CN113076835B - Teaching Evaluation Method and System Based on Regression Analysis - Google Patents

Abstract

The invention provides a regression analysis-based teaching evaluation method and a regression analysis-based teaching evaluation system, which are used for acquiring the adjacent two-time score ranking of students to be evaluated and all student panoramic video data of time periods to be evaluated; obtaining a psychological state recognition result of each student according to the acquired panoramic video data of all students and a preset student emotional expression recognition model; ranking and regression analysis are carried out according to the adjacent two achievements to obtain a ranking increment model, and the ranking increment model and the average basic ranking of all students are combined to obtain a ranking increment after the basic influence is removed; weighting and summing the acquired ranking increment without the basic influence and the mental state recognition result of the student to obtain a final evaluation result; the method and the device eliminate the influence of the basic difference of the teaching class on the increment evaluation, combine the psychological state evaluation result of each class of each student, and realize more accurate and comprehensive teaching evaluation.

Description

Teaching Evaluation Method and System Based on Regression Analysis

technical field

The present disclosure relates to the technical field of teaching evaluation, in particular to a method and system for teaching evaluation based on regression analysis.

Background technique

The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art.

Teaching evaluation is an important part of teaching management. Scientific and effective teaching evaluation can reasonably guide teachers' work direction and improve teachers' work enthusiasm, thereby promoting the improvement of education quality and the development of students. Establishing a scientific and reasonable teaching evaluation system is crucial to the quality of education for teachers themselves, students, and the entire country. However, due to different teaching objects and different teaching subjects, the teaching evaluation of teachers is a common problem faced by schools at present. Especially since the reform of the new college entrance examination, the difference in the student base caused by the "walking classes" of students has been too large, which has brought challenges to teacher evaluation and assessment.

At present, the teaching evaluation method system is gradually improving, and the incremental evaluation of rankings has been widely accepted. Rank incremental evaluation is a way to evaluate teachers' teaching based on the increment between the current rank and the basic rank. However, the increase in ranking is also affected by the basic ranking of the students. It is difficult for the students who rank high to obtain the increase in rank, while the students who rank lower are relatively easy to obtain the increase in rank. If the class taught by a teacher has the best students in the grade, no matter how good the teacher is and how hard he works, the increase in the average rank of the class taught by the teacher cannot be positive; on the contrary, if the class taught by the teacher concentrates the students in the worst grade , then the average rank increment of the class taught by the teacher cannot be negative. Therefore, under the objective fact that there are differences in the student base, it is obviously unscientific to directly use the increment of the ranking; moreover, the existing teaching evaluation methods do not effectively combine the classroom effect, and simply use the level of grades to evaluate, and the evaluation factors It is relatively single, without combining factors such as emotional changes in the course teaching process, it is impossible to achieve a more comprehensive teaching evaluation.

Contents of the invention

In order to solve the shortcomings of the existing technology, the present disclosure provides a teaching evaluation method and system based on regression analysis, which eliminates the influence of the basic differences of teaching classes on incremental evaluation, and combines the psychological state evaluation results of each student in each class, A more accurate and comprehensive teaching evaluation has been realized.

In order to achieve the above purpose, the present disclosure adopts the following technical solutions:

The first aspect of the present disclosure provides a teaching evaluation method based on regression analysis.

A teaching evaluation method based on regression analysis, comprising the following steps:

Obtain the two adjacent grades of the students to be evaluated and the panoramic video data of all students in the time period to be evaluated;

According to the acquired panoramic video data of all students and the preset student emotional expression recognition model, the recognition result of each student's mental state is obtained;

According to the two adjacent performance rankings and regression analysis, the ranking increment model is obtained, and the ranking increment model is combined with the average basic ranking of all students to obtain the ranking increment after removing the basic influence;

The final evaluation result is obtained by weighting and summing the obtained ranking increment after removing the basic influence and the identification result of the student's mental state.

As an optional implementation, according to the average basic ranking and ranking increment model of all students gathered by a certain student, the influence of the basic ranking on the ranking increment is obtained, and the influence of the basic ranking on the ranking increment is subtracted from the original average ranking increment , get the ranking increment after removing the base influence.

