. The educators, guidance, and questioning behaviors mentioned below can also be annotated in the same way.

When there are two or more educators in a predetermined offline space, the educator detection unit 210 may determine each educator and detect the determined characteristic point of the educator. The educator detection unit 210 may perform annotation processing for each of two or more educators. The educator detection unit 210 may annotate two or more educators as main educators and sub-educators, and annotate educators as equivalent educators.

The educator detection unit 220 defines a person other than an educator on a predetermined offline space as an educated person based on video data. The educator detection unit 220 defines a person other than the educator detected by the educator detection unit 210 on a predetermined offline space as the educated person based on the video data. The trainee detection unit 220 may detect a defined feature point of the trainee.

The trainee detection unit 220 may determine a trainee on a predetermined offline space and annotate the determined trainee as an trainee.

The instruction performance detection unit 230 detects a feature point of the instruction performance to map the trainee defined by the detected educator based on the video data.

The instruction performance detection unit 230 determines that the educator conducts the instruction for the educator and detects a feature point of the instruction execution when the educator and the educator are within a predetermined distance for a predetermined time or longer.

The instruction execution detection unit 230 detects the center point of the educator and the center point of the trainee, and if the center point of the educator and the center point of the trainee are less than a predetermined distance for a predetermined time or more, it is determined that the educator conducts the trainee , It detects the feature points of the map performance.

The object center point detection unit (not shown) calculates the x and y coordinates of one or more connection lines recognized in each joint unit for a human object in a predetermined offline space based on video data. The object center point detector (not shown) calculates the average values of the calculated one or more x-coordinates and y-coordinates, and defines the average values of the x- and y-coordinates as the center points of the corresponding object.

In addition, the object center point detection unit (not shown) may define a center of a portion corresponding to the head of a human object on a predetermined offline space as a center point based on video data. For example, the object center point detection unit (not shown) calculates x and y coordinates of both eyes of a human object on a predetermined offline space based on video data. The object center point detector (not shown) may calculate the average values of the calculated two x-coordinates and y-coordinates, and define the average values of the x and y coordinates as the center points of the corresponding object.

The question behavior detection unit 240 detects feature points of the question behavior of asking the detected educator among the defined educators based on the video data.

When the question behavior detecting unit 240 maintains a question attitude of any one of the predetermined one or more question attitudes among the trainees for a predetermined time or more, it determines that the learner acts as a question and detects a feature point of the question behavior. The predetermined question posture may include a posture of raising a hand and the like.

In addition, the question behavior detection unit 240 maintains a question posture of one or more of the predetermined question postures among the trainees for a predetermined time or more, and when the educator conducts instruction to the corresponding trainee, it is determined that the learner acts as a question Then, it is possible to detect the characteristic point of the question behavior.

The instruction performance evaluation unit 250 evaluates the degree of instruction performance by the educator based on at least one of the characteristic point of the educator, the characteristic point of the instruction execution, and the characteristic point of the question behavior.

The degree of guidance performance can be evaluated by a plurality of items. For example, the degree of instructional performance includes the distance that the educator has moved in a predetermined offline space, the number of instructional executions due to the questions of the educator, the number of voluntary instructional executions, the average instructional execution time, the number of instructional executions per educator, It may include the average time taken from conducting questions to conducting guidance.

It is possible to provide the educator's terminal or the educator's terminal with the instructional performance of a plurality of items evaluated by the instructional performance evaluation unit 250. Accordingly, the educator or the manager of the educator can determine which of the plurality of instructional performance items should improve instructional performance to improve the quality of instructional performance provided to the educated person.

The instruction performance evaluation unit 250 may provide a predetermined target value to a terminal of the educator or a terminal of the educator's manager for a plurality of evaluated items. For example, the instruction performance evaluation unit 250 may set the target value to 20 times for the number of times the educator voluntarily performs instruction, and provide it to the terminal of the educator or the terminal of the educator. Therefore, the educator or the manager of the educator may determine that the quality of the class can be improved only if the instruction performed by the educator does not reach the target value by performing the instruction as much as the target value.

The educator database manages and stores the features of one or more educators. The educator database may store and manage the educator's feature points when the educator is registered as an educator in a predetermined online or offline. For example, the educator database 260 may receive a picture of the educator, detect and manage and store the feature points of the educator. In addition, the educator database may detect and characterize the educator's feature points for a person object judged as an educator according to a predetermined rule on video data photographing a predetermined offline space and manage and store it in real time.

