Interactive teaching system integrated with multi-modal perception

Abstract

The invention relates to an interactive teaching system fused with multi-modal perception, which comprises a data acquisition module used for collecting a multi-source original teaching data set and carrying out primary processing on the multi-source original teaching data set to obtain an initial teaching data set, and a data processing module used for carrying out deep analysis on the initial teaching data set to determine a student learning state reference value, the students are classified; the data storage module is used for storing an initial teaching data set and a student learning state reference value; and the access control module is used for analyzing the access data and establishing user permission. According to the invention, through multi-source data acquisition and deep analysis, the learning state of the student can be accurately positioned, a personalized teaching strategy basis is provided for teachers, for example, corresponding teaching suggestions are pushed according to the classification of the student, and the teaching pertinence is improved. Secondly, data storage adopts encryption and block chain technologies, data security and integrity are guaranteed, information leakage and tampering are prevented, and student privacy and teaching data credibility are maintained.

Description

Technical Field

[0001] The present invention belongs to the technical field of teaching systems, and in particular, relates to an interactive teaching system integrating multimodal perception. Background Art

[0002] In today's era of digital education, traditional teaching models face many challenges in meeting students' personalized learning needs, accurately assessing students' learning status, and efficiently integrating teaching resources. Existing teaching systems often lack the deep integration and effective use of students' multi-dimensional learning information, and it is difficult to perceive key factors such as students' learning behavior, knowledge mastery, and learning attitudes in real time and comprehensively, resulting in teaching decisions that lack sufficient pertinence and timeliness, and it is difficult for teaching results to reach the ideal level. The present invention aims to deeply integrate multimodal perception technology into the teaching system through an innovative technical architecture, effectively solve the above problems, and inject new vitality and efficiency into educational and teaching activities. Summary of the invention

[0003] The purpose of the present invention is to provide an interactive teaching system integrating multimodal perception, which solves the problems existing in the prior art.

[0004] The purpose of the present invention can be achieved by the following technical solutions:

[0005] An interactive teaching system integrating multimodal perception, comprising:

[0006] The data collection module is responsible for collecting multi-source original teaching data sets and performing preliminary processing on them to obtain the initial teaching data sets. The original teaching data covers students' basic information data, learning process data, and learning outcome data. Students' basic information data includes students' name data, students' gender data, students' grade and class data, students' age data, and students' contact information data; learning process data includes students' class participation data and after-school learning time data; learning outcome data includes students' homework score data and test score data;

[0007] The data processing module conducts in-depth analysis of the initial teaching data set to determine the reference value of students' learning status and classify students;

[0008] A data storage module, used to store the initial teaching data set and the reference value of the student's learning status;

[0009] The access control module is used to analyze access data and establish user permissions.

[0010] Prioritized, the way to collect the original teaching data set is as follows:

[0011] Through the school information management system, collect student name data, student gender data, student grade and class data, student age data and student contact information data;

[0012] With the help of smart classroom monitoring system, students’ classroom participation data is collected, such as the number of times students speak, the number of questions they ask, the frequency of interaction with teachers and classmates, etc., and the data on students’ after-school study time is recorded through the online learning platform;

[0013] Obtain student homework and test score data from the school teaching evaluation system;

[0014] The preprocessing methods include data cleaning and data standardization to remove noise and outliers in the data, and unify the format of various types of data to obtain the initial teaching data set.

[0015] Preferred methods for analyzing the initial teaching dataset include:

[0016] Preset the grade learning ability threshold interval [G1, G2, G3, G4, G5, G6, G7, G8, G9, G10, G11, G12], substitute the student grade for comparison, and get the grade learning ability value. The grade learning ability threshold interval here corresponds to the student grade from the first grade of elementary school to the third grade of high school, and the value is a natural number from 1 to 12. Different grades correspond to different expected learning ability ranges. By comparing with the threshold interval of the student's grade, the relative learning ability level of the student in the same grade group is preliminarily evaluated;

[0017] A learning behavior health interval table is preset, and the students' classroom participation data and after-school learning time data are substituted into it for comparison to obtain the learning behavior health value. This interval table quantifies learning behavior according to its positive degree;

