A learning mentoring method by using an Eye Tracking of learner

Abstract

The present invention enables a video marker corresponding to the learning gaze of a learner studying a video-based learning object to be displayed on the video-based learning object, analyzes the movement pattern of the video marker on the video-based learning object to analyze the learner's learning habits or learning abilities, and provides learning diagnostic mentoring information, which is the analysis result information, to the learner, thereby providing the effect of mentoring regarding learning habits such as reading methods and concentration, or learning abilities such as literacy and problem-solving skills, which are basic competencies of the learner necessary for learning.

Description

Technology Field

[0001] The present invention relates to a learning mentoring method based on learner gaze processing, and more specifically, to a technology for a learning mentoring method in which a video marker corresponding to the learning gaze of a learner studying a video-based learning object is displayed on the video-based learning object, the movement pattern of the video marker on the video-based learning object is analyzed to analyze the learner's learning habits or learning abilities, and learning diagnostic mentoring information, which is the analysis result information, is provided to the learner. Background Technology

[0003] Conventional learning methods include offline and online methods, and due to geographical distance issues and the recent COVID-19 pandemic, the online learning market is growing rapidly.

[0004] Conventional online learning methods involved learners paying for online lectures and purchasing textbooks to take video lectures online.

[0005] These conventional online lectures merely deliver the instructor's one-sided lecture content to the learner, and had the problem that learning mentoring to correct the learner's learning abilities or errors was impossible.

[0006] Recently, an online learning method capable of providing learning mentoring to correct learning errors has been developed. However, in the conventional online learning method capable of providing learning mentoring to correct learning errors, learners solve problems related to the lecture content and receive problem-solving mentoring after sending the results (answer sheets) to the instructor. The learning mentoring for the problems solved by the learners was a method in which the instructor provided only simple grading results to the learners.

[0007] Therefore, in the case of conventional online learning methods capable of providing learning mentoring to correct learning errors, learners are unable to identify the specific errors in incorrect answers. This leads to the repetition of the same mistakes when encountering similar problems in the future. Furthermore, since questions answered by guessing are graded as correct, learning mentoring for such guessed answers was also impossible.

[0008] Meanwhile, although learning habits such as reading methods and concentration, as well as learning abilities such as literacy and problem-solving skills, vary among learners and are the most basic competencies of a learner, there has been a problem in providing mentoring regarding individual learning habits and abilities due to the lack of methods or systems to analyze and mentor them; therefore, it can be said that the development of technology to resolve this issue is necessary.

[0009] Meanwhile, eye-tracking devices have recently been developed. These devices measure eye movements to determine people's interest and attention levels, and are being utilized in various fields, including medical applications such as analyzing text processing characteristics of children with reading disabilities or correcting strabismus patients, as well as precisely analyzing the effectiveness of products displayed in stores, packaging images, and advertising images, evaluating the usability of mobile devices, and measuring the behavior of automobile drivers.

[0010] Meanwhile, basic learning competencies necessary for learning, such as reading methods and study habits like concentration, as well as learning abilities like literacy and problem-solving skills, can all be assessed based on how the learner views the subject matter and engages in learning.

[0011] Therefore, this invention was created by combining an eye-tracking device, which was previously used in a field separate from learning, with the principle that a learner's basic competencies—such as reading methods, concentration, learning habits, literacy, and problem-solving skills—can all be assessed based on how the learner views the learning object.

[0012] The present invention addresses the problem and necessity described above by proposing a technology for a learning mentoring method that displays a video marker corresponding to the learning gaze of a learner studying a video-based learning object on the object, analyzes the movement pattern of the video marker on the object to analyze the learner's learning habits or abilities, and provides the learner with learning diagnostic mentoring information, which is the result of the analysis. The following are prior art related to this. Prior art literature

[0014] 1. Korean Registered Patent Publication No. 10-1180383: System and method for linking online and offline learning 2. Korean Registered Patent Publication No. 10-1652500: Real-time knowledge-based sharing smart learning system and method 3. Korean Registered Patent Publication No. 10-2099943: Learning service system for learners using mentors 4. Korean Registered Patent Publication No. 10-2099945: Learning system using mentors capable of judging the level of a learner's concentration The problem to be solved

