A data analysis-based education remote monitoring management system and method

Abstract

The present invention discloses an education remote monitoring management system and method based on data analysis, which relates to the technical field of education monitoring management. The method discloses the following steps: Step 1: After the teacher starts remote teaching, the teacher's teaching explanation is recorded to obtain multiple groups of teaching sentences; Step 2: The multiple groups of teaching sentences are divided into multiple teaching monitoring segments; Step 3: The teaching emotion value of each teaching monitoring segment is obtained, and the teaching monitoring segment is marked as an efficient teaching segment or an inefficient teaching segment according to the comparison result of the teaching emotion value and its threshold value; Step 4: The efficient teaching segment and the inefficient teaching segment are managed. The method of the present invention intelligently divides the teacher's remote teaching audio into monitoring segments through natural language processing, and can mark the teaching monitoring segment as an efficient teaching segment or an inefficient teaching segment according to the comparison result of the teaching emotion value and its threshold value, and dynamically manages the teacher and the remote students according to the marking situation.

Description

Technical Field

[0001] The present invention relates to the technical field of education monitoring management, and more specifically, to an education remote monitoring management system and method based on data analysis. Background Art

[0002] With the rapid development of information technology, educational methods are becoming increasingly diverse, with an increasing number of schools opting for online distance education. Online distance education offers the characteristics of sharing, openness, and autonomy. However, current online distance education teachers cannot fully understand each student's learning progress on each knowledge point, resulting in low teaching efficiency at some teaching locations. Furthermore, the Academic Affairs Office is unable to keep up with each teacher's teaching progress, making it difficult to control teaching quality. Therefore, effective remote monitoring of education is a pressing issue that needs to be addressed. Summary of the Invention

[0003] In view of the shortcomings of the existing technology, the purpose of the present invention is to provide an education remote monitoring management system and method based on data analysis.

[0004] To achieve the above object, the present invention provides the following technical solutions:

[0005] An education remote monitoring and management system based on data analysis, comprising a lecture recording segmentation module, a lecture sentence processing module, a lecture segment division module, and an education monitoring and management module;

[0006] The teaching recording segmentation module records the teacher's teaching after the teacher starts remote teaching to obtain multiple groups of teaching sentences;

[0007] The teaching statement processing module is used to divide multiple groups of teaching statements into multiple teaching monitoring segments;

[0008] The teaching segment division module obtains the teaching emotion value of each teaching monitoring segment, and marks the teaching monitoring segment as an efficient teaching segment or an inefficient teaching segment according to the comparison result of the teaching emotion value and its threshold value;

[0009] The education monitoring management module manages efficient teaching segments and inefficient teaching segments.

[0010] Furthermore, the teaching sentences are obtained in the following way: the teaching audio is generated in real time, and the recordings in the teaching audio are segmented and processed by natural language processing to obtain multiple groups of teaching sentences, which are marked as DGi, where i is the sequential number of the teaching sentence, i=1, 2, ..., N, and N is a positive integer.

[0011] Furthermore, multiple groups of teaching statements are divided into multiple teaching monitoring segments, specifically: obtaining the teaching vector value of each group of teaching statements, obtaining the course association value of adjacent numbered teaching statements, setting the course association threshold, when the course association value is greater than or equal to the course association threshold, the adjacent numbered teaching statements are classified into the same teaching monitoring segment, when the course association value is less than the course association threshold, the adjacent numbered teaching statements are divided into two teaching monitoring segments.

[0012] Furthermore, the teaching vector value of each group of teaching sentences is obtained in the following way: the word vector of each word in the teaching sentence is obtained through the Word2Vec model, and the word vector of each word in the teaching sentence is summed up to obtain the teaching vector value.

[0013] Furthermore, the course association value of adjacent numbered teaching sentences is obtained by obtaining the teaching vector value of the adjacent numbered teaching sentences, performing cosine similarity calculation on the teaching vector value of the adjacent numbered teaching sentences, and obtaining the course association value of the adjacent numbered teaching sentences.

[0014] Furthermore, the teaching emotion value of the teaching monitoring segment is obtained by the following method: obtain all the teaching sentences in the teaching monitoring segment, use the corresponding audio of the teaching sentences as the input data of the teaching emotion mobilization model to obtain the teaching emotion mobilization value, sum up all the teaching emotion mobilization values ​​and take the average to obtain the average teaching emotion mobilization value, and mark it as HCZ, obtain the total number of emotion mobilization changes, and mark it as GFS, obtain the average emotion richness mobilization density, and mark it as WSE, and use the formula Get the teaching emotion value of the teaching monitoring segment.