[0018] Preset the grade threshold [H1, H2, H3], substitute the student's homework grade data and test grade data for comparison, and get the grade performance value. The grade performance value judges the student's knowledge mastery based on the grade data, and takes a natural number from 0 to 2. For example, 0 means that the grade is poor and does not meet the basic requirements, 1 means that the grade is average and meets general expectations, and 2 means that the grade is excellent and exceeds expectations;

[0019] The threshold interval of the comprehensive learning index is preset [I 1, I2, I 3, I4], and the comprehensive learning index value Hi is obtained by substituting into the calculation formula: Hd = Xa*P1+Xb*P2+Xc*P3, where Xa, Xb and Xc are the grade learning ability value, learning behavior health value and performance value respectively, and P1, P2 and P3 are the preset weight factors respectively. The threshold of the comprehensive learning index here is quantified based on the comprehensive learning performance of the students, and the corresponding values ​​are 1, 2, 3, 4, four natural numbers, which are used to comprehensively evaluate the learning status of students;

[0020] By substituting into the calculation formula: Da=Ha*K1+(Ha+Hc)*K2+Hd*K3, the reference value of the student's learning status is obtained, where K1, K2 and K3 are preset weight factors respectively.

[0021] The preferred classification methods include:

[0022] The student reference threshold interval [Ja, Jb, Jc] is preset, and the student learning status reference value is brought into the comparison student reference threshold interval to obtain the student learning status classification value.

[0023] When the reference value of the student's learning status is greater than Ja and less than Jb, it is recorded as J1, and a basic improvement reminder SMS is sent to the teacher's receiving end. When the reference value of the student's learning status is greater than Jb and less than Jc, it is recorded as J2, and a reinforcement and consolidation reminder SMS is sent to the teacher's receiving end. When the reference value of the student's learning status is greater than Jc, a expansion and deepening reminder SMS is sent to the teacher's receiving end.

[0024] The basic improvement reminder text message includes explaining that students need to further strengthen certain basic knowledge or learning habits, and it is recommended that teachers provide more basic tutoring and learning method guidance, and require teachers to pay attention to subtle changes in students' learning status; the reinforcement and consolidation reminder text message includes explaining that students' learning status is relatively stable but there is still some knowledge that needs to be strengthened, and it is recommended that teachers provide targeted practice and review plans, and require teachers to check students' learning progress regularly; the expansion and deepening reminder text message includes explaining that students have strong learning ability, and it is recommended that teachers provide extended learning resources and challenging learning tasks.

[0025] Preferably, the steps of storing the initial teaching data set and the student learning status reference value include:

[0026] Based on the student information data, the initial teaching data set and the student learning status reference value are packaged to obtain the personal learning set;

[0027] Based on the students' grade and class data, the individual learning sets are divided into ordinary individual learning sets, focused individual learning sets and excellent individual learning sets. The basis for the division can be the overall learning level of the class and the relative position of the individual students in the class.

[0028] Based on the AES encryption algorithm, the key personal learning set and the excellent personal learning set are encrypted to obtain the ciphertext;

[0029] Based on the hash algorithm, the ciphertext is calculated to obtain the hash value of the ciphertext, and the hash value and the ciphertext are packaged and uploaded to the blockchain network and stored in the corresponding block.

[0030] Initially, the steps to establish user permissions include:

[0031] Connect to the school staff management system to obtain the personal information of all teachers, including the teacher's name, title and teacher photo, and divide each teacher's name, title and K teacher photos into a group;

[0032] Collect K teacher photos as the sample set, and divide the sample set into 70%K training set, 15%K test set and 15%K validation set;

[0033] Based on the image processing algorithm, the teacher photos in the sample set are preprocessed. The preprocessing methods include enlarging, reducing, rotating and cropping. The image feature data of each teacher photo is obtained through the image processing algorithm to obtain the initial teacher data set;

[0034] A classification model is established based on the initial teacher data set. The normal identity cluster N and the abnormal identity cluster C are preset. Two data points N and C are selected as the center of the first cluster. By substituting into the calculation formula: Get the distance between the data points, calculate the distance between the sub-data item and the first cluster center in the training set, assign the sub-data item to the first cluster center with the closest distance, and recalculate the mean of the two first cluster centers as the new preliminary cluster center for secondary calculation, where x and y are the coordinate values ​​of the data points, xi and yi are the values ​​of the two data points on the sub-data items respectively, and n is the number of sub-data items;

[0035] Through V rounds of iteration, the output results of the classification model tend to be stable;

[0036] Input the teacher photos in the training set into the classification model. When the output result is a normal identity cluster, it is verified as a teacher. When the output result is an abnormal identity cluster, it is verified as a non-teacher.