[0015] The present invention aims to display a video marker corresponding to the learning gaze of a learner studying a video-based learning object on the video-based learning object, analyze the movement pattern of the video marker on the video-based learning object to analyze the learner's learning habits or learning abilities, and provide learning diagnostic mentoring information, which is the analysis result information, to the learner. means of solving the problem

[0017] The learner eye-tracking-based learning mentoring method of the present invention for solving the above problem is,

[0018] A first step (S100) in which a learning mentee terminal (1) generates learning video information based on learner eye processing regarding the content to be received for learning mentoring;

[0019] A second step (S200) in which the learning mentee terminal (1) transmits learning video information based on learner eye processing to the learning mentor terminal (2) through a communication network;

[0020] A third step (S300) in which a learning mentor terminal (2) generates learning diagnostic mentoring information using learning video information based on learner eye processing;

[0021] The method includes a fourth step (S400) in which the learning mentor terminal (2) transmits learning diagnostic mentoring information to the learning mentee terminal (1) through a communication network, and

[0022] The above-described learner gaze processing-based learning video information is characterized by being video-based learning object video information in which a video marker is displayed that moves along the learning gaze focus of a learner learning a video-based learning object. Effects of the invention

[0024] The present invention enables a video marker corresponding to the learning gaze of a learner studying a video-based learning object to be displayed on the video-based learning object, analyzes the movement pattern of the video marker on the video-based learning object to analyze the learner's learning habits or learning abilities, and provides learning diagnostic mentoring information, which is the analysis result, to the learner, thereby providing the effect of mentoring regarding learning habits such as reading methods and concentration, or learning abilities such as literacy and problem-solving skills, which are basic competencies of the learner necessary for learning. Brief explanation of the drawing

[0026] FIG. 1 is an explanatory diagram of the overall steps of the present invention. FIG. 2 is an explanatory diagram of the first step of the present invention. FIG. 3 is an example of the first pattern of the present invention. FIG. 4 is an example of a second pattern of the present invention. FIG. 5 is an example of the third pattern of the present invention. FIG. 6 is an exemplary diagram of the fourth pattern of the present invention. FIG. 7 is an explanatory diagram of the third step of the present invention. Specific details for implementing the invention

[0027] Embodiments of the present invention will be described in detail with reference to the attached drawings.

[0028] The present invention, a learning mentoring method based on learner eye processing (hereinafter referred to as the present invention), is an invention that provides the effect of providing mentoring on learning habits such as reading methods and concentration, or learning abilities such as literacy and problem-solving skills, which are basic competencies of a learner required for learning, by enabling a video marker corresponding to the learning gaze of a learner learning a video-based learning object to be displayed on the video-based learning object, analyzing the movement pattern of the video marker on the video-based learning object to analyze the learner's learning habits or learning abilities, and providing learning diagnostic mentoring information, which is the analysis result information, to the learner. As illustrated in FIG. 1, the invention is characterized by comprising a first step (S100), a second step (S200), a third step (S300), and a fourth step (S400).

[0030] Specifically, the learner eye-tracking-based learning mentoring method of the present invention, as illustrated in FIG. 1,

[0031] A first step (S100) in which a learning mentee terminal (1) generates learning video information based on learner eye processing regarding the content to be received for learning mentoring;

[0032] A second step (S200) in which the learning mentee terminal (1) transmits learning video information based on learner eye processing to the learning mentor terminal (2) through a communication network;

[0033] A third step (S300) in which a learning mentor terminal (2) generates learning diagnostic mentoring information using learning video information based on learner eye processing;

[0034] The method includes a fourth step (S400) in which the learning mentor terminal (2) transmits learning diagnostic mentoring information to the learning mentee terminal (1) through a communication network, and

[0035] The above-described learner gaze processing-based learning video information is characterized by being video-based learning object video information in which a video marker is displayed that moves along the learning gaze focus of a learner learning a video-based learning object.