As an optional implementation manner, a ranking increment model is constructed according to two adjacent performance ranking results, and parameters of the ranking increment model are obtained by using a regression analysis method.

As an optional implementation, receiving the student's panoramic video and performing distortion correction and grid mapping on the student's panoramic video to obtain the student target video of each student;

Import the student target video of each student into the student's emotional expression recognition model to identify the activation and joy of each student;

Calculate the activity level of each student based on the student target video of each student;

Receive the audio of each student collected by the preset audio collection device and calculate the negative emotions of each student according to the audio of each student;

Analyze the psychological state of each student according to the activation, joy, activity and negative emotions.

As an optional implementation, performing distortion correction and grid mapping on the student panoramic video to obtain the student target video of each student includes:

Using a video dynamic distortion correction algorithm to perform distortion correction on the student's panoramic video to obtain a standard student's panoramic video;

Carrying out frame processing to the standard student panoramic video to obtain multiple student panoramic images;

According to the ID background image of the student's seat position, network mapping is performed on each student's panoramic image to obtain the student's ID panoramic image;

Segment the target student image corresponding to each student ID in each student ID panorama image;

All target student images corresponding to the same student ID are stitched into a student target video in chronological order.

As an optional implementation, calculating the activity of each student according to the student target video of each student includes:

Obtain the time of capture in the student target video for each of said students;

Determine the get out of class end time in the shooting time according to the get out of class timetable;

Extracting the student video segment corresponding to the end of get out of class time from the student target video;

Detecting the duration of student appearance in the student video segment and calculating the total duration as the first degree of activity;

Calling the action type identification model to identify multiple action types in the student target video of each student, and determine the corresponding activity of each action type according to the mapping relationship between the preset action type and activity;

Calculate the average activity of the activities corresponding to all action types as the second activity;

Calculate the activity level of each student according to the first activity level and the second activity level.

As an optional implementation manner, said calculating the negative emotion of each student according to the audio of each student includes:

Calling a speech recognition algorithm to convert the audio into text and performing word segmentation on the text to obtain a plurality of words;

matching the plurality of words with a preset keyword library;

Using the keyword identical to any one of the plurality of words matched from the preset keyword library as the target keyword, and constructing a target keyword vector based on the matched target construction word;

extracting a plurality of acoustic features in the audio, and comparing each acoustic feature with a corresponding acoustic feature threshold;

Taking voiceprint features greater than the acoustic feature threshold as target acoustic features, and constructing an acoustic feature vector based on the target acoustic features;

The target keyword vector and the acoustic feature vector are used as the student's negative emotion vector.

As an optional implementation, the analysis of the mental state of each student according to the degree of activation, joy, activity and negative emotion includes:

Selecting the first student list corresponding to the activity lower than the preset activity threshold, and analyzing the mental state of the students in the first student list;

Mapping the activation and pleasure corresponding to the students in the first student list to the Arousal-Valence continuous emotional latitude model, determining the second student list corresponding to the preset latitude space, and analyzing the second student The mental state of the students on the list;

Judging whether the students in the second student list correspond to negative emotion vectors;

Determine a third student list corresponding to students with negative emotion vectors, and analyze the mental states of the students in the third student list.

The second aspect of the present disclosure provides a teaching evaluation method based on regression analysis, including the following steps:

Obtain the two adjacent grades of the students to be evaluated;

According to the two adjacent performance rankings and regression analysis, the ranking increment model is obtained, and the ranking increment model is combined with the average basic ranking of all students to obtain the ranking increment after removing the basic influence;

According to the ranking increment after removing the basic influence, the final evaluation result is obtained.