[0037] Preset access permission ranges, enter the job title in personal information, and divide it into junior teachers, intermediate teachers and senior teachers according to the job title. Access permissions include basic access permission, intermediate access permission and advanced access permission.

[0038] Based on the data storage module, junior teachers can query ordinary personal learning sets, intermediate teachers can query ordinary personal learning sets and focused personal learning sets, and senior teachers can query ordinary personal learning sets, focused personal learning sets and excellent personal learning sets;

[0039] The principal can enter the teaching data management system through face recognition and obtain the highest management authority. By sending a custom instruction to the central server, the central server can cancel the teacher's access rights after passing the face recognition verification;

[0040] Enter the personal information of all teachers and establish user permissions based on E2, E3 and E4.

[0041] Beneficial effects of the present invention: Through multi-source data collection and in-depth analysis, the present invention can accurately locate the learning status of students and provide teachers with a basis for personalized teaching strategies, such as pushing corresponding teaching suggestions based on student classification to improve the pertinence of teaching. Secondly, data storage uses encryption and blockchain technology to ensure data security and integrity, prevent information leakage and tampering, and maintain student privacy and the credibility of teaching data. Furthermore, the access control module establishes an identity model and divides permissions based on image processing, reasonably controls teacher data access, avoids abuse of permissions, and promotes standardized teaching management. The overall system realizes intelligent teaching management, reduces the burden of manual operations on teachers, improves teaching efficiency, and helps promote the improvement of education and teaching quality and the process of education informatization. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for describing the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0043] Figure 1 This is a system block diagram of an interactive teaching system integrating multimodal perception according to the present invention. DETAILED DESCRIPTION

[0044] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0045] See also Figure 1 As shown, the present invention is an interactive teaching system integrating multimodal perception, comprising:

[0046] The data collection module is responsible for collecting multi-source original teaching data sets and performing preliminary processing on them to obtain the initial teaching data sets. The original teaching data covers students' basic information data, learning process data, and learning outcome data. Students' basic information data includes students' name data, students' gender data, students' grade and class data, students' age data, and students' contact information data; learning process data includes students' class participation data and after-school learning time data; learning outcome data includes students' homework score data and test score data;

[0047] The method of collecting the original teaching data set is as follows:

[0048] Through the school information management system, collect student name data, student gender data, student grade and class data, student age data and student contact information data;

[0049] With the help of smart classroom monitoring system, students’ classroom participation data is collected, such as the number of times students speak, the number of questions they ask, the frequency of interaction with teachers and classmates, etc., and the data on students’ after-school study time is recorded through the online learning platform;

[0050] Obtain student homework and test score data from the school teaching evaluation system;

[0051] The preprocessing methods include data cleaning and data standardization to remove noise and outliers in the data, and unify the format of various types of data to obtain the initial teaching data set.

[0052] The data processing module conducts in-depth analysis of the initial teaching data set to determine the reference value of students' learning status and classify students;

[0053] The analysis of the initial teaching data set includes:

[0054] Preset the grade learning ability threshold interval [G1, G2, G3, G4, G5, G6, G7, G8, G9, G10, G11, G12], substitute the student grade for comparison, and get the grade learning ability value. The grade learning ability threshold interval here corresponds to the student grade from the first grade of elementary school to the third grade of high school, and the value is a natural number from 1 to 12. Different grades correspond to different expected learning ability ranges. By comparing with the threshold interval of the student's grade, the relative learning ability level of the student in the same grade group is preliminarily evaluated;

[0055] A learning behavior health interval table is preset, and the students' classroom participation data and after-school learning time data are substituted into it for comparison to obtain the learning behavior health value. This interval table quantifies learning behavior according to its positive degree;

[0056] Preset the grade threshold [H1, H2, H3], substitute the student's homework grade data and test grade data for comparison, and get the grade performance value. The grade performance value judges the student's knowledge mastery based on the grade data, and takes a natural number from 0 to 2. For example, 0 means that the grade is poor and does not meet the basic requirements, 1 means that the grade is average and meets general expectations, and 2 means that the grade is excellent and exceeds expectations;