[0037] The above first step (S100) is a step of generating learning video information based on learner eye processing regarding the content that the learning mentee terminal (1) wants to receive learning mentoring from.

[0038] The above-mentioned learning mentee terminal (1) is a type of electronic device used by a learner for learning, and may be a PC, laptop, tablet PC, etc., and the learner learns by reading a video-based learning object displayed on the learning mentee terminal (1), and learning video information based on the learner's eye processing is generated as a learning result.

[0039] The above-mentioned video-based learning object studied by the learner includes video-based textbooks, self-study books, reference books, and test papers, and the above-mentioned learner eye-tracking-based learning video information is characterized by being video-based learning object video information in which a video marker is displayed that moves on the video-based learning object along the learning eye focus of the learner studying the video-based learning object.

[0040] The learner learns by reading the video-based learning object displayed on the learning mentee terminal (1), and during learning, the learner's gaze focus moves on the video-based learning object.

[0041] An eye-tracking device (11) is installed in the learning mentee terminal (1) so that a video marker moving on the video-based learning object can be displayed on the video-based learning object in correspondence with the eye focus of a learner moving on the video-based learning object.

[0042] That is, the eye tracking device (11) performs the task of displaying an image marker at a location on the image-based learning object corresponding to the eye focus of the learner looking at the image-based learning object displayed on the screen of the learning mentee terminal (1), as the learner's eye focus moves.

[0043] The display of image markers on the image-based learning object by the eye tracking device (11) is saved in real time, and when the learner's learning is finished, learning image information based on the learner's eye processing is generated as a learning result.

[0045] Specifically, the first step (S100) is, as illustrated in FIG. 2,

[0046] Step 1-1 (S110) of displaying a video-based learning object on the screen of a learning mentee terminal (1), and

[0047] A first-2 step (S120) in which a video marker is displayed on a video-based learning object displayed on a screen, wherein the video marker is displayed at a position on the video-based learning object corresponding to the learning gaze focus of the learner studying the video-based learning object, and

[0048] The method is characterized by including a first-3 step (S130) of generating learning video information based on learner gaze processing, which is video information of a video-based learning object with a video marker displayed on it that moves along the learning gaze focus of a learner learning the video-based learning object in real time, by performing real-time storage processing of video-based learning object screen information with a video marker displayed on it until the learner finishes learning.

[0050] The above 1-1 step (S110) is a step of displaying video-based learning objects on the screen of the learning mentee terminal (1). The video-based learning objects displayed on the screen of the learning mentee terminal (1) include video-based textbooks, self-study books, reference books, and question papers, and FIG. 2 illustrates an example in which a video-based Korean language question paper is displayed.

[0052] The above 1-2 step (S120) is a step of displaying an image marker on an image-based learning object displayed on a screen, wherein the image marker is displayed at a position on the image-based learning object corresponding to the learning gaze focus of a learner studying the image-based learning object, and the image-based learning object screen information on which the image marker is displayed is saved and processed in real time.

[0053] As shown in FIG. 2, an eye tracking device (11) is installed in a learning mentee terminal (1) so that a video marker moving on a video-based learning object can be displayed on the video-based learning object in correspondence with the gaze focus of a learner moving on the video-based learning object, and the eye tracking device (11) causes a video marker to be displayed at a position on the video-based learning object corresponding to the gaze focus of a learner looking at the video-based learning object displayed on the screen of the learning mentee terminal (1).

[0055] That is, the eye tracking device (11) identifies the direction of the learner's gaze focus when looking at a video-based learning object displayed on the screen of the learning mentee terminal (1), identifies the position on the screen corresponding to the identified learner's gaze focus, and identifies the position on the video-based learning object displayed on the screen corresponding to the identified position on the screen. When the position on the video-based learning object corresponding to the learner's gaze focus is identified, a video marker is displayed at the corresponding position on the video-based learning object.