The educator detection unit 210 determines a person object as an educator by measuring the similarity between the educator's characteristic point on the educator database and the characteristic point of the human object on a predetermined offline space, and detects the educator's characteristic point. For example, the educator detection unit 210 measures the similarity between the feature points of the educator's face on the educator database and the feature points of the face of the human object on the predetermined offline space. Thereafter, the educator detection unit 210 determines a person object in a predetermined offline space as an educator when the similarity is higher than a predetermined criterion based on the measured similarity.

The educator evaluation unit 270 evaluates the educator based on one or more instruction performance levels, and transmits the educator evaluation results to the educator's manager terminal. Therefore, the manager of the educator can receive the evaluation result of the educator evaluated by the objectified data, and instruct the educator to improve the performance of the instruction.

3 is an exemplary view of an educator and an educator image detected by an educator detection unit and an educator detection unit of a system for evaluating educators using artificial intelligence.

Referring to FIG. 3, the educator detection unit 210 and the educator detection unit 220 of a system for evaluating an educator using artificial intelligence detects feature points of the educator 211 in a predetermined offline space, and trains the educator 221. define.

The educator detection unit 210 is a connecting line connecting the joints of the human object on the video data, and may determine whether the human object is standing or sitting. The educator detection unit 210 may determine the person object as the educator 211 when the specific person object stands for a predetermined time or more. In addition, the educator detection unit may box and annotate the determined educator.

The educator detection unit 220 may determine and define a person object as the educator 221 except for the educator 211 detected by the educator detection unit 210 among a plurality of human objects on video data.

4 is an exemplary view of a map performance image detected by a map performance detection unit of a system for evaluating educators using artificial intelligence.

Referring to FIG. 4, the instruction performing detection unit 230 of the system for evaluating an educator using artificial intelligence detects a feature point of instruction execution by the educator 211 to instruct the educator 221.

The instruction performing detection unit 230 determines that the educator 211 performs the instruction for the educator 221 when the educator 211 and the educator 221 are within a predetermined distance for a predetermined time or longer, and performs the instruction To detect the feature point.

The instruction performing detection unit 230 detects the center point of the educator 211 and the center point of the trainee 221, and for a predetermined time or longer, the center point of the educator 211 and the center point of the trainee 221 are below a predetermined distance. In this case, the educator 211 determines that the educator 221 conducts the instruction, and detects a feature point of the instruction execution. Also, the map performance detection unit 230 may box and annotate the determined map performance.

5 is an exemplary view of a question behavior image detected by a question behavior detection unit of a system for evaluating educators using artificial intelligence.

Referring to FIG. 5, the question behavior detection unit 240 of the system for evaluating educators using artificial intelligence detects feature points of the question behavior 241 that the educator asks.

The question behavior detection unit 240 determines that the educated person conducts the question action 241 when the question posture of one of the predetermined one or more question positions among the trainees is maintained for a predetermined time or longer, and the characteristic points of the question action 241 Can be detected.

6 is a schematic flowchart of a method for evaluating an educator using artificial intelligence according to another embodiment of the present invention.

Referring to FIG. 6, a method of evaluating an educator using artificial intelligence includes: generating video data (S310), detecting a feature point of the educator (S320), defining an educator (S330), and performing instruction It includes the step of detecting the feature point (S340), the step of detecting the feature point of the question behavior (S350) and the step of evaluating the degree of map performance (S360).

The educator's evaluation method using artificial intelligence in FIG. 6 corresponds to the components of the evaluation system described with reference to FIGS. 1 to 5. Hereinafter, in describing the method, duplicate description of components of the evaluation system of FIGS. 1 to 5 will be omitted.

In step S310, the camera module 100 photographs a predetermined offline space to generate video data.

In step S320, the educator detection unit 210 detects a feature point of the educator based on the video data.

In step S330, the educator detecting unit 220 defines a person other than the educator on a predetermined offline space as the educated person based on the video data.

In step S340, the instruction execution detection unit 230 detects a feature point of instruction execution to instruct the trainee defined by the detected educator based on the video data.

In step S350, the question action detection unit 240 detects a feature point of the question action that asks the detected educator among the defined trainees based on the video data.

In step S360, the instruction performance evaluation unit 250 evaluates the degree of instruction performance of the educator based on at least one of the characteristic point of the educator, the characteristic point of the instruction execution, and the characteristic point of the question behavior.

7 is a schematic configuration diagram of an auxiliary system for improving educational achievement using artificial intelligence according to another embodiment of the present invention.

Referring to FIG. 7, an auxiliary system for improving educational achievement using artificial intelligence includes a lecture analysis unit 410, a coding analysis unit 420, a hint list generation unit 430, and a message guidance generation unit 440 for close instructional guidance. And a communication unit 450.