[0015] Furthermore, the average emotion-rich mobilization density is obtained in the following way: a teaching emotion mobilization threshold is set; when the teaching emotion mobilization value is greater than or equal to the teaching emotion mobilization threshold, the teaching emotion mobilization value is marked as an emotion-rich mobilization value; when the teaching emotion mobilization value is less than the teaching emotion mobilization threshold, no corresponding processing is performed; all emotion-rich mobilization values ​​are sorted in the order of the numbering of the teaching sentences; the difference between the numbers of the two adjacent teaching sentences after sorting is calculated to obtain the emotion-rich mobilization density; all emotion-rich mobilization densities are summed and the average is taken to obtain the average emotion-rich mobilization density.

[0016] Furthermore, the total number of mobilization changes is obtained in the following way: all teaching emotion mobilization values ​​are sorted in the order of the number of the teaching sentences, the difference between the two adjacent teaching emotion mobilization values ​​after sorting is calculated and the absolute value is taken to obtain the emotion mobilization change value, and the emotion mobilization change threshold is set. When the emotion mobilization change value is greater than or equal to the emotion mobilization change threshold, the number of emotion mobilization changes in the teaching monitoring segment is increased by one, and all the emotion mobilization change numbers in the teaching monitoring segment are summed up to obtain the total number of emotion mobilization changes. When the emotion mobilization change value is less than the emotion mobilization change threshold, no corresponding processing is performed.

[0017] Furthermore, efficient and inefficient teaching segments are managed as follows:

[0018] When the teaching monitoring segment is marked as an inefficient teaching segment, a warning message of the inefficient teaching segment is sent to the teacher, and an inefficient warning record is generated synchronously. The inefficient warning record includes the warning time. All inefficient warning records of the teacher before the current system time are obtained, and all inefficient warning records are sorted in the order of the warning time. The time difference between the two adjacent warning times after sorting is calculated to obtain the warning interval. All warning intervals are summed and averaged to obtain the average warning interval, which is marked as YED. The total number of inefficient warning records is obtained and marked as SZK. The formula is used Get the teacher's education evaluation value QWA, where c1 is the average warning interval coefficient and c2 is the total number of inefficient warning records coefficient. Set the education evaluation threshold. When the education evaluation value is greater than or equal to the education evaluation threshold, mark the teacher as an inefficient education teacher and upload the information of the inefficient education teacher to the school's academic affairs system.

[0019] When the teaching monitoring segment is marked as an efficient teaching segment, the teaching monitoring start time and teaching monitoring end time of the efficient teaching segment are obtained, and the lecture video of each remote lecturer is intercepted according to the teaching monitoring start time and the teaching monitoring end time. The effective teaching value of each remote lecturer is obtained according to the lecture video, and an effective teaching threshold is set. When the effective teaching value is less than the effective teaching threshold, the remote lecturer is marked as an invalid lecturer, and the invalid lecturers are sorted from small to large according to the effective teaching value. After sorting, an education attention table for the efficient teaching segment is generated, and the teacher can focus on teaching the content of the efficient teaching segment to the invalid lecturers according to the education attention table;

[0020] The effective teaching value of remote teaching students is obtained by the following method: converting the intercepted teaching video into video frame images, further processing the video frame images, extracting the eye image data of the remote teaching students in the video frame images, obtaining the listening concentration value of the eye image data, sorting the video frame images in chronological order, obtaining the listening concentration value of each video frame image, setting the listening concentration threshold, when the listening concentration value is greater than or equal to the listening concentration threshold, marking the video frame image as a valid teaching image, obtaining the total number of valid teaching images and marking it as PLK, when the listening concentration value is less than the listening concentration threshold, marking the video frame image as an invalid teaching image, sorting all invalid teaching images in chronological order, calculating the time difference between the two adjacent invalid teaching images after sorting, obtaining the invalid teaching interval, summing up all invalid teaching intervals and taking the average, obtaining the average invalid teaching interval, and marking it as MBZ, using the formula The effective teaching value HVL of the remote teaching students is obtained, where b1 is the total number coefficient of effective teaching images and b2 is the average invalid teaching interval coefficient;

[0021] The lecture concentration value of the eye image data is obtained by the following method: using the eye image data as input data of the eye analysis model to obtain the lecture concentration value of the eye image data;

[0022] Get the teaching sentence that ranks first in the efficient teaching segment, mark the time of the first word in the teaching sentence that ranks first as the start time of teaching monitoring; get the teaching sentence that ranks last in the efficient teaching segment, mark the time of the last word in the teaching sentence that ranks last as the start time of teaching monitoring.