[0056] Figure 2 illustrates an example in which a video marker corresponding to the learner's gaze focus is displayed on a video-based Korean language test paper, and the video-based learning object screen information with the video marker displayed is stored and processed in real time by the learning mentee terminal (1).

[0058] The above 1-3 steps (S130) are steps of processing and saving video-based learning object screen information with video markers displayed in real time until the end of the learner's learning, thereby generating learning video information based on learner gaze processing, which is video-based learning object video information with video markers displayed that moves along the learning gaze focus of the learner learning the video-based learning object.

[0059] As the learner's learning progresses on the video-based learning object, the learner's focus of gaze moves on the video-based learning object, and the video marker displayed on the video-based learning object also moves in conjunction with the movement of the learner's focus of gaze, and the screen information of the video-based learning object with the video marker displayed is saved in real time until the learner finishes learning.

[0060] Accordingly, when the learner's learning is finished, learning video information based on learner eye processing is generated, which is video-based learning object video information with video markers displayed that move along the learning gaze focus of the learner learning the video-based learning object. FIG. 2 illustrates an example of learning video information based on learner eye processing with video markers displayed that move on a video-based Korean language test paper in conjunction with the movement of the learner's learning gaze focus.

[0062] The above second step (S200) is a step in which the learning mentee terminal (1) transmits learning video information based on learner eye processing to the learning mentor terminal (2) through a communication network, and as illustrated in FIG. 1, the learning video information based on learner eye processing generated through the first step (S100) is transmitted to the learning mentor terminal (2) at a remote location through a communication network.

[0064] The above third step (S300) is a step in which a learning mentor terminal (2) generates learning diagnostic mentoring information using learning video information based on learner eye processing, characterized by analyzing the movement pattern of video markers in the learning video based on learner eye processing on a video-based learning object to identify the learner's learning habits or learning abilities, and generating learning diagnostic mentoring information regarding the identified learner's learning habits or learning abilities.

[0065] Learning habits such as reading methods and concentration, as well as learning abilities such as literacy and problem-solving skills, are the most basic fundamental competencies for a learner and vary from learner to learner.

[0066] Therefore, the movement pattern of video markers moving on video-based learning objects varies for each learner as they follow the learning gaze focus of the learner studying the video-based learning objects. By analyzing the movement pattern of video markers, it is possible to determine whether learning habits such as reading methods and concentration are good or bad, and whether learning abilities such as literacy and problem-solving skills are good or bad. Additionally, the results of the analysis of the video marker movement pattern can be used to generate learning diagnostic mentoring information for the learner.

[0068] Below, the third step (S300) regarding the analysis of the movement pattern of video markers and the generation of learning diagnostic mentoring information for learners is described in detail.

[0070] In the above third step (S300), if the movement pattern of the image marker on the image-based learning object is at least one of the following first patterns, it is determined that the learner has poor reading habits, and learning diagnostic mentoring information is generated instructing the learner to read line by line in order while maintaining a constant reading speed from the beginning to the end of the sentence when reading, and the first pattern is characterized by including a pattern in which the image marker moves in a diagonal direction and a pattern in which the movement speed of the image marker irregularly repeats fast and slow within the sentence.

[0071] For learners with normal reading habits, reading line by line maintains a consistent reading speed from the beginning to the end of a sentence. However, reading diagonally instead of line by line, or reading with irregularly fast and slow speeds within a sentence, is an incorrect reading habit.

[0072] Therefore, if the movement pattern of the image marker on the image-based learning object is such that the movement speed is maintained at a constant level from the beginning to the end of the sentence and does not move line by line, and at least one of the following is observed: a pattern of moving diagonally as shown in Fig. 3, or a pattern in which the movement speed is irregularly repeated between fast and slow, it is determined that the learner has a bad reading learning habit, and learning diagnostic mentoring information is generated that instructs the learner to read line by line while maintaining a constant reading speed from the beginning to the end of the sentence when reading.