The third aspect of the present disclosure provides a teaching evaluation system based on regression analysis, including:

The data acquisition module is configured to: acquire the two adjacent grades of the students to be evaluated and the panoramic video data of all students in the time period to be evaluated;

The mental state recognition module is configured to: obtain the mental state recognition result of each student according to the obtained panoramic video data of all students and the preset student emotional expression recognition model;

The ranking incremental acquisition module is configured to: obtain the ranking incremental model according to the two adjacent grade rankings and regression analysis, combine the ranking incremental model and the average basic ranking of all students, and obtain the ranking increment after removing the basic influence;

The teaching evaluation module is configured to: perform a weighted summation of the obtained ranking increment after removing the basic influence and the identification result of the mental state of the students to obtain the final evaluation result.

The fourth aspect of the present disclosure provides a computer-readable storage medium on which a program is stored, and when the program is executed by a processor, the steps in the regression analysis-based teaching evaluation method described in the first aspect of the present disclosure are implemented.

The fifth aspect of the present disclosure provides an electronic device, including a memory, a processor, and a program stored on the memory and operable on the processor. When the processor executes the program, the implementation of the first aspect of the present disclosure is achieved. Steps in the teaching evaluation method based on regression analysis.

Compared with the prior art, the beneficial effects of the present disclosure are:

1. The method, system, medium or electronic device described in the present disclosure uses the method of regression analysis to determine the relationship between the ranking increment and the basic ranking, thereby eliminating the influence of the basic difference of the teaching class on the incremental evaluation.

2. The method, system, medium or electronic device described in this disclosure collects the panoramic video of the students in the classroom through the video collection equipment, avoiding the individual collection of video for each student, which has strong concealment and high degree of subjectivity, so the collection The authenticity of the video is high, and the accuracy of the mental state of the students obtained based on the collected video analysis is high.

3. The method, system, medium or electronic device described in this disclosure eliminates the influence of basic differences in teaching classes on incremental evaluation, and combines the psychological state evaluation results of each student in each class to achieve more accurate and comprehensive teaching Evaluation.

Advantages of additional aspects of the disclosure will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the disclosure.

Description of drawings

The accompanying drawings constituting a part of the present disclosure are used to provide a further understanding of the present disclosure, and the exemplary embodiments and descriptions of the present disclosure are used to explain the present disclosure, and do not constitute improper limitations to the present disclosure.

FIG. 1 is a schematic flow chart of a teaching evaluation method based on regression analysis provided by Embodiment 1 of the present disclosure.

FIG. 2 is a scatter diagram of basic nouns and ranking increments provided by Embodiment 1 of the present disclosure.

Detailed ways

The present disclosure will be further described below in conjunction with the accompanying drawings and embodiments.

It should be noted that the following detailed description is exemplary and intended to provide further explanation of the present disclosure. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure belongs.

It should be noted that the terminology used herein is only for describing specific embodiments, and is not intended to limit the exemplary embodiments according to the present disclosure. As used herein, unless the context clearly dictates otherwise, the singular is intended to include the plural, and it should also be understood that when the terms "comprising" and/or "comprising" are used in this specification, they mean There are features, steps, operations, means, components and/or combinations thereof.

In the case of no conflict, the embodiments in the present disclosure and the features in the embodiments can be combined with each other.

Example 1:

As shown in Figure 1, Embodiment 1 of the present disclosure provides a teaching evaluation method based on regression analysis, including the following steps:

Obtain the two adjacent grades of the students to be evaluated and the panoramic video data of all students in the time period to be evaluated;

According to the acquired panoramic video data of all students and the preset student emotional expression recognition model, the recognition result of each student's mental state is obtained;

According to the two adjacent performance rankings and regression analysis, the ranking increment model is obtained, and the ranking increment model is combined with the average basic ranking of all students to obtain the ranking increment after removing the basic influence;

The final evaluation result is obtained by weighting and summing the obtained ranking increment after removing the basic influence and the identification result of the student's mental state.

Specifically, the method to obtain the ranking increment after removing the basic influence is:

(1) Suppose there are n students in the grade, and the basic ranking of student i is recorded as x _i , i=1,2,...,n. After a period of teaching, the grade ranking of student i becomes x′ _i , i=1 ,2,...,n.

(2) The rank increment obtained by student i is y _i =xi _-x ' _i , i=1,2,...,n.