[0023] Furthermore, a method for remote monitoring and management of education based on data analysis includes the following steps:

[0024] Step 1: After the teacher starts teaching remotely, record the teacher's lectures and obtain multiple sets of teaching sentences;

[0025] Step 2: Divide multiple groups of teaching sentences into multiple teaching monitoring segments;

[0026] Step 3: Obtain the teaching emotion value of each teaching monitoring segment, and mark the teaching monitoring segment as an efficient teaching segment or an inefficient teaching segment based on the comparison result of the teaching emotion value and its threshold;

[0027] Step 4: Manage efficient and inefficient teaching segments.

[0028] Compared with the prior art, the present invention has the following beneficial effects:

[0029] 1. The method of the present invention uses natural language processing to intelligently divide the teacher's remote teaching audio into monitoring segments. The teaching monitoring segments can be marked as efficient teaching segments or inefficient teaching segments based on the comparison results of the teaching emotion value and its threshold value, and the teacher and remote students can be dynamically managed and controlled based on the marking situation.

[0030] 2. Set up a teaching recording segmentation module and a teaching sentence processing module, which can segment and reassemble the teacher's remote teaching audio through natural language processing, and intelligently divide the teacher's remote teaching audio into monitoring segments, so as to facilitate subsequent targeted intelligent analysis of different monitoring segments. Setting up a teaching segment division module and an education monitoring management module can allow teachers to fully understand the listening situation of each remote student on different knowledge points, and also allow the Academic Affairs Office to fully understand the teacher's teaching situation, thereby improving the supervision quality and efficiency of remote teaching. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 This is a flowchart of an educational remote monitoring and management method based on data analysis;

[0032] Figure 2 A flowchart of the process divided into teaching and monitoring sections;

[0033] Figure 3 A flowchart for managing efficient and inefficient teaching segments. DETAILED DESCRIPTION Example 1

[0034] Reference Figure 1 , a remote monitoring and management method for education based on data analysis, comprising the following steps:

[0035] Step 1: After the teacher starts teaching remotely, record the teacher's lectures and obtain multiple sets of teaching sentences;

[0036] Step 2: Divide multiple groups of teaching sentences into multiple teaching monitoring segments;

[0037] Step 3: Obtain the teaching emotion value of each teaching monitoring segment, and mark the teaching monitoring segment as an efficient teaching segment or an inefficient teaching segment based on the comparison result of the teaching emotion value and its threshold;

[0038] Step 4: Manage efficient and inefficient teaching segments.

[0039] The above method can intelligently divide the teacher's remote teaching audio into monitoring segments through natural language processing, and can mark the teaching monitoring segments as efficient teaching segments or inefficient teaching segments based on the comparison results of the teaching emotion value and its threshold, and dynamically manage the teacher and remote students based on the marking situation. Example 2

[0040] Reference Figures 2 to 3 , an education remote monitoring and management system based on data analysis, including a teaching recording segmentation module, a teaching sentence processing module, a teaching segment division module, and an education monitoring management module.

[0041] The teaching recording segmentation module records the teacher's lectures after the teacher starts remote teaching, generates teaching audio in real time, and segments the recordings in the teaching audio through natural language processing to obtain multiple groups of teaching sentences, which are marked as DGi, where i is the sequential number of the teaching sentences, i=1, 2, ..., N, and N is a positive integer.

[0042] The teaching statement processing module obtains the teaching vector value of each group of teaching statements and obtains the course association value of the teaching statements with adjacent numbers.

[0043] Example: In a lecture audio, the audio reads, "Good afternoon, students. Before class today, the teacher would like to show you some pictures. After you look at the pictures, the teacher would like to ask you what you see in them." Natural language processing is used to segment the lecture audio into two groups of sentences: "Good afternoon, students. Before class today, the teacher would like to show you some pictures," and "After you look at the pictures, the teacher would like to ask you what you see in them." The sentence "Good afternoon, students. Before class today, the teacher would like to show you some pictures" is labeled DG1, and the sentence "After you look at the pictures, the teacher would like to ask you what you see in them" is labeled DG2.