[0057] The threshold interval of the comprehensive learning index is preset [I 1, I2, I 3, I4], and the comprehensive learning index value Hi is obtained by substituting into the calculation formula: Hd = Xa*P1+Xb*P2+Xc*P3, where Xa, Xb and Xc are the grade learning ability value, learning behavior health value and performance value respectively, and P1, P2 and P3 are the preset weight factors respectively. The threshold of the comprehensive learning index here is quantified based on the comprehensive learning performance of the students, and the corresponding values ​​are 1, 2, 3, 4, four natural numbers, which are used to comprehensively evaluate the learning status of students;

[0058] By substituting into the calculation formula: Da=Ha*K1+(Ha+Hc)*K2+Hd*K3, the reference value of the student's learning status is obtained, where K1, K2 and K3 are preset weight factors respectively.

[0059] The classification methods include:

[0060] The student reference threshold interval [Ja, Jb, Jc] is preset, and the student learning status reference value is brought into the comparison student reference threshold interval to obtain the student learning status classification value.

[0061] When the reference value of the student's learning status is greater than Ja and less than Jb, it is recorded as J1, and a basic improvement reminder SMS is sent to the teacher's receiving end. When the reference value of the student's learning status is greater than Jb and less than Jc, it is recorded as J2, and a reinforcement and consolidation reminder SMS is sent to the teacher's receiving end. When the reference value of the student's learning status is greater than Jc, a expansion and deepening reminder SMS is sent to the teacher's receiving end.

[0062] The basic improvement reminder text message includes explaining that students need to further strengthen certain basic knowledge or learning habits, and it is recommended that teachers provide more basic tutoring and learning method guidance, and require teachers to pay attention to subtle changes in students' learning status; the reinforcement and consolidation reminder text message includes explaining that students' learning status is relatively stable but there is still some knowledge that needs to be strengthened, and it is recommended that teachers provide targeted practice and review plans, and require teachers to check students' learning progress regularly; the expansion and deepening reminder text message includes explaining that students have strong learning ability, and it is recommended that teachers provide extended learning resources and challenging learning tasks.

[0063] A data storage module, used to store the initial teaching data set and the reference value of the student's learning status;

[0064] The steps of storing the initial teaching data set and the reference value of the student learning status include:

[0065] Based on the student information data, the initial teaching data set and the student learning status reference value are packaged to obtain the personal learning set;

[0066] Based on the students' grade and class data, the individual learning sets are divided into ordinary individual learning sets, focused individual learning sets and excellent individual learning sets. The basis for the division can be the overall learning level of the class and the relative position of the individual students in the class.

[0067] Based on the AES encryption algorithm, the key personal learning set and the excellent personal learning set are encrypted to obtain the ciphertext;

[0068] Based on the hash algorithm, the ciphertext is calculated to obtain the hash value of the ciphertext, and the hash value and the ciphertext are packaged and uploaded to the blockchain network and stored in the corresponding block.

[0069] The access control module is used to analyze access data and establish user permissions.

[0070] The steps to establish user permissions include:

[0071] Connect to the school staff management system to obtain the personal information of all teachers, including the teacher's name, title and teacher photo, and divide each teacher's name, title and K teacher photos into a group;

[0072] Collect K teacher photos as the sample set, and divide the sample set into 70%K training set, 15%K test set and 15%K validation set;

[0073] Based on the image processing algorithm, the teacher photos in the sample set are preprocessed. The preprocessing methods include enlarging, reducing, rotating and cropping. The image feature data of each teacher photo is obtained through the image processing algorithm to obtain the initial teacher data set;

[0074] A classification model is established based on the initial teacher data set. The normal identity cluster N and the abnormal identity cluster C are preset. Two data points N and C are selected as the center of the first cluster. By substituting into the calculation formula: Get the distance between the data points, calculate the distance between the sub-data item and the first cluster center in the training set, assign the sub-data item to the first cluster center with the closest distance, and recalculate the mean of the two first cluster centers as the new preliminary cluster center for secondary calculation, where x and y are the coordinate values ​​of the data points, xi and yi are the values ​​of the two data points on the sub-data items respectively, and n is the number of sub-data items;

[0075] Through V rounds of iteration, the output results of the classification model tend to be stable;

[0076] Input the teacher photos in the training set into the classification model. When the output result is a normal identity cluster, it is verified as a teacher. When the output result is an abnormal identity cluster, it is verified as a non-teacher.