(3) The ranking increment is related to the basic ranking, and the functional relationship is set as y=f(x), where x represents the basic ranking, and y represents the ranking increment.

(4) The function y=f(x) contains unknown parameters, using regression analysis combined with the given data ( _xi , y _i ), i=1, 2,...,n to determine the unknown parameters in the function.

(5) Suppose there are m classes in the grade, the average basic ranking of all students in class k is X _k , k=1,2,...,m, after a period of teaching, the original average ranking of all students in class i The increment is Y _k , k=1, 2, . . . , m.

(6) Substitute the average basic ranking X _k of class k into the function y=f(x), and calculate the influence of the basic ranking on the ranking increment Y _k ^* = f(X _k ), k=1,2,..., m.

(7) Remove the influence Y _k ^* of the basic rank from the increment Y _k of the original ranking, record Z _k ＝ Y _k －Y _k ^* , k=1,2,...,m, then Z _k is to remove the basic influence of the students in the class Subsequent ranking increments.

(8) Use Z _k to evaluate the teaching quality of different teaching classes or teachers.

If the rankings mentioned in step (1) are tied, use "rank" to replace the concept of traditional rankings. After the results are ranked from high to low, the sequence number is called "rank". If there is a tie in the rankings, the average value of the sequence numbers is called "rank".

The following is an example to illustrate:

(1) A school has a total of 361 students in the whole year, record the basic ranking of student i as x _i , i=1,2,…,361, after a period of teaching, the grade ranking of student i becomes x′ _i , i =1,2,...,361.

(2) It is assumed that after a period of teaching, the rank increment obtained by student i is y _i = _xi -x' _i , i=1,2,...,361. For example, the basic rank of student 10 is x ₁₀ =15, and after a period of learning, the rank becomes x′ ₁₀ =12, then the increment of the student's rank is y ₁₀ =15-13=2.

(3) Due to the existence of a large number of parallel rankings, there are 27 different values of _xi after sorting, and the increments corresponding to the same _xi are averaged and recorded as y _i , ( _xi , y _i ), i=1, The scatter plot of 2,...,27 is shown in Figure 2.

(4) It can be seen from Figure 1 that the increment of students ranked in the first half is generally negative, and the increment of students ranked in the second half is generally positive. Except for some data, as the basic ranking increases, the increment also increases, and the relationship between the two is roughly linear.

(5) Suppose the functional relationship between the ranking increment and the basic ranking is y=f(x)=a+bx, where x represents the basic ranking, y represents the ranking increment, and a and b are unknown parameters.

(6) Using linear regression, a = -111.2, b = 0.62, so the functional relationship between the ranking increment and the basic ranking is y = -111.23+0.62x.

(7) There are 9 classes in this grade. Table 1 shows the average basic ranking X _k and original increment Y _k of each class.

Due to the large differences in the basis of each class, the original increment cannot be directly used for evaluation.

(8) Use the function Y _k ^* ＝－111.2+0.62X _k to calculate the influence Y _k ^* of the basic ranking on the ranking increment, and set Z _i ＝Y _i －Y _i ^* , which can be obtained after removing the influence of the basic ranking of the students in the class rank increment.

class 1 2 3 4 5 6 7 8 9 Y_k^* -22.1 -20.0 -16.4 5.2 8.9 18.8 9.2 6.4 3.9 Z_k -14.4 6.6 27.6 56.4 -32.7 23.8 -28.3 -52.2 22.8

(9) Using Z _k , where k=1, 2, .

(10) Among the 361 students, there are many students with the same grades, and the ranks are used to replace the rankings in the calculation. The calculation of the ranks is as follows: There are 7 students, and the grades are sorted from large to small: 138, 132, 132, 128, 128, 128, 120, the ranks of the 7 students are 1, 2.5, 2.5, 5, 5, 5, 7 respectively.