[0074] In addition, if the movement pattern of the image marker on the image-based learning object in the third step (S300) is at least one of the following second patterns, it is determined that the learner has a learning habit of reading text without concentration, and learning diagnostic mentoring information is generated instructing the learner to read line by line in order, maintaining a constant reading speed from the beginning to the end of the sentence without the gaze staying in one place for a long time or shaking up and down to adjacent lines. The second pattern is characterized by including a pattern in which the image marker moves back and forth between the beginning and end of the sentence, a pattern in which the image marker shakes up and down to adjacent lines, a pattern in which the image marker stays in one place for a certain period of time or longer, and a pattern in which the image marker suddenly disappears.

[0075] For learners with focused reading habits, reading is done in order line by line, maintaining a constant reading speed from the beginning to the end of a sentence without the gaze lingering in one place for a long time or swaying up and down to adjacent lines. However, if the gaze lingers in one place for a long time while reading (so-called zoning out), sways up and down to adjacent lines (reading in an anxious state), moves back and forth between the beginning and end of a sentence (moving the gaze without the intention of understanding the content of the sentence), or suddenly turns the gaze to another place while reading (getting distracted while reading), these are habits of reading without concentration.

[0076] Therefore, if the movement pattern of the image marker on the image-based learning object does not stay in one place for a long time, does not shake up and down, and does not move line by line in order while maintaining a constant reading speed from the beginning to the end of the sentence, and if at least one of the following patterns is present as shown in Fig. 4—such as the image marker moving back and forth between the beginning and end of the sentence, the image marker shaking up and down to adjacent lines, the image marker staying in one place for a certain period of time or longer, or the image marker suddenly disappearing—it is determined that the learner has a learning habit of reading without concentration, and learning diagnostic mentoring information is generated that instructs the learner to read line by line in order while maintaining a constant reading speed from the beginning to the end of the sentence, without the gaze staying in one place for a long time or shaking up and down to adjacent lines.

[0078] In addition, if the movement pattern of the video marker on the video-based learning object in the third step (S300) is at least one of the following third patterns, it is determined that the learner has poor literacy, and learning diagnostic mentoring information is generated instructing the learner to read at a certain speed in the order of the current sentence or paragraph while thinking about the content of the previous sentence or paragraph when reading the text, and the third pattern is characterized by including a pattern in which the time the video marker stays in one part is longer than in other parts, a pattern in which the movement of the video marker is repeated only within one sentence, and a pattern in which the video marker passes quickly through an important part (a part containing key content or vocabulary).

[0079] Literacy is the learning ability to grasp the content of a text. Literacy learners read at a steady pace, following the order of the current sentence or paragraph while thinking about the content of the previous sentence or paragraph. However, if one lacks literacy, they may spend more time in one part of the text than in other parts, repeat reading only within a single sentence, or quickly skip over important parts (e.g., parts containing key content or vocabulary).

[0080] Specifically, referring to Figure 5, in the case of a learner with poor literacy, the learner's gaze lingers longer on the part called utility, which is considered difficult vocabulary, compared to other parts, and repeatedly reads the parts that are not understood (e.g., Herbert Simon presented two perspectives regarding rational action: comprehensive rationality and bounded rationality), and quickly passes by important parts (e.g., the perspective that the agent acknowledges that there are limitations) without reading them properly.

[0081] Accordingly, as illustrated in Fig. 5, if at least one of the following is observed in the movement pattern of the image marker on the image-based learning object—such as a pattern where the time spent in one part is longer than in another, a pattern where movement is repeated only within one sentence, or a pattern where the learner quickly passes through important parts (parts containing key content or vocabulary)—it is determined that the learner lacks literacy, and learning diagnostic mentoring information is generated instructing the learner to read at a constant speed in the order of the current sentence or paragraph while thinking about the content of the previous sentence or paragraph when reading.

[0083] In addition, if the movement pattern of the image marker on the image-based learning object in the third step (S300) is at least one of the following fourth patterns, it is determined that the learner lacks problem-solving ability, and learning diagnostic mentoring information is generated instructing the learner to read the question and answer choices sequentially first, and then read the entire passage at a certain speed when solving problems, wherein the fourth pattern is characterized by including a pattern in which the image marker moves back and forth between the passage and the question, a pattern in which the image marker stays on the answer choices for too long, and a pattern in which the image marker stays on one problem for too long.