[0044] The teaching vector value for each set of teaching sentences is obtained as follows: the word vector of each word in the teaching sentence is obtained through the Word2Vec model, and the word vector value is obtained by summing up the word vectors of each word in the teaching sentence. The word vector of each word is obtained by training the Word2Vec model with a large amount of teaching sentence data.

[0045] The course association values ​​of adjacent numbered teaching sentences are obtained by obtaining the teaching vector values ​​of adjacent numbered teaching sentences, performing cosine similarity calculation on the teaching vector values ​​of adjacent numbered teaching sentences, and obtaining the course association values ​​of adjacent numbered teaching sentences.

[0046] Set the course association threshold. The course association threshold is a system preset value and can be modified according to actual needs.

[0047] When the course association value is greater than or equal to the course association threshold, the teaching statements with adjacent numbers are classified as the same teaching monitoring segment.

[0048] When the course association value is less than the course association threshold, the adjacent numbered teaching statements are divided into two teaching monitoring segments.

[0049] Setting up a teaching recording segmentation module and a teaching sentence processing module can segment and reassemble the teacher's remote teaching audio through natural language processing, and intelligently divide the teacher's remote teaching audio into monitoring segments, which is convenient for subsequent targeted intelligent analysis of different monitoring segments.

[0050] The teaching segment division module obtains the teaching emotion value of each teaching monitoring segment and sets the teaching emotion threshold. The teaching emotion threshold is a system preset value and can be modified according to actual needs. When the teaching emotion value is greater than or equal to the teaching emotion threshold, the teaching monitoring segment is marked as an efficient teaching segment.

[0051] When the teaching emotion value is less than the teaching emotion threshold, the teaching monitoring segment is marked as an inefficient teaching segment.

[0052] When the teaching monitoring segment is marked as an inefficient teaching segment, the education monitoring management module sends a warning message of the inefficient teaching segment to the teacher and simultaneously generates an inefficient warning record. The inefficient warning record includes the warning time, obtains all inefficient warning records of the teacher before the current time of the system, sorts all inefficient warning records in the order of the warning time, calculates the time difference between the two adjacent warning times after sorting, obtains the warning interval, sums up all the warning intervals and takes the average to obtain the average warning interval, and marks it as YED, obtains the total number of inefficient warning records, and marks it as SZK, and uses the formula Get the teacher's education evaluation QWA, where c1 is the average warning interval coefficient and c2 is the total number coefficient of inefficient warning records. The value of c1 is 0.57 and the value of c2 is 0.62. Set the education evaluation threshold. The education evaluation threshold is a system preset value and can be modified according to actual needs. When the education evaluation value is greater than or equal to the education evaluation threshold, the teacher will be marked as an inefficient education teacher and the information of the inefficient education teacher will be uploaded to the school's academic affairs system. When the education evaluation value is less than the education evaluation threshold, no corresponding processing will be performed.

[0053] When the teaching monitoring segment is marked as an efficient teaching segment, the teaching monitoring start time and teaching monitoring end time of the efficient teaching segment are obtained, and the lecture video of each remote student is intercepted according to the teaching monitoring start time and the teaching monitoring end time. The effective teaching value of each remote student is obtained according to the lecture video, and the effective teaching threshold is set. The effective teaching threshold is a system preset value and can be modified according to actual needs. When the effective teaching value is greater than or equal to the effective teaching threshold, no corresponding processing is performed. When the effective teaching value is less than the effective teaching threshold, the remote student is marked as an invalid student, and the invalid students are sorted from small to large according to the effective teaching value value. After sorting, an education attention table for the efficient teaching segment is generated. The teacher can focus on teaching the content of the efficient teaching segment to the invalid students according to the education attention table.