Mental states are identified by:

Obtain an open source emotional expression data set and a private emotional expression data set and carry out transfer learning training students' emotional expression recognition model based on the open source emotional expression data set and the private emotional expression data set;

Receive the student panoramic video collected by the preset video capture device and perform distortion correction and grid mapping on the student panoramic video to obtain the student target video of each student;

Import the student target video of each student into the student's emotional expression recognition model to identify the activation and joy of each student;

Calculate the activity level of each student based on the student target video of each student;

Receive the audio of each student collected by the preset audio collection device and calculate the negative emotions of each student according to the audio of each student;

Analyze the psychological state of each student according to the activation, joy, activity and negative emotions.

According to an optional embodiment of this embodiment, the transfer learning training of students' emotional expression recognition model based on the open source emotional expression data set and the private emotional expression data set includes:

Using the migration principal component analysis algorithm to perform dimension reduction processing on the open source emotional expression data set to obtain a target open source emotional expression data set and to perform dimension reduction processing on the private emotional expression data set to obtain a target private emotional expression data set;

Based on the target open source emotional expression dataset training CNN neural network to obtain the basic emotional expression recognition model;

According to the Arousal-Valence continuous emotion latitude model, the target private emotional expression data set is marked, and based on the marked target private emotional expression data set, the basic emotional expression recognition model is transferred and learned to obtain the student emotional expression recognition model.

According to an optional embodiment of this embodiment, performing distortion correction and grid mapping on the student panoramic video to obtain the student target video of each student includes:

Using a video dynamic distortion correction algorithm to perform distortion correction on the student's panoramic video to obtain a standard student's panoramic video;

Carrying out frame processing to the standard student panoramic video to obtain multiple student panoramic images;

According to the ID background image of the student's seat position, network mapping is performed on each student's panoramic image to obtain the student's ID panoramic image;

Segment the target student image corresponding to each ID in each student ID panorama image;

All target student images corresponding to the same ID are spliced into a student target video in chronological order.

According to an optional embodiment of this embodiment, the calculating the activity of each student according to the student target video of each student includes:

Obtain the time of capture in the student target video for each of said students;

Determine the get out of class end time in the shooting time according to the get out of class timetable;

Extracting the student video segment corresponding to the end of get out of class time from the student target video;

Detecting the duration of student appearance in the student video segment and calculating the total duration as the first degree of activity;

Calling the action type identification model to identify multiple action types in the student target video of each student, and determine the corresponding activity of each action type according to the mapping relationship between the preset action type and activity;

Calculate the average activity of the activities corresponding to all action types as the second activity;

Calculate the activity level of each student according to the first activity level and the second activity level.

According to an optional embodiment of this embodiment, the calculating the negative emotion of each student according to the audio of each student includes:

Calling a speech recognition algorithm to convert the audio into text and performing word segmentation on the text to obtain a plurality of words;

matching the plurality of words with a preset keyword library;

Using the keyword identical to any one of the plurality of words matched from the preset keyword library as the target keyword, and constructing a target keyword vector based on the matched target construction word;

extracting a plurality of acoustic features in the audio, and comparing each acoustic feature with a corresponding acoustic feature threshold;

Taking voiceprint features greater than the acoustic feature threshold as target acoustic features, and constructing an acoustic feature vector based on the target acoustic features;

The target keyword vector and the acoustic feature vector are used as the student's negative emotion vector.

According to an optional embodiment of this embodiment, the analyzing the mental state of each student according to the degree of activation, joy, activity and negative emotion includes:

Selecting the first student list corresponding to the activity lower than the preset activity threshold, and analyzing the mental state of the students in the first student list;

Mapping the activation and pleasure corresponding to the students in the first student list to the Arousal-Valence continuous emotional latitude model, determining the second student list corresponding to the preset latitude space, and analyzing the second student The mental state of the students on the list.

According to an optional embodiment of this embodiment, the method for analyzing students' mental states based on audio and video further includes:

Judging whether the students in the second student list correspond to negative emotion vectors;

Determine a third student list corresponding to students with negative emotion vectors, and analyze the mental states of the students in the third student list.

In this embodiment, a grade is assigned to the increment of the students' performance in the whole class, a grade is assigned to the mental state of the students, and finally the weighted summation is carried out according to the preset weights to obtain the final evaluation result of the whole class.