The lecture analysis unit 410 analyzes a result of performing a video lecture taken by the trainee in a predetermined online space. The lecture analysis unit 410 analyzes log data of the video lectures taken by the trainees in a predetermined online space, and analyzes the results of the video lectures performed by the trainees using the analyzed log data. For example, the lecture analysis unit 410 may display the result of performing the video lecture based on at least one of a video lecture play date, a play count, an average play completion rate, a full screen button click count, a play bar click count, and a loading time. Analyze.

The lecture analysis unit 410 analyzes the results of the video lecture based on the quiz result performed during the video lecture. The quiz may be provided to the trainee in the middle of the video lecture, and may be provided to the trainee after the video lecture ends. Therefore, the quiz provided to the trainee during the video lecture may serve to check whether the trainee is watching the video lecture well, or to check whether the lecture is understood after the lecture ends.

The coding analysis unit 420 analyzes a result of performing a coding problem corresponding to a video lecture of the trainee in a predetermined online space.

The coding analysis unit 420 compiles the number of codings performed on a predetermined online space, the number of compile errors, the algorithm design time, the algorithm design evaluation result, the coding execution time, the number of questions / answers for a given problem, and the correct answer for a given problem Based on at least one of whether or not the result of coding, the size of the memory to which the code is allocated, and whether or not the correct answer of the code is present, a result of performing the coding problem corresponding to the video lecture is analyzed.

The educator performs coding on a predetermined online space, and the coding analysis unit 420 may count the number of times compilation is performed for the corresponding coding. Also, the coding analysis unit 420 may count the number of compile errors when the trainee compiles.

The coding analysis unit 420 may provide a coding problem to the trainee in a predetermined online space and provide an algorithm for the problem. The coding analysis unit 420 may measure the time for the trainee to design a coding problem with the provided algorithm. In addition, the coding analysis unit 420 may evaluate coding designed by the provided algorithm according to a predetermined rule, and score results for the designed coding.

The coding analysis unit 420 may give a coding problem to an educated person in a predetermined online space, and measure an execution time for the educator to code for the coding problem. Also, the coding analysis unit 420 may determine whether a correct answer or incorrect answer is given to the coding problem. In addition, the coding analysis unit 420 may count the number of times that the educator questions or answers the coding analysis unit 420 while coding the coding problem.

The coding analysis unit 420 may give a coding problem to the trainee in a predetermined online space, and determine a coding result submitted by the trainee. The coding result can determine not only the correct answer to the coding problem, but also the grade of coding submitted by the algorithm, library, and function used. Also, the coding analysis unit 420 may determine the grade of the submitted coding by the allocated memory size of the coding submitted by the trainee. Therefore, the coding analysis unit 420 may determine that the smaller the allocated memory size of coding, the higher the grade of the submitted coding.

The hint list generating unit 430 calculates the need to provide a hint based on at least one of the result of performing the analyzed video lecture and the result of performing the analyzed coding problem, and when the need to provide the hint exceeds a threshold value, the hint Create a list.

The hint list may be provided in the form of a help or question list. The hint list generation unit 430 may generate a list of help or questions corresponding to the content and coding problems corresponding to the video lecture that the educator is listening to according to a predetermined rule. The hint list generator 430 may load a list of help and questions stored in the hint list database and provide them to the trainee.

The close instruction map request message generation unit 440 calculates the need for performing the map based on at least one of the analyzed video lecture log file and the analyzed coding, and the calculated need for performing the map exceeds a threshold value. In this case, a message requiring a close map is generated.

The message that needs close instruction is a message indicating the degree to which the trainee needs close instruction. Messages requiring a close map may be displayed in percentage, text, or color, but are not limited thereto. For example, a message requiring a close map can be displayed as a percentage of how close the map is required, can be displayed as a text that a close map is required, and if a close map is required, the management window of the corresponding student boxed in red It can be displayed as.

The communication unit 450 transmits a hint list to the terminal of the trainee. The communication unit 450 transmits the hint list generated by the hint list generation unit 430 to the trainee's terminal so that the trainee can check the hint list online. Accordingly, when a difficulty arises while an educator performs video lecture or coding in an online space, the hint list generation unit 430 may automatically provide the hint list to the educator through the communication unit 450.

In addition, the communication unit 450 transmits a notification message for performing the instruction to the terminal of the educator. The communication unit 450 may automatically provide a map execution notification message corresponding to the content of the contact map required message generated by the contact map required message generation unit 440 to the educator's terminal.

8 is an exemplary view of a hint list generated by a hint list generation unit of an auxiliary system for improving educational achievement using artificial intelligence.