[0054] The effective teaching value of remote teaching students is obtained by the following method: converting the intercepted teaching video into video frame images, further processing the video frame images, extracting the eye image data of the remote teaching students in the video frame images, obtaining the listening concentration value of the eye image data, sorting the video frame images in chronological order, obtaining the listening concentration value of each video frame image, setting the listening concentration threshold, when the listening concentration value is greater than or equal to the listening concentration threshold, marking the video frame image as a valid teaching image, obtaining the total number of valid teaching images and marking it as PLK, when the listening concentration value is less than the listening concentration threshold, marking the video frame image as an invalid teaching image, sorting all invalid teaching images in chronological order, calculating the time difference between the two adjacent invalid teaching images after sorting, obtaining the invalid teaching interval, summing up all invalid teaching intervals and taking the average, obtaining the average invalid teaching interval, and marking it as MBZ, using the formula The effective teaching value HVL of the remote teaching students is obtained, where b1 is the total number coefficient of effective teaching images, b2 is the average invalid teaching interval coefficient, the value of b1 is 0.76, and the value of b2 is 0.43.

[0055] The lecture concentration value of the eye expression image data is obtained by the following method: the eye expression image data is used as input data of the eye expression analysis model to obtain the lecture concentration value of the eye expression image data.

[0056] The eye gaze analysis model is developed by obtaining multiple eye gaze image data sets collected from students attending remote classes during actual remote instruction. This data set is then used to construct an eye gaze image dataset. Different eye gaze images are assigned different attention values. The dataset is then divided into a training set and a validation set. The eye gaze analysis model is then developed through iterative neural network training. A higher attention value indicates a more focused eye.

[0057] The teaching emotion value of the teaching monitoring segment is obtained by the following method: obtain all the teaching sentences in the teaching monitoring segment, use the corresponding audio of the teaching sentences as the input data of the teaching emotion mobilization model to obtain the teaching emotion mobilization value, sum up all the teaching emotion mobilization values ​​and take the average to obtain the average teaching emotion mobilization value, and mark it as HCZ, obtain the total number of emotion mobilization changes, and mark it as GFS, obtain the average emotion richness mobilization density, and mark it as WSE, and use the formula The teaching emotion values ​​of the teaching monitoring segment are obtained, and the value of a1 is 0.53, the value of a2 is 0.39, and the value of a3 is 1.02.

[0058] The total number of mobilization changes is obtained in the following way: sort all teaching emotion mobilization values ​​according to the numbering order of the teaching sentences, calculate the difference between the two adjacent teaching emotion mobilization values ​​after sorting and take the absolute value to obtain the emotion mobilization change value, set the emotion mobilization change threshold, the emotion mobilization change threshold is a system preset value, and can be modified according to actual needs. When the emotion mobilization change value is greater than or equal to the emotion mobilization change threshold, the number of emotion mobilization changes in the teaching monitoring segment is increased by one, and all the emotion mobilization change times in the teaching monitoring segment are summed up to obtain the total number of emotion mobilization changes. When the emotion mobilization change value is less than the emotion mobilization change threshold, no corresponding processing is performed.

[0059] The average emotion-rich mobilization density is obtained in the following way: setting the teaching emotion mobilization threshold, which is a system preset value and can be modified according to actual needs; when the teaching emotion mobilization value is greater than or equal to the teaching emotion mobilization threshold, the teaching emotion mobilization value is marked as an emotion-rich mobilization value; when the teaching emotion mobilization value is less than the teaching emotion mobilization threshold, no corresponding processing is performed; all emotion-rich mobilization values ​​are sorted in the order of the numbering of the teaching sentences; the difference between the numbers of the two adjacent teaching sentences after sorting is calculated to obtain the emotion-rich mobilization density; all emotion-rich mobilization densities are summed and the average is taken to obtain the average emotion-rich mobilization density.

[0060] The teaching emotion mobilization model is obtained in the following way: multiple sets of teaching sentence audios are obtained, and the teaching sentence audios are all audios intercepted from the teacher's previous remote teaching. The teaching sentence audios are used as training data for the convolutional neural network model, and the training data is assigned a teaching emotion mobilization value. The training data is iteratively trained through the training set and the validation set to obtain the teaching emotion mobilization model. The larger the teaching emotion mobilization value, the richer the teacher's emotions in the teaching sentence.

[0061] Get the teaching sentence that ranks first in the efficient teaching segment, mark the time of the first word in the teaching sentence that ranks first as the start time of teaching monitoring; get the teaching sentence that ranks last in the efficient teaching segment, mark the time of the last word in the teaching sentence that ranks last as the start time of teaching monitoring.

[0062] The camera of the remote student is pointed at the remote student, and the video of the class can be obtained through the camera.