Example 2:

Embodiment 2 of the present disclosure provides a teaching evaluation method based on regression analysis, comprising the following steps:

Obtain the two adjacent grades of the students to be evaluated;

According to the two adjacent performance rankings and regression analysis, the ranking increment model is obtained, and the ranking increment model is combined with the average basic ranking of all students to obtain the ranking increment after removing the basic influence;

According to the ranking increment after removing the basic influence, the final evaluation result is obtained.

Specifically, the following steps are included:

Specifically, the method to obtain the ranking increment after removing the basic influence is:

(1) Suppose there are n students in the grade, and the basic ranking of student i is recorded as x _i , i=1,2,...,n. After a period of teaching, the grade ranking of student i becomes x′ _i , i=1 ,2,...,n.

(2) The rank increment obtained by student i is y _i =xi _-x ' _i , i=1,2,...,n.

(3) The ranking increment is related to the basic ranking, and the functional relationship is set as y=f(x), where x represents the basic ranking, and y represents the ranking increment.

(4) The function y=f(x) contains unknown parameters, using regression analysis combined with the given data ( _xi , y _i ), i=1, 2,...,m to determine the unknown parameters in the function.

(5) Suppose there are m classes in the grade, the average basic ranking of all students in class k is X _k , k=1,2,...,m, after a period of teaching, the original average ranking of all students in class i The increment is Y _k , k=1, 2, . . . , m.

(6) Substitute the average basic ranking X _k of class k into the function y=f(x), and calculate the influence of the basic ranking on the ranking increment Y _k ^* = f(X _k ), k=1,2,..., m.

(7) Remove the influence Y _k ^* of the basic rank from the increment Y _k of the original ranking, record Z _k ＝ Y _k －Y _k ^* , k=1,2,...,m, then Z _k is to remove the basic influence of the students in the class Subsequent ranking increments.

(8) Use Z _k to evaluate the teaching quality of different teaching classes or teachers.

Example 3:

Embodiment 3 of the present disclosure provides a teaching evaluation system based on regression analysis, including:

The data acquisition module is configured to: acquire the two adjacent grades of the students to be evaluated and the panoramic video data of all students in the time period to be evaluated;

The mental state recognition module is configured to: obtain the mental state recognition result of each student according to the obtained panoramic video data of all students and the preset student emotional expression recognition model;

The ranking incremental acquisition module is configured to: obtain the ranking incremental model according to the two adjacent grade rankings and regression analysis, combine the ranking incremental model and the average basic ranking of all students, and obtain the ranking increment after removing the basic influence;

The teaching evaluation module is configured to: perform a weighted summation of the obtained ranking increment after removing the basic influence and the identification result of the mental state of the students to obtain the final evaluation result.

The working method of the system is the same as the teaching evaluation method based on regression analysis provided in Embodiment 1, and will not be repeated here.

Example 4:

Embodiment 4 of the present disclosure provides a teaching evaluation system based on regression analysis, including:

The data acquisition module is configured to: acquire the two adjacent grades of the students to be evaluated;

The ranking incremental acquisition module is configured to: obtain the ranking incremental model according to the two adjacent grade rankings and regression analysis, combine the ranking incremental model and the average basic ranking of all students, and obtain the ranking increment after removing the basic influence;

The teaching evaluation module is configured to obtain the final evaluation result according to the ranking increment after removing the basic influence.

Example 5:

Embodiment 5 of the present disclosure provides a computer-readable storage medium, on which a program is stored. When the program is executed by a processor, the teaching evaluation method based on regression analysis as described in Embodiment 1 or Embodiment 2 of the present disclosure is implemented. A step of.

Embodiment 6:

Embodiment 6 of the present disclosure provides an electronic device, including a memory, a processor, and a program stored in the memory and operable on the processor. When the processor executes the program, the implementation as described in Embodiment 1 of the present disclosure is achieved. Steps in the teaching evaluation method based on regression analysis.