[0084] Learners with problem-solving skills read both the question and the answer choices first, and then read the entire passage at a certain speed when solving problems. However, learners without problem-solving skills may read back and forth between the passage and the question, read only the answer choices for a long time, or when they encounter a difficult problem, they become fixated on that problem and are unable to move on to the next problem.

[0085] Accordingly, as shown in Fig. 6, if the movement pattern of the image marker on the image-based learning object is at least one of the following: a pattern where the image marker moves back and forth between the text and the question, a pattern where the image marker stays on the answer choices for too long, or a pattern where the image marker stays on one problem for too long, it is determined that the learner lacks problem-solving ability, and learning diagnostic mentoring information is generated that instructs the learner to read the question and answer choices sequentially first, and then read the entire text at a certain speed when solving the problem.

[0087] Meanwhile, in the above third step (S300), artificial intelligence may be used to analyze the movement pattern of the image marker and to generate learning diagnostic mentoring information for the learner.

[0088] Specifically, the artificial intelligence, which has learned through deep learning normal / abnormal movement patterns regarding reading methods / concentration / literacy / problem-solving skills of video markers on video-based learning objects, uses the learning results to analyze the movement patterns of video markers on video-based learning objects within a learning video based on learner eye processing to identify the learner's learning habits or learning abilities, and generates learning diagnostic mentoring information regarding the identified learner's learning habits or learning abilities.

[0089] Referring to Fig. 7, the artificial intelligence learns normal / abnormal movement patterns regarding the reading method / concentration / literacy / problem-solving ability of image markers on an image-based learning object through deep learning, and pre-constructs a judgment model capable of determining whether the reading method / concentration / literacy / problem-solving ability is normal / abnormal according to the movement pattern of the image marker, and a mentoring model capable of generating mentoring information to improve the reading method / concentration / literacy / problem-solving ability when it is abnormal.

[0090] When learning video information based on learner eye processing is provided as input, the artificial intelligence uses a judgment model to determine whether reading methods, concentration, literacy, and problem-solving skills are normal or abnormal based on the movement patterns of video markers within the learning video. If the result indicates abnormality, it uses a mentoring model to generate learning diagnostic mentoring information regarding the learner's learning habits or abilities that can improve reading methods, concentration, literacy, and problem-solving skills.

[0092] The above fourth step (S400) is a step in which the learning mentor terminal (2) transmits learning diagnostic mentoring information to the learning mentee terminal (1) through a communication network, and as illustrated in FIG. 1, the learning diagnostic mentoring information generated through the third step (S300) is transmitted to the learning mentee terminal (1) at a remote location through a communication network.

[0094] Although the technical concept of the present invention has been described above together with the accompanying drawings, this is merely an illustrative description of preferred embodiments of the present invention and is not intended to limit the invention. It is obvious that the scope of the rights of the present invention is not limited to the embodiments but includes modifications made by those skilled in the art within the scope of the technical concept of the present invention. Explanation of the symbols

[0096] 1 : Learning Mentee Terminal 11: Eye tracking device 2 : Learning Mentor Terminal