[0063] Setting up the teaching segment division module and the education monitoring management module can allow teachers to fully understand the listening situation of each remote student on different knowledge points, and can also allow the Academic Affairs Office to fully understand the teacher's teaching situation, thereby improving the supervision quality and efficiency of remote teaching.

[0064] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formulas are set by technicians in this field according to actual conditions.

[0065] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer program are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center via wired or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0066] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0067] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0068] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0069] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0070] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0071] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. An education remote monitoring and management system based on data analysis, characterized in that: It includes lecture recording segmentation module, lecture sentence processing module, lecture segment division module, and education monitoring and management module; The teaching recording segmentation module records and segments the teacher's teaching after the teacher starts remote teaching to obtain multiple groups of teaching sentences; The teaching statement processing module is used to divide multiple groups of teaching statements into multiple teaching monitoring segments; The teaching segment division module obtains the teaching emotion value of each teaching monitoring segment, and marks the teaching monitoring segment as an efficient teaching segment or an inefficient teaching segment according to the comparison result of the teaching emotion value and its threshold value; The education monitoring management module manages efficient and inefficient teaching segments; Manage efficient and inefficient teaching segments, specifically: When the teaching monitoring segment is marked as an inefficient teaching segment, a warning message of the inefficient teaching segment is sent to the teacher, and an inefficient warning record is generated synchronously. The inefficient warning record includes the warning time. All inefficient warning records of the teacher before the current system time are obtained, and all inefficient warning records are sorted in the order of the warning time. The time difference between the two adjacent warning times after sorting is calculated to obtain the warning interval. All warning intervals are summed and averaged to obtain the average warning interval, which is marked as YED. The total number of inefficient warning records is obtained and marked as SZK. The formula is used Get the teacher's education evaluation value QWA, where c1 is the average warning interval coefficient and c2 is the total number of inefficient warning records coefficient. Set the education evaluation threshold. When the education evaluation value is greater than or equal to the education evaluation threshold, mark the teacher as an inefficient education teacher and upload the information of the inefficient education teacher to the school's academic affairs system. When the teaching monitoring segment is marked as an efficient teaching segment, the teaching monitoring start time and teaching monitoring end time of the efficient teaching segment are obtained, and the lecture video of each remote lecturer is intercepted according to the teaching monitoring start time and the teaching monitoring end time. The effective teaching value of each remote lecturer is obtained according to the lecture video, and an effective teaching threshold is set. When the effective teaching value is less than the effective teaching threshold, the remote lecturer is marked as an invalid lecturer, and the invalid lecturers are sorted from small to large according to the effective teaching value. After sorting, an education attention table for the efficient teaching segment is generated, and the teacher can focus on teaching the content of the efficient teaching segment to the invalid lecturers according to the education attention table; The effective teaching value of remote teaching students is obtained by the following method: converting the intercepted teaching video into video frame images, further processing the video frame images, extracting the eye image data of the remote teaching students in the video frame images, obtaining the listening concentration value of the eye image data, sorting the video frame images in chronological order, obtaining the listening concentration value of each video frame image, setting the listening concentration threshold, when the listening concentration value is greater than or equal to the listening concentration threshold, marking the video frame image as a valid teaching image, obtaining the total number of valid teaching images and marking it as PLK, when the listening concentration value is less than the listening concentration threshold, marking the video frame image as an invalid teaching image, sorting all invalid teaching images in chronological order, calculating the time difference between the two adjacent invalid teaching images after sorting, obtaining the invalid teaching interval, summing up all invalid teaching intervals and taking the average, obtaining the average invalid teaching interval, and marking it as MBZ, using the formula The effective teaching value HVL of the remote teaching students is obtained, where b1 is the total number coefficient of effective teaching images and b2 is the average invalid teaching interval coefficient; The lecture concentration value of the eye image data is obtained by the following method: using the eye image data as input data of the eye analysis model to obtain the lecture concentration value of the eye image data; Get the teaching sentence that ranks first in the efficient teaching segment, mark the time of the first word in the teaching sentence that ranks first as the start time of teaching monitoring, get the teaching sentence that ranks last in the efficient teaching segment, mark the time of the last word in the teaching sentence that ranks last as the end time of teaching monitoring.

2. The education remote monitoring and management system based on data analysis according to claim 1 is characterized in that: The teaching sentences are obtained in the following way: the teaching audio is generated in real time, and the recordings in the teaching audio are segmented by natural language processing to obtain multiple groups of teaching sentences, which are marked as DGi, where i is the sequential number of the teaching sentence, i=1, 2, ..., N, and N is a positive integer.