Those skilled in the art should understand that the embodiments of the present disclosure may be provided as methods, systems, or computer program products. Accordingly, the present disclosure may take the form of a hardware embodiment, a software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present disclosure may take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, optical storage, etc.) having computer-usable program code embodied therein.

The present disclosure is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the present disclosure. It should be understood that each procedure and/or block in the flowchart and/or block diagram, and a combination of procedures and/or blocks in the flowchart and/or block diagram can be realized by computer program instructions. These computer program instructions may be provided to a general purpose computer, special purpose computer, embedded processor, or processor of other programmable data processing equipment to produce a machine such that the instructions executed by the processor of the computer or other programmable data processing equipment produce a An apparatus for realizing the functions specified in one or more procedures of the flowchart and/or one or more blocks of the block diagram.

These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing apparatus to operate in a specific manner, such that the instructions stored in the computer-readable memory produce an article of manufacture comprising instruction means, the instructions The device realizes the function specified in one or more procedures of the flowchart and/or one or more blocks of the block diagram.

These computer program instructions can also be loaded onto a computer or other programmable data processing device, causing a series of operational steps to be performed on the computer or other programmable device to produce a computer-implemented process, thereby The instructions provide steps for implementing the functions specified in the flow chart or blocks of the flowchart and/or the block or blocks of the block diagrams.

Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be implemented through computer programs to instruct related hardware, and the programs can be stored in a computer-readable storage medium. During execution, it may include the processes of the embodiments of the above-mentioned methods. Wherein, the storage medium may be a magnetic disk, an optical disk, a read-only memory (Read-Only Memory, ROM) or a random access memory (Random Access Memory, RAM) and the like.

The above descriptions are only preferred embodiments of the present disclosure, and are not intended to limit the present disclosure. For those skilled in the art, the present disclosure may have various modifications and changes. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present disclosure shall be included within the protection scope of the present disclosure.

Claims (11)