Referring to FIG. 8, the hint list generation unit 430 of the auxiliary system for improving educational achievement generates a hint list.

The hint list generation unit 430 provides a hint list to the terminal of the trainee in the form of a help or a list of questions. The hint list generation unit 430 may provide a hint list 431 related to a video lecture or coding problem to the hint box 431. In addition, the trainee may input a question in the question box 432 when there is a question in the video lecture or coding problem. The hint list generation unit 430 may provide a list of hints corresponding to the question of the trainee to the terminal of the trainee.

9 is an exemplary diagram of a message requiring a close map generated by a message generating unit requiring close maps of an auxiliary system for improving educational achievement using artificial intelligence.

Referring to FIG. 9, the message generating unit 440 of the instruction for close instruction of the auxiliary system for improving the educational achievement using artificial intelligence may provide the message for the instruction of instruction to the terminal of the educator.

The message generation unit 440 that needs close instruction may determine the degree of learning recognition of the trainee based on at least one of the analyzed video lecture log file and the analyzed coding. The contact instruction-needing message generation unit 440 may provide an instructional instruction-addressing message to the educator's terminal so that the educator can instruct the educator closely. For example, an educator in an offline space may receive a 95% close instructional message for a specific educator through the educator's terminal, and an educator who receives the message may determine that the educator needs close instruction. Subsequently, the educator may move to the position of the educated person and individually and intensively conduct instruction regarding the part, the vulnerable part, and the part lacking in solving the coding problem of the educated person.

Therefore, the educator can immediately see and recognize the terminal of the educator that there is a part suffering from a situation in which the specific educator codes.

10 is a schematic flowchart of an auxiliary method for improving educational achievement using artificial intelligence according to another embodiment of the present invention.

Referring to FIG. 10, an assistive method for improving educational achievement using artificial intelligence includes: analyzing a result of performing a video lecture (S510), analyzing a result of performing a coding problem (S520), and generating a hint list It includes a step (S530), a step of generating a close contact map message (S540) and a step of transmitting a map execution notification message (S550).

The auxiliary method for improving educational achievement using artificial intelligence of FIG. 10 corresponds to the components of the auxiliary system described with reference to FIGS. 7 to 9. Hereinafter, in describing the method, duplicate description of components of the auxiliary systems of FIGS. 7 to 9 will be omitted.

In step S510, the lecture analysis unit 410 analyzes the result of performing the video lecture taken by the trainee in a predetermined online space.

In step S520, the coding analysis unit 420 analyzes the result of performing the coding problem corresponding to the video lecture of the trainee in a predetermined online space.

In step S530, the hint list generation unit 430 calculates the need to provide a hint based on at least one of the result of performing the analyzed video lecture and the result of performing the analyzed coding problem, and the need to provide a hint determines a threshold value. If exceeded, a hint list is generated.

In step S540, the close map necessary message generation unit 440 calculates the need for performing the map based on at least one of the analyzed video lecture log file and the analyzed coding, and the calculated need for performing the map is critical If the value is exceeded, a message requiring a close map is generated.

In step S550, the communication unit 450 transmits a hint list to the terminal of the trainee, and transmits a notification message for performing the instruction to the terminal of the educator.

The steps of a method or algorithm described in connection with an embodiment of the present invention may be implemented directly in hardware, a software module executed by hardware, or a combination thereof. The software modules may include Random Access Memory (RAM), Read Only Memory (ROM), Erasable Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), Flash Memory, Hard Disk, Removable Disk, CD-ROM, or It may reside on any type of computer readable recording medium well known in the art.

The embodiments of the present invention have been described above with reference to the accompanying drawings, but a person skilled in the art to which the present invention pertains may implement the present invention in other specific forms without changing its technical spirit or essential features. You will understand. Therefore, it should be understood that the above-described embodiments are illustrative in all respects and not restrictive.

100: camera module
200: server

Claims (12)