3. The education remote monitoring and management system based on data analysis according to claim 1 is characterized in that: Multiple groups of teaching statements are divided into multiple teaching monitoring segments. Specifically, the teaching vector value of each group of teaching statements is obtained, the course association value of adjacent numbered teaching statements is obtained, and a course association threshold is set. When the course association value is greater than or equal to the course association threshold, the adjacent numbered teaching statements are classified into the same teaching monitoring segment. When the course association value is less than the course association threshold, the adjacent numbered teaching statements are divided into two teaching monitoring segments.

4. The education remote monitoring and management system based on data analysis according to claim 3 is characterized in that: The teaching vector value of each group of teaching sentences is obtained in the following way: the word vector of each word in the teaching sentence is obtained through the Word2Vec model, and the word vector of each word in the teaching sentence is summed up to obtain the teaching vector value.

5. The education remote monitoring and management system based on data analysis according to claim 3 is characterized in that: The course association values ​​of adjacent numbered teaching sentences are obtained by obtaining the teaching vector values ​​of adjacent numbered teaching sentences, performing cosine similarity calculation on the teaching vector values ​​of adjacent numbered teaching sentences, and obtaining the course association values ​​of adjacent numbered teaching sentences.

6. The education remote monitoring and management system based on data analysis according to claim 1, characterized in that: The teaching emotion value of the teaching monitoring segment is obtained by the following method: obtain all the teaching sentences in the teaching monitoring segment, use the corresponding audio of the teaching sentences as the input data of the teaching emotion mobilization model to obtain the teaching emotion mobilization value, sum up all the teaching emotion mobilization values ​​and take the average to obtain the average teaching emotion mobilization value, and mark it as HCZ, obtain the total number of emotion mobilization changes, and mark it as GFS, obtain the average emotion richness mobilization density, and mark it as WSE, and use the formula The teaching emotion value of the teaching monitoring segment is obtained, a1 is the average teaching emotion mobilization value coefficient, a2 is the total number of emotion mobilization changes coefficient, and a3 is the average emotion rich mobilization density coefficient.

7. The education remote monitoring and management system based on data analysis according to claim 6 is characterized in that: The average emotion-rich mobilization density is obtained in the following way: setting a teaching emotion mobilization threshold, when the teaching emotion mobilization value is greater than or equal to the teaching emotion mobilization threshold, marking the teaching emotion mobilization value as an emotion-rich mobilization value, when the teaching emotion mobilization value is less than the teaching emotion mobilization threshold, no corresponding processing is performed, all emotion-rich mobilization values ​​are sorted according to the numbering order of the teaching sentences, and the difference between the numbers of the two adjacent teaching sentences after sorting is calculated to obtain the emotion-rich mobilization density, and all emotion-rich mobilization densities are summed and averaged to obtain the average emotion-rich mobilization density.

8. The education remote monitoring and management system based on data analysis according to claim 6 is characterized in that: The total number of mobilization changes is obtained in the following way: all teaching emotion mobilization values ​​are sorted in the order of the number of the teaching sentences, the difference between the two adjacent teaching emotion mobilization values ​​after sorting is calculated and the absolute value is taken to obtain the emotion mobilization change value, and the emotion mobilization change threshold is set. When the emotion mobilization change value is greater than or equal to the emotion mobilization change threshold, the number of emotion mobilization changes in the teaching monitoring segment is increased by one, and all the emotion mobilization change numbers in the teaching monitoring segment are summed up to obtain the total number of emotion mobilization changes. When the emotion mobilization change value is less than the emotion mobilization change threshold, no corresponding processing is performed.

9. A data analysis-based education remote monitoring and management method, applied to the data analysis-based education remote monitoring and management system according to claim 1, characterized in that: The steps include: Step 1: After the teacher starts teaching remotely, record and segment the teacher's lecture to obtain multiple groups of teaching sentences; Step 2: Divide multiple groups of teaching sentences into multiple teaching monitoring segments; Step 3: Obtain the teaching emotion value of each teaching monitoring segment, and mark the teaching monitoring segment as an efficient teaching segment or an inefficient teaching segment based on the comparison result of the teaching emotion value and its threshold; Step 4: Manage efficient and inefficient teaching segments.

Images