1. a kind of teaching evaluation method based on regression analysis, it is characterized in that: comprise the following steps: Obtain the two adjacent grades of the students to be evaluated and the panoramic video data of all students in the time period to be evaluated; According to the acquired panoramic video data of all students and the preset student emotional expression recognition model, the recognition result of each student's mental state is obtained; According to the two adjacent grade rankings and regression analysis, the ranking increment model is obtained, and according to the average basic ranking and ranking increment model of all students gathered by a certain student, the influence of the basic ranking on the ranking increment is obtained; using the original average ranking increment Subtract the influence of the basic ranking on the ranking increment to obtain the ranking increment after removing the basic influence; The final evaluation result is obtained by weighting and summing the obtained ranking increment after removing the basic influence and the identification results of the students' mental state; The regression analysis method is used to determine the relationship between the ranking increment and the basic ranking, thereby eliminating the influence of the basic differences of the teaching classes on the incremental evaluation, and combining the psychological state evaluation results of each class of each student to achieve a more accurate and comprehensive evaluation. Teaching Evaluation; Among them, the method to obtain the ranking increment after removing the basic influence is: (1) Suppose there are n students in the grade, and the basic ranking of student i is recorded as x _i , i=1,2,...,n. After a period of teaching, the grade ranking of student i becomes x′ _i , i=1 ,2,...,n; (2) The ranking increment obtained by student i is y _i =xi _-x ′ _i , i=1,2,...,n; (3) ranking increment is related to basic ranking, assuming that the functional relationship is y=f(x), wherein x represents the basic ranking, and y represents the ranking increment; (4) The function y=f(x) contains unknown parameters, using regression analysis combined with the given data ( _xi , y _i ), i=1,2,...,n to determine the unknown parameters in the function; (5) Suppose there are m classes in the grade, the average basic rank of all students in class k is X _k , k=1,2,...,m, after a period of teaching, the original average rank obtained by all students in class k The increment is Y _k , k=1,2,...,m; (6) Substitute the average basic ranking X _k of class k into the function y=f(x), and calculate the influence of the basic ranking on the ranking increment Y _k ^* = f(X _k ), k=1,2,..., m; (7) Remove the influence Y _k ^* of the basic rank from the increment Y _k of the original ranking, record Z _k ＝ Y _k －Y _k ^* , k=1,2,...,m, then Z _k is to remove the basic influence of the students in the class Post-rank increments; (8) Use Z _k to evaluate the teaching quality of different teaching classes or teachers. 2. the teaching evaluation method based on regression analysis as claimed in claim 1, is characterized in that: According to the ranking results of two adjacent grades, a ranking incremental model is constructed, and the parameters of the ranking incremental model are obtained by using the method of regression analysis. 3. the teaching evaluation method based on regression analysis as claimed in claim 1, is characterized in that: Receive the student's panoramic video and perform distortion correction and grid mapping on the student's panoramic video to obtain the student's target video of each student; Import the student target video of each student into the student's emotional expression recognition model to identify the activation and joy of each student; Calculate the activity level of each student based on the student target video of each student; Receive the audio of each student collected by the preset audio collection device and calculate the negative emotions of each student according to the audio of each student; Analyze the psychological state of each student according to the activation, joy, activity and negative emotions. 4. the teaching evaluation method based on regression analysis as claimed in claim 3, is characterized in that: Distortion correction and grid mapping are carried out to the student's panoramic video to obtain the student target video of each student including: Using a video dynamic distortion correction algorithm to perform distortion correction on the student's panoramic video to obtain a standard student's panoramic video; Carrying out frame processing to the standard student panoramic video to obtain multiple student panoramic images; According to the ID background image of the student's seat position, network mapping is performed on each student's panoramic image to obtain the student's ID panoramic image; Segment the target student image corresponding to each student ID in each student ID panorama image; All target student images corresponding to the same student ID are stitched into a student target video in chronological order. 5. the teaching evaluation method based on regression analysis as claimed in claim 3, is characterized in that: Calculating each student's activity based on said student target video for each student includes: Obtain the time of capture in the student target video for each of said students; Determine the get out of class end time in the shooting time according to the get out of class timetable; Extracting the student video segment corresponding to the end of get out of class time from the student target video; Detecting the duration of student appearance in the student video segment and calculating the total duration as the first degree of activity; Calling the action type identification model to identify multiple action types in the student target video of each student, and determine the corresponding activity of each action type according to the mapping relationship between the preset action type and activity; Calculate the average activity of the activities corresponding to all action types as the second activity; Calculate the activity level of each student according to the first activity level and the second activity level. 6. the teaching evaluation method based on regression analysis as claimed in claim 3, is characterized in that: The calculating the negative emotion of each student according to the audio of each student includes: Calling a speech recognition algorithm to convert the audio into text and performing word segmentation on the text to obtain a plurality of words; matching the plurality of words with a preset keyword library; Using the keyword identical to any one of the plurality of words matched from the preset keyword library as the target keyword, and constructing a target keyword vector based on the matched target construction word; extracting a plurality of acoustic features in the audio, and comparing each acoustic feature with a corresponding acoustic feature threshold; Taking voiceprint features greater than the acoustic feature threshold as target acoustic features, and constructing an acoustic feature vector based on the target acoustic features; The target keyword vector and the acoustic feature vector are used as the student's negative emotion vector. 7. the teaching evaluation method based on regression analysis as claimed in claim 3, is characterized in that: The analysis of each student's psychological state according to the degree of activation, joy, activity and negative emotion includes: Selecting the first student list corresponding to the activity lower than the preset activity threshold, and analyzing the mental state of the students in the first student list; Map the activation and pleasure corresponding to the students in the first student list to the Arousal-Valence continuous emotional latitude model, determine the second student list corresponding to the preset latitude space, and analyze the second student list the mental state of the students; Judging whether the students in the second student list correspond to negative emotion vectors; Determine a third student list corresponding to students with negative emotion vectors, and analyze the mental states of the students in the third student list. 8. A teaching evaluation method based on regression analysis, characterized in that: comprising the following steps: Obtain the two adjacent grades of the students to be evaluated; According to the two adjacent grade rankings and regression analysis, the ranking increment model is