A camera module that photographs a predetermined offline space to generate video data; And
Includes a server that processes the video data to evaluate the educator's teaching performance
The server,
An educator detector configured to detect a feature point of the educator based on the video data;
An educator detecting unit defining a person other than the educator as the educator on the basis of the video data;
A map performance detection unit based on the video data, the detected educator detects a feature point of a map performance for guiding the defined trainee;
A question action detection unit that detects a feature point of a question action questioning the detected educator among the defined trainees based on the video data; And
And an instructional performance evaluation unit evaluating the degree of instructional performance of the educator based on at least one of the characteristic point of the educator, the characteristic point of the instruction execution, and the characteristic point of the question behavior,
System for evaluating educators using artificial intelligence. According to claim 1,
The server further includes an educator database for managing and storing one or more feature points of the educator,
The educator detection unit,
The human object standing and moving among the human objects in the predetermined offline space is determined as the educator, or the similarity between the characteristic point of the educator in the educator database and the characteristic point of the human object in the predetermined offline space is measured. Judging by the educator, detecting the feature point of the educator,
System for evaluating educators using artificial intelligence. According to claim 1,
The map performance detection unit,
If the educator and the trainee are within a predetermined distance or more for a predetermined period of time, the educator determines that the educator is teaching, and detects a feature point of the teaching,
System for evaluating educators using artificial intelligence. According to claim 3,
The map performance detection unit,
The center point of the educator and the center point of the trainee are detected, and for a predetermined time or more, when the center point of the educator and the center point of the trainee are less than or equal to a predetermined distance, it is determined that the educator conducts the trainee, Detecting a feature point of the map execution,
System for evaluating educators using artificial intelligence. According to claim 1,
The question behavior detection unit,
If one of the pre-determined at least one question posture is maintained for a predetermined time or more, it is determined that the educator is in question and detects a feature point of the question,
System for evaluating educators using artificial intelligence. According to claim 1,
The server further includes an educator evaluation unit,
The educator evaluation unit,
Evaluating the educator and transmitting the evaluation result of the educator to a terminal of the educator's manager, based on one or more of the teaching performance levels,
System for evaluating educators using artificial intelligence. As a method realized by computer,
Generating video data by photographing a predetermined offline space;
Detecting a feature point of the educator based on the video data;
Defining a person other than the educator on the predetermined offline space as an educated person based on the video data;
Detecting, based on the video data, the detected educator's feature point of conducting an instruction to lead the defined educator;
Based on the video data, detecting a feature point of a question behavior of asking the detected educator among the defined educators; And
Evaluating the educator's degree of instruction performance based on at least one of the educator's feature point, the instruction performance characteristic point, and the question behavior characteristic point,
How to evaluate educators using artificial intelligence. A lecture analysis unit that analyzes a result of a video lecture taken by a trainee in a predetermined online space;
A coding analysis unit for analyzing a result of performing a coding problem corresponding to the video lecture of the trainee in the predetermined online space;
A hint for calculating a need to provide a hint and generating a hint list when the need to provide a hint exceeds a threshold based on at least one of the result of performing the analyzed video lecture and the result of performing the analyzed coding problem List generator;
Based on at least one of the analyzed video lecture log file and the analyzed coding, a necessity of performing the map is calculated, and a need for a close map is generated when the calculated necessity of performing the map exceeds a threshold value. A message generating unit that needs a close contact map; And
And a communication unit transmitting the hint list to the terminal of the trainee, and transmitting the instruction execution notification message to the terminal of the educator,
Auxiliary system to improve educational achievement using artificial intelligence. The method of claim 8,
The coding analysis unit,
The number of compilations of the coding performed on the predetermined online space, the number of compilation errors, the algorithm design time, the algorithm design evaluation result, the coding execution time, the number of questions / answers for the given problem, the correct answer for the given problem, the coding execution result, Analyzing a result of performing a coding problem corresponding to the video lecture, based on at least one of a size of a memory to which a code is allocated and a presence or absence of a correct answer of the code,
Auxiliary system to improve educational achievement using artificial intelligence. The method of claim 8,
The lecture analysis unit,
Analyzing the result of performing the video lecture based on at least one of the playback date, the number of playback times, the average playback completion rate, the full screen button click count, the playback bar click count, and the loading time of the video lecture,
Auxiliary system to improve educational achievement using artificial intelligence. The method of claim 8,
The lecture analysis unit,
Based on the quiz results performed during the video lecture, analyzing the results of the video lecture,
Auxiliary system to improve educational achievement using artificial intelligence. As a method realized by computer,
Analyzing a result of performing a video lecture taken by a trainee in a predetermined online space;
Analyzing a result of performing a coding problem corresponding to the video lecture of the trainee in the predetermined online space;
Calculating a need to provide a hint based on at least one of a result of performing the analyzed video lecture and a result of performing the analyzed coding problem, and generating a hint list when the need to provide a hint exceeds a threshold value ;
Based on at least one of the analyzed video lecture log file and the analyzed coding, a necessity of performing the map is calculated, and a need for a close map is generated when the calculated necessity of performing the map exceeds a threshold value. To do; And
Transmitting the hint list to the terminal of the trainee, and transmitting the instruction execution notification message to the terminal of the educator,
Auxiliary methods to improve educational achievement using artificial intelligence.

Images