Online education system and method based on generative artificial intelligence

Abstract

The present invention relates to the field of online education technology, and in particular to an online education system and method based on generative artificial intelligence, including a teaching end, a learning end and a resource processing end. By taking a teaching group as a unit, the learning behavior of students is analyzed to obtain the learning habits of the students, and then processing is performed according to the learning habits of the students to obtain the common value of the habits of the students in the teaching group. After that, the teaching resources are processed according to the common value of the students to obtain the common resources. Finally, taking the common resources as the center, the learning deviation rate of the students' learning habits and the common resources is processed, so as to obtain the students' personalized learning scenes, avoiding the process that the system has to repeatedly process the data every time the students learn online, reducing the amount of data processing by the system, enabling the common resources to adapt to most students, and further improving the processing efficiency of the system and the learning efficiency of the students.

Description

Technical Field

[0001] The present invention relates to the field of online education technology, and in particular to an online education system and method based on generative artificial intelligence. Background Art

[0002] The online education system is a platform that uses Internet technology to provide remote teaching and learning for students and teachers.

[0003] Prior art CN113129665A discloses an artificial intelligence online teaching interactive method, system, storage medium and related equipment, including real-time understanding of students' learning dynamics and learning effects, rapid and timely monitoring and reminding of the students, targeted control of the students' learning behavior, improving students' learning efficiency, improving the quality of teachers' teaching in online courses, and strengthening the interaction between students and teachers; including artificial intelligence online teaching interactive method, artificial intelligence online teaching interactive server, artificial intelligence online teaching interactive terminal, artificial intelligence online teaching interactive system, computer storage medium, computer equipment;

[0004] In the above-mentioned teaching system, generally one teacher teaches a class of students. However, this process cannot take into account the learning habits of each student. If the teacher and the student are given individual personalized teaching one by one, the amount of data processed by the online education system will be too large. When there are too many students learning online, it will easily cause system failure. Summary of the invention

[0005] The purpose of the present invention is to solve the problems in the background technology and to propose an online education system and method based on generative artificial intelligence.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] Online education systems and methods based on generative artificial intelligence, including teaching, learning and resource processing ends;

[0008] The teaching end is used to collect teaching information and transmit it to the resource processing end, where the teaching information includes educational resources and teaching groups;

[0009] The learning end is used to collect students' behavior information and transmit it to the resource processing end. The behavior information includes learning time, interaction tracking and interaction frequency.

[0010] The resource processing end is used to receive teaching information and behavior information. The behavior information is first processed inertially to obtain learning habits. Learning habits include inertial time, concentration time and inertial interaction frequency. The maximum value of the mode is extracted according to the student's learning time, and the time corresponding to the maximum value is marked as inertial time. Then, according to the teaching screen, a sight line designated area is set for the teaching screen. Then, the real-time sight line points of the students are detected, and the real-time sight line points and the sight line designated area are comprehensively processed to obtain the concentration value. According to the concentration value, the student's concentration time is obtained. Finally, the interaction frequency between the student and the teacher is averaged during the concentration time to obtain the inertial interaction frequency.

[0011] Then, for the learning habits of students in the teaching group, the common value of the learning habits of the teaching group is first taken, and then the educational resources are split according to the common value to obtain the common resources;

[0012] The learning deviation rate is then calculated using the public resources and the learning habits of individual students, and the student's personalized learning scenario is obtained based on the learning deviation rate. The resource processing end then transmits the personalized learning scenario to the learning end.

[0013] As a further solution of the present invention, the teaching end includes a teacher behavior management module and a teaching data collection module;

[0014] The teacher behavior management module is used to upload educational resources, where educational resources refer to materials related to teaching;

[0015] The teaching data collection module is used to collect teaching information, which includes educational resources and teaching groups.

[0016] As a further solution of the present invention, the learning end includes a student behavior collection module and a personalized scene application module;

[0017] The student behavior collection module is used to collect students' learning behaviors during the learning process and transmit them to the resource processing end;

[0018] The personalized scene application module is used to receive personalized learning scenes and display them on students' learning terminals for students to study.