[0077] Preset access permission ranges, enter the job title in personal information, and divide it into junior teachers, intermediate teachers and senior teachers according to the job title. Access permissions include basic access permission, intermediate access permission and advanced access permission.

[0078] Based on the data storage module, junior teachers can query ordinary personal learning sets, intermediate teachers can query ordinary personal learning sets and focused personal learning sets, and senior teachers can query ordinary personal learning sets, focused personal learning sets and excellent personal learning sets;

[0079] The principal can enter the teaching data management system through face recognition and obtain the highest management authority. By sending a custom instruction to the central server, the central server can cancel the teacher's access rights after passing the face recognition verification;

[0080] Enter the personal information of all teachers and establish user permissions based on E2, E3 and E4.

[0081] In the description of the specification, the description with reference to the terms "one embodiment", "example", "specific example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0082] The above contents are merely examples and explanations of the concept of the present invention. The technicians in this technical field may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined by the claims, they should all fall within the protection scope of the present invention.

Claims

1. An interactive teaching system integrating multimodal perception, characterized in that: include: The data collection module is responsible for collecting multi-source original teaching data sets and performing preliminary processing on them to obtain the initial teaching data sets. The original teaching data covers students' basic information data, learning process data, and learning outcome data. Students' basic information data includes students' name data, students' gender data, students' grade and class data, students' age data, and students' contact information data; learning process data includes students' class participation data and after-school learning time data; learning outcome data includes students' homework score data and test score data; The data processing module conducts in-depth analysis of the initial teaching data set to determine the reference value of students' learning status and classify students; A data storage module, used to store the initial teaching data set and the reference value of the student's learning status; The access control module is used to analyze access data and establish user permissions.

2. The interactive teaching system integrating multimodal perception according to claim 1, characterized in that: The original teaching data set is collected in the following way: through the school information management system, collect student name data, student gender data, student grade and class data, student age data and student contact information data; With the help of smart classroom monitoring system, students’ classroom participation data is collected, such as the number of times students speak, the number of questions they ask, the frequency of interaction with teachers and classmates, etc., and the data on students’ after-school study time is recorded through the online learning platform; Obtain student homework and test score data from the school teaching evaluation system; The preprocessing methods include data cleaning and data standardization to remove noise and outliers in the data, and unify the format of various types of data to obtain the initial teaching data set.

3. The interactive teaching system integrating multimodal perception according to claim 1, characterized in that: The analysis of the initial teaching data set includes: Preset the grade learning ability threshold interval [G1, G2, G3, G4, G5, G6, G7, G8, G9, G10, G11, G12], substitute the student grade for comparison, and get the grade learning ability value. The grade learning ability threshold interval here corresponds to the student grade from the first grade of elementary school to the third grade of high school, and the value is a natural number from 1 to 12. Different grades correspond to different expected learning ability ranges. By comparing with the threshold interval of the student's grade, the relative learning ability level of the student in the same grade group is preliminarily evaluated; A learning behavior health interval table is preset, and the students' classroom participation data and after-school learning time data are substituted into it for comparison to obtain the learning behavior health value. This interval table quantifies learning behavior according to its positive degree; Preset the grade threshold [H1, H2, H3], substitute the student's homework grade data and test grade data for comparison, and get the grade performance value. The grade performance value judges the student's knowledge mastery based on the grade data, and takes a natural number from 0 to 2. For example, 0 means that the grade is poor and does not meet the basic requirements, 1 means that the grade is average and meets general expectations, and 2 means that the grade is excellent and exceeds expectations; The threshold interval of the comprehensive learning index is preset [I 1, I2, I 3, I4], and the comprehensive learning index value Hi is obtained by substituting it into the calculation formula: Hd = Xa*P1+Xb*P2+Xc*P3, where Xa, Xb and Xc are the grade learning ability value, learning behavior health value and performance value respectively, and P1, P2 and P3 are the preset weight factors respectively. The threshold of the comprehensive learning index here is quantified based on the comprehensive learning performance of the students, and the corresponding values ​​are 1, 2, 3, 4, four natural numbers, which are used to comprehensively evaluate the learning status of students; By substituting into the calculation formula: Da=Ha*K1+(Ha+Hc)*K2+Hd*K3, the reference value of the student's learning status is obtained, where K1, K2 and K3 are preset weight factors respectively.