Claims

Claim 1 A learning mentoring method based on learner eye processing comprises: a first step (S100) in which a learning mentee terminal (1) generates learning video information based on learner eye processing regarding content to be received for learning mentoring; a second step (S200) in which the learning mentee terminal (1) transmits the learning video information based on learner eye processing to a learning mentor terminal (2) through a communication network; a third step (S300) in which the learning mentor terminal (2) generates learning diagnostic mentoring information using the learning video information based on learner eye processing; and a fourth step (S400) in which the learning mentor terminal (2) transmits the learning diagnostic mentoring information to the learning mentee terminal (1) through a communication network, wherein the learning video information based on learner eye processing is video information of a learning object based on a video, wherein a video marker is displayed that moves on the learning object based on a video, following the learning gaze focus of a learner learning the learning object based on a video. Claim 2 A learning mentoring method based on learner eye processing according to claim 1, characterized in that the image-based learning object includes an image-based textbook, tutorial, reference book, and test paper. Claim 3 A learning mentoring method based on learner gaze processing according to claim 1, wherein the first step (S100) comprises: a first-1 step (S110) of displaying a video-based learning object on the screen of a learning mentee terminal (1); a first-2 step (S120) of displaying a video marker on the video-based learning object displayed on the screen, wherein the video marker is displayed at a position on the video-based learning object corresponding to the learning gaze focus of a learner learning the video-based learning object; and a first-3 step (S130) of generating learning video information based on learner gaze processing, wherein the video marker is displayed on the video-based learning object video information that moves along the learning gaze focus of a learner learning the video-based learning object, by performing real-time storage processing of the video-based learning object screen information with the video marker displayed until the end of the learner's learning. Claim 4 A learning mentoring method based on learner eye processing according to claim 1, wherein the third step (S300) is characterized by analyzing the movement pattern of an image marker in a learning video based on learner eye processing on an image-based learning object to identify the learner's learning habits or learning abilities, and generating learning diagnostic mentoring information regarding the identified learner's learning habits or learning abilities. Claim 5 A learning mentoring method based on learner eye processing according to claim 4, wherein the third step (S300) identifies that the learner has a bad reading learning habit if the movement pattern of an image marker on an image-based learning object is at least one of the following first patterns, and generates learning diagnostic mentoring information instructing the learner to read line by line while maintaining a constant reading speed from the beginning to the end of the sentence when reading, and wherein the first pattern includes a pattern in which the image marker moves in a diagonal direction and a pattern in which the movement speed of the image marker irregularly repeats fast and slow within the sentence. Claim 6 In claim 4, the third step (S300) is characterized by identifying that if the movement pattern of an image marker on an image-based learning object is at least one of the following second patterns, the learner has a learning habit of reading text without concentration, and generating learning diagnostic mentoring information that instructs the learner to read line by line in order while maintaining a constant reading speed from the beginning to the end of the sentence without the gaze staying in one place for a long time or shaking up and down to adjacent lines, and the second pattern includes a pattern in which the image marker moves back and forth between the beginning and end of the sentence, a pattern in which the image marker shakes up and down to adjacent lines, a pattern in which the image marker stays in one place for a certain period of time or longer, and a pattern in which the image marker suddenly disappears. Claim 7 In claim 4, the third step (S300) is characterized by generating learning diagnostic mentoring information that indicates that if the movement pattern of an image marker on an image-based learning object is at least one of the following third patterns, the learner has poor literacy, and that when reading a text, the learner reads at a certain speed in accordance with the order of the current sentence or paragraph while thinking about the content of the previous sentence or paragraph, and the third pattern includes a pattern in which the time the image marker stays in one part is longer than in other parts, a pattern in which the movement of the image marker is repeated only within one sentence, and a pattern in which the image marker passes quickly through an important part (a part containing key content or vocabulary). Claim 8 In claim 4, the third step (S300) identifies that the learner lacks problem-solving ability if the movement pattern of the image marker on the image-based learning object is at least one of the following four patterns, and generates learning diagnostic mentoring information instructing the learner to read the question sheet and answer choices sequentially first, and then read the entire passage at a certain speed when solving a problem, wherein the fourth pattern includes a pattern in which the image marker moves back and forth between the passage and the question sheet, a pattern in which the image marker stays on an answer choice for too long, and a pattern in which the image marker stays on one problem for too long. Claim 9 A learning mentoring method based on learner eye processing according to claim 4, wherein the third step (S300) is characterized by an artificial intelligence that has learned through deep learning normal / abnormal movement patterns regarding text reading method / concentration / literacy / problem-solving ability of a video marker on a video-based learning object using the learning results to analyze the movement patterns of a video marker in a learning video based on learner eye processing on a video-based learning object to identify the learner's learning habits or learning abilities, and to generate learning diagnostic mentoring information regarding the identified learner's learning habits or learning abilities.

Images