[0019] As a further solution of the present invention, the resource processing end includes a teaching information processing module, a student action processing module and a comprehensive information processing module;

[0020] The student action processing module is used to receive behavior information and perform inertial processing on the behavior information to obtain the student's learning habits;

[0021] The teaching information processing module is used to receive learning habits, obtain the common value of the learning habits of the teaching group, and then split the educational resources according to the common value to obtain the common resources;

[0022] The comprehensive information processing module is used to calculate the learning deviation rate based on public resources and the learning habits of individual students, and to obtain the students' personalized learning scenarios based on the learning deviation rate.

[0023] As a further solution of the present invention, a method for acquiring learning habits includes:

[0024] Select a target student, take the current time as the node, and extract the learning information of the target student's previous N online learning times, where N is the threshold;

[0025] According to the learning information of the target students, the mode of online learning time is extracted and marked as inertial duration;

[0026] Identify the learning resources of the target students during online learning, select the teaching screen in the online teaching video at this time, and establish a plane coordinate system with the lower left corner of the teaching screen as the origin;

[0027] In chronological order, set the corresponding sightline designated area for each time period in the teaching screen and mark it as Wherein (Xa, Xb) refers to the region formed between Xa and Xb in the X-axis direction, and (Ya, Yb) refers to the region formed between Ya and Yb in the Y-axis direction;

[0028] Extract interactive tracking data in learning behaviors and mark them as real-time sight points (SXi, SYi), where i represents different times;

[0029] Set the interval time, the interval time is the threshold, first detect the number of sight deviations Cs within the interval time, the number of sight deviations refers to the number of times the real-time sight point is not in the specified sight area, at the same time, when the real-time sight point is not in the specified sight area, collect its duration and mark it as the reaction time Tf, based on the formula The offset value Dp is obtained, where (SX1, SY1) is the point where the real-time sight point stays the longest during the reaction time, and (Xz, Yz) is the center point in the specified sight area;

[0030] Afterwards, The concentration value MZ of the target student in the interval time is obtained, DPa is the mean of the offset value Dp in the interval time, Tfa is the mean of the reaction time in the interval time, α1 is the correction coefficient, b1 and b2 are both constants, and 0<b1<1, 0<b2<1;

[0031] When the concentration value MZ is less than or equal to My, the corresponding time point is marked as a rest point. Then, the rest point of each online learning of the target student is detected, and the time between the start time of the target student's study and the rest point is marked as the focused learning time. The average of the previous N focused learning times of the target student is taken to obtain the target student's focused time, where My is the threshold;

[0032] Taking the concentration time as the unit, the interaction frequency between the target students and the teacher during online learning is extracted, and the interaction frequency of the target students during the previous N online learning times is averaged to obtain the inertial interaction frequency.

[0033] As a further solution of the present invention, a method for acquiring public resources includes:

[0034] An educational resource is used as designated analysis information. First, the teaching group targeted by the designated analysis information is identified. According to the learning habits of students in the teaching group, any learning habit is selected as the target habit. The mode value of the target habit in the teaching group is detected, and the maximum mode value is extracted. The corresponding data is used as the common value of the target habit in this teaching group.

[0035] The remaining data in the learning habits are taken as target habits in turn and subjected to majority value detection to obtain the corresponding common values;

[0036] According to the educational resources uploaded in the teacher behavior management module, a teaching group is first selected as the target group. According to the common value of the inertia duration, the educational resources are divided into several teaching segments according to the common value of the inertia duration. Then, according to the common value of the concentration time, pause points are set in the teaching segments according to the common value of the concentration time, and corresponding interaction information is set in the pause points according to the inertia interaction frequency, so as to obtain public resources.

[0037] As a further solution of the present invention, a method for acquiring a personalized learning scenario includes:

[0038] First, extract students' learning habits, combine them with public resources, and use The learning deviation rate Gp is ​​obtained, where Sj represents the student's learning habits, j represents different learning habits, Sjg represents the common value corresponding to the learning habits, and βj represents the weight coefficient corresponding to the learning habit j;

[0039] When the learning deviation rate is less than or equal to the standard deviation rate, the public resources are transmitted to the personalized scenario application module as the student's personalized learning scenario. If the learning deviation rate is greater than the standard deviation rate, the public resources are adjusted according to the student's learning habits to generate a personalized learning scenario and transmit it to the personalized scenario application module. The standard deviation rate is the threshold.

[0040] The online education method based on generative artificial intelligence specifically comprises the following steps:

[0041] Step 1: Collect teaching information, which refers to educational resources and teaching groups, and then collect students' learning behaviors;

[0042] Step 2: Analyze students’ learning behaviors and obtain their learning habits, which include inertial duration, concentration time, and inertial interaction frequency;

[0043] Step 3: Based on the students’ learning habits, conduct a mode test on the learning habits of a teaching group to obtain the common value of the learning habits in the teaching group;

[0044] Step 4: Then, according to the common value of learning habits, the teaching resources are processed, and the common resources are obtained according to the processing results. Then, when students study, the common resources are calculated with the students' learning habits to obtain the learning deviation rate. Finally, the corresponding student's personalized learning scenario is obtained according to the learning deviation rate.

[0045] Compared with the prior art, the advantages of the present invention are:

[0046] The present invention analyzes the learning behavior of students by taking the teaching group as the unit to obtain the learning habits of students, and then processes them according to the learning habits of students to obtain the common value of the habits of students in the teaching group, and then processes the teaching resources according to the common values ​​of students to obtain common resources, and finally takes the common resources as the center to process the learning deviation rate of students' learning habits and common resources, so as to obtain the personalized learning scene of students, avoiding the process that the system has to repeatedly process the data every time the students learn online, reducing the amount of data processed by the system, so that the common resources can adapt to most students, and further improve the processing efficiency of the system and the learning efficiency of students;

[0047] The present invention uses generative artificial intelligence technology to process data, further improving resource processing efficiency and helping to optimize and update teaching resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 It is a schematic diagram of the system structure of the present invention;

[0049] Figure 2 It is a schematic diagram of the method flow of the present invention. DETAILED DESCRIPTION

[0050] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0051] Reference Figure 1 and Figure 2 , an online education system based on generative artificial intelligence, including teaching end, learning end, and resource processing end;

[0052] The teaching end is mainly used for the management of the online education platform. The specific teaching end includes the teacher behavior management module and the teaching data collection module. The learning end is used to manage the student's learning process. The specific learning end includes the student behavior collection module and the personalized scene application module. The resource processing end is used for comprehensive processing of the data collected from the teaching end and the learning end. The specific resource processing end includes the teaching information processing module, the student action processing module, and the comprehensive information processing module.

[0053] The teacher behavior management module is mainly used to upload educational resources. Specifically, teachers upload teaching resources to the online education platform through the teacher behavior management module according to the teaching needs of the course and store them. The teaching resources include teaching videos and teaching materials, etc., and the online education platform in this embodiment mainly provides services for entrepreneurship guidance related courses;

[0054] The teaching data collection module is used to collect teaching information, where the teaching information includes the teacher's educational resources and teaching groups. The educational resources are the information uploaded by the teacher in the teacher behavior management module. The teaching group refers to the group of students that the teacher is responsible for teaching. In the teaching system of colleges and universities, the teaching groups are divided into classes, that is, a teacher is responsible for teaching multiple classes, and one class is a teaching group. Then the teaching data collection module transmits the teaching information to the teaching information processing module;

[0055] The student behavior collection module is used to collect the student's behavior during the learning process to obtain the learning behavior, which includes login frequency, interactive tracking, learning time, course progress, learning results and browsing path, etc. After that, the student behavior collection module transmits the learning behavior to the student action processing module. It should be further explained that the student behavior collection module is provided with an eye tracking device, and its interactive tracking is to track the line of sight of the target student through the eye tracking device;

[0056] The student action processing module performs inertial processing on the student's behavior information based on the received behavior information to obtain the student's learning habits. The specific inertial processing method is:

[0057] Select a target student, take the current time as the node, and extract the learning information of the target student's previous N online learning times, where N is a threshold. In this embodiment, N is 30;

[0058] According to the online learning time of each time in the learning information of the target students, first extract the mode of the online learning time and mark it as the inertia duration;

[0059] Then, the learning concentration value of the target students is calculated to obtain the learning concentration time of the target students. The specific calculation method of the concentration time includes:

[0060] First, the learning resources for the target students during online learning are identified. The learning resources selected here are online teaching videos. The teaching screen in the online teaching video is selected, and the plane coordinate system is established with the lower left corner of the teaching screen as the origin.

[0061] Then, in chronological order, set the corresponding sightline designated area for each time period in the teaching screen and mark it as Wherein (Xa, Xb) refers to the region formed between Xa and Xb in the X-axis direction, and (Ya, Yb) refers to the region formed between Ya and Yb in the Y-axis direction;

[0062] Extract interactive tracking data from learning behaviors. The interactive tracking data is the point where the target student's sight lines are on the teaching screen, and mark it as a real-time sight point (SXi, SYi), where i represents different times.

[0063] Then set the interval time, which is the threshold. First, detect the number of sight deviations Cs during the interval time. The number of sight deviations refers to the number of times the real-time sight point is not in the specified sight area. At the same time, when the real-time sight point is not in the specified sight area, collect its duration and mark it as the reaction time Tf. Further, based on the formula The offset value Dp is obtained, where (SX1, SY1) is the point where the real-time sight point stays the longest during the reaction time, and (Xz, Yz) is the center point in the specified sight area;

[0064] Afterwards, The concentration value MZ of the target student in the interval time is obtained, DPa is the mean value of the offset value Dp in the interval time, Tfa is the mean value of the reaction time in the interval time, α1 is the correction coefficient, in this embodiment, α1 is 0.974, b1 and b2 are both constants, and 0<b1<1, 0<b2<1;

[0065] When the concentration value MZ is less than or equal to My, the corresponding time point is marked as a rest point. Then, the rest point of each online learning of the target student is detected, and the time between the start time of the target student's learning and the rest point is marked as the focused learning time. The average of the previous N focused learning times of the target student is taken to obtain the focused time of the target student, where My is a threshold value, and its specific value is set by a technician in this field;

[0066] Then, the interaction frequency between the target students and the teacher during online learning was extracted based on the focus time. The interaction frequency of the target students during the previous N online learning sessions was averaged to obtain the inertial interaction frequency.

[0067] Afterwards, the student action processing module integrates the inertia duration, concentration time, and interaction frequency into the learning habits of the target students and transmits them to the teaching information processing module;

[0068] The teaching information processing module is used to process teaching resources according to learning habits. The specific processing method is as follows:

[0069] Taking a teaching group that a teacher is responsible for as an example, an educational resource is selected as designated analysis information. The teaching group targeted by the designated analysis information is first identified, and the common habits of the teaching group are extracted. The method for extracting the common habits of students includes:

[0070] According to the students' learning habits, any learning habit is selected as the target habit, the mode value of the target habit in the teaching group is detected, the maximum value is extracted, and the corresponding data is used as the common value of the target habit in this teaching group;

[0071] The remaining data in the learning habits are taken as target habits in turn and subjected to majority value detection to obtain the corresponding common values;

[0072] Afterwards, the educational resources are processed according to the public value of the learning habits, and the processed resources are marked as public resources for this teaching group. The specific method for obtaining public resources includes:

[0073] According to the educational resources uploaded in the teacher behavior management module, a teaching group is first selected as the target group, and the common values ​​of learning habits in the target group are extracted;

[0074] Then the educational resources are split according to the common values. Specifically, first, the educational resources are divided into several teaching segments according to the common value of the inertia duration. Then, according to the common value of the concentration time, pause points are set in the teaching segments according to the common value of the concentration time, and corresponding interaction information is set in the pause points according to the inertia interaction frequency, so as to obtain the common resources.

[0075] It should be further explained that one teaching group corresponds to one public resource, so that the teaching resources in the online education platform can adapt to the learning habits of most students. After that, the teaching information processing module transmits the public resources to the comprehensive information processing module;

[0076] The comprehensive information processing module performs comprehensive processing on the public resources and learning habits based on the received public resources. The specific method of comprehensive processing is as follows:

[0077] First, extract students' learning habits, combine them with public resources, and use The learning deviation rate Gp is ​​obtained, where Sj represents the student's learning habits, j represents different learning habits, Sjg represents the common value corresponding to the learning habits, and βj represents the weight coefficient corresponding to the learning habit j;

[0078] When the learning deviation rate is ≤ the standard deviation rate, the public resources are transmitted to the personalized scene application module as the student's personalized learning scene. If the learning deviation rate is greater than the standard deviation rate, the public resources are adjusted according to the student's learning habits to generate a personalized learning scene and transmit it to the personalized scene application module. The standard deviation rate is a threshold value, which is specifically set by a technician in this field.