4. The interactive teaching system integrating multimodal perception according to claim 1, characterized in that: The classification methods include: The student reference threshold interval [Ja, Jb, Jc] is preset, and the student learning status reference value is brought into the comparison student reference threshold interval to obtain the student learning status classification value. When the reference value of the student's learning status is greater than Ja and less than Jb, it is recorded as J1, and a basic improvement reminder SMS is sent to the teacher's receiving end. When the reference value of the student's learning status is greater than Jb and less than Jc, it is recorded as J2, and a reinforcement and consolidation reminder SMS is sent to the teacher's receiving end. When the reference value of the student's learning status is greater than Jc, a expansion and deepening reminder SMS is sent to the teacher's receiving end. The basic improvement reminder text message includes explaining that students need to further strengthen certain basic knowledge or learning habits, and it is recommended that teachers provide more basic tutoring and learning method guidance, and require teachers to pay attention to subtle changes in students' learning status; the reinforcement and consolidation reminder text message includes explaining that students' learning status is relatively stable but there is still some knowledge that needs to be strengthened, and it is recommended that teachers provide targeted practice and review plans, and require teachers to check students' learning progress regularly; the expansion and deepening reminder text message includes explaining that students have strong learning ability, and it is recommended that teachers provide extended learning resources and challenging learning tasks.

5. The interactive teaching system integrating multimodal perception according to claim 1, characterized in that: The steps of storing the initial teaching data set and the reference value of the student learning status include: Based on the student information data, the initial teaching data set and the student learning status reference value are packaged to obtain the personal learning set; Based on the students' grade and class data, the individual learning sets are divided into ordinary individual learning sets, focused individual learning sets and excellent individual learning sets. The basis for the division can be the overall learning level of the class and the relative position of the individual students in the class. Based on the AES encryption algorithm, the key personal learning set and the excellent personal learning set are encrypted to obtain the ciphertext; Based on the hash algorithm, the ciphertext is calculated to obtain the hash value of the ciphertext, and the hash value and the ciphertext are packaged and uploaded to the blockchain network and stored in the corresponding block.

6. The interactive teaching system integrating multimodal perception according to claim 1, characterized in that: The steps to establish user permissions include: Connect to the school staff management system to obtain the personal information of all teachers, including the teacher's name, title and teacher photo, and divide each teacher's name, title and K teacher photos into a group; Collect K teacher photos as the sample set, and divide the sample set into 70%K training set, 15%K test set and 15%K validation set; Based on the image processing algorithm, the teacher photos in the sample set are preprocessed. The preprocessing methods include enlarging, reducing, rotating and cropping. The image feature data of each teacher photo is obtained through the image processing algorithm to obtain the initial teacher data set; A classification model is established based on the initial teacher data set. The normal identity cluster N and the abnormal identity cluster C are preset. Two data points N and C are selected as the center of the first cluster. By substituting into the calculation formula: Get the distance between the data points, calculate the distance between the sub-data item and the first cluster center in the training set, assign the sub-data item to the first cluster center with the closest distance, and recalculate the mean of the two first cluster centers as the new preliminary cluster center for secondary calculation, where x and y are the coordinate values ​​of the data points, xi and yi are the values ​​of the two data points on the sub-data items respectively, and n is the number of sub-data items; Through V rounds of iteration, the output results of the classification model tend to be stable; Input the teacher photos in the training set into the classification model. When the output result is a normal identity cluster, it is verified as a teacher. When the output result is an abnormal identity cluster, it is verified as a non-teacher. Preset access permission ranges, enter the job title in personal information, and divide it into junior teachers, intermediate teachers and senior teachers according to the job title. Access permissions include basic access permission, intermediate access permission and advanced access permission. Based on the data storage module, junior teachers can query ordinary personal learning sets, intermediate teachers can query ordinary personal learning sets and focused personal learning sets, and senior teachers can query ordinary personal learning sets, focused personal learning sets and excellent personal learning sets; The principal can enter the teaching data management system through face recognition and obtain the highest management authority. By sending a custom instruction to the central server, the central server can cancel the teacher's access rights after passing the face recognition verification; Enter the personal information of all teachers and establish user permissions based on E2, E3 and E4.

Images