[0079] The personalized scene application module displays the received personalized learning scenes on the students' learning terminals for students to learn, further improving the teaching results.

[0080] It should be further explained that the present invention processes data based on generative artificial intelligence technology, further improving resource processing efficiency and helping to optimize and update teaching resources.

[0081] The online education method based on generative artificial intelligence specifically comprises the following steps:

[0082] Step 1: Collect teaching information, which refers to educational resources and teaching groups, and then collect students' learning behaviors;

[0083] Step 2: Analyze students’ learning behaviors and obtain their learning habits, which include inertial duration, concentration time, and inertial interaction frequency;

[0084] Step 3: Based on the students’ learning habits, conduct a mode test on the learning habits of a teaching group to obtain the common value of the learning habits in the teaching group;

[0085] Step 4: Then, according to the common value of learning habits, the teaching resources are processed, and the common resources are obtained according to the processing results. Then, when students study, the common resources are calculated with the students' learning habits to obtain the learning deviation rate. Finally, the corresponding student's personalized learning scenario is obtained according to the learning deviation rate.

[0086] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. An online education system based on generative artificial intelligence, characterized by: Including teaching end, learning end and resource processing end; The teaching end is used to collect teaching information and transmit it to the resource processing end, where the teaching information includes educational resources and teaching groups; The learning end is used to collect students' behaviors, obtain behavioral information, and transmit it to the resource processing end. The behavioral information includes learning time, interaction tracking, and interaction frequency; The resource processing end is used to receive teaching information and behavior information. The behavior information is first processed inertially to obtain learning habits. Learning habits include inertial time, concentration time and inertial interaction frequency. The maximum value of the mode is extracted according to the student's learning time, and the time corresponding to the maximum value is marked as inertial time. Then, according to the teaching screen, a sight line designated area is set for the teaching screen. Then, the real-time sight line points of the students are detected, and the real-time sight line points and the sight line designated area are comprehensively processed to obtain the concentration value. According to the concentration value, the student's concentration time is obtained. Finally, the interaction frequency between students and teachers is averaged during the concentration time to obtain the inertial interaction frequency. The specific processing process of inertial processing includes: Select a target student, take the current time as the node, and extract the learning information of the target student's previous N online learning times, where N is the threshold; According to the learning information of the target students, the mode of online learning time is extracted and marked as inertial duration; Identify the learning resources used by the target students during online learning, select the teaching screen in the online teaching video at this time, and establish a plane coordinate system with the lower left corner of the teaching screen as the origin; In chronological order, set the corresponding sightline designated area for each time period in the teaching screen and mark it as Wherein (Xa, Xb) refers to the region formed between Xa and Xb in the X-axis direction, and (Ya, Yb) refers to the region formed between Ya and Yb in the Y-axis direction; Extract interactive tracking data in learning behaviors and mark them as real-time sight points (SXi, SYi), where i represents different times; Set the interval time, the interval time is the threshold, first detect the number of sight deviations Cs within the interval time, the number of sight deviations refers to the number of times the real-time sight point is not in the specified sight area, at the same time, when the real-time sight point is not in the specified sight area, collect its duration and mark it as the reaction time Tf, based on the formula The offset value Dp is obtained, where (SX1, SY1) is the point where the real-time sight point stays the longest during the reaction time, and (Xz, Yz) is the center point in the specified sight area; Afterwards, The concentration value MZ of the target student in the interval time is obtained, DPa is the mean of the offset value Dp in the interval time, Tfa is the mean of the reaction time in the interval time, α1 is the correction coefficient, b1 and b2 are both constants, and 0<b1<1, 0<b2<1; When the concentration value MZ is less than or equal to My, the corresponding time point is marked as a rest point. Then, the rest point of each online learning of the target student is detected, and the time between the start time of the target student's study and the rest point is marked as the focused learning time. The average of the previous N focused learning times of the target student is taken to obtain the target student's focused time, where My is the threshold; Taking the concentration time as the unit, the interaction frequency between the target students and the teacher during online learning is extracted, and the interaction frequency of the target students during the previous N online learning times is averaged to obtain the inertial interaction frequency; Then, for the learning habits of students in the teaching group, the common value of the learning habits of the teaching group is first obtained, and then the educational resources are split according to the common value to obtain the common resources. Specifically, the method for obtaining the common resources includes: Randomly select an educational resource as designated analysis information, first identify the teaching group targeted by the designated analysis information, select any learning habit as the target habit according to the learning habits of students in the teaching group, detect the mode value of the target habit in the teaching group, extract the maximum mode value, and use the corresponding data as the common value of the target habit in this teaching group; The remaining data in the learning habits are taken as target habits in turn and subjected to majority value detection to obtain the corresponding common values; According to the educational resources uploaded in the teacher behavior management module, a teaching group is first selected as the target group. According to the common value of the inertia duration, the educational resources are divided into several teaching segments according to the common value of the inertia duration. Then, according to the common value of the concentration time, the pause points are set in the teaching segments according to the common value of the concentration time, and the corresponding interaction information is set in the pause points according to the inertia interaction frequency, so as to obtain the public resources; Then, the learning deviation rate is calculated for the public resources and the learning habits of individual students, and the individual learning scenario of the student is obtained according to the learning deviation rate. Then, the resource processing end transmits the individual learning scenario to the learning end. The method for determining the individual learning scenario includes: First, extract students' learning habits, combine them with public resources, and use The learning deviation rate Gp is ​​obtained, where Sj represents the student's learning habits, j represents different learning habits, Sjg represents the common value corresponding to the learning habits, and βj represents the weight coefficient corresponding to the learning habit j; When the learning deviation rate is less than or equal to the standard deviation rate, the public resources are transmitted to the personalized scenario application module as the student's personalized learning scenario. If the learning deviation rate is greater than the standard deviation rate, the public resources are adjusted according to the student's learning habits to generate a personalized learning scenario and transmit it to the personalized scenario application module. The standard deviation rate is the threshold.

2. The online education system based on generative artificial intelligence according to claim 1, characterized in that: The teaching end includes the teacher behavior management module and the teaching data collection module; The teacher behavior management module is used to upload educational resources, where educational resources refer to materials related to teaching; The teaching data collection module is used to collect teaching information, which includes educational resources and teaching groups.

3. The online education system based on generative artificial intelligence according to claim 1, characterized in that: The learning end includes a student behavior collection module and a personalized scenario application module; The student behavior collection module is used to collect students' learning behaviors during the learning process and transmit them to the resource processing end; The personalized scene application module is used to receive personalized learning scenes and display them on students' learning terminals for students to study.

4. The online education system based on generative artificial intelligence according to claim 1, characterized in that: The resource processing end includes a teaching information processing module, a student action processing module, and a comprehensive information processing module; The student action processing module is used to receive behavior information and perform inertial processing on the behavior information to obtain the student's learning habits; The teaching information processing module is used to receive learning habits, obtain the common value of the learning habits of the teaching group, and then split the educational resources according to the common value to obtain the common resources; The comprehensive information processing module is used to calculate the learning deviation rate based on public resources and the learning habits of individual students, and to obtain the students' personalized learning scenarios based on the learning deviation rate.

5. An online education method based on generative artificial intelligence, the method being applied to the online education system of any one of claims 1 to 4 above, characterized in that: The method specifically comprises the following steps: Step 1: Collect teaching information, which refers to educational resources and teaching groups, and then collect students' learning behaviors; Step 2: Analyze students’ learning behaviors and obtain their learning habits, which include inertial duration, concentration time, and inertial interaction frequency; Step 3: Based on the students’ learning habits, conduct a mode test on the learning habits of a teaching group to obtain the common value of the learning habits in the teaching group; Step 4: Then, according to the common value of learning habits, the teaching resources are processed, and the common resources are obtained according to the processing results. Then, when students study, the common resources are calculated with the students' learning habits to obtain the learning deviation rate. Finally, the corresponding student's personalized learning scenario is obtained according to the learning deviation rate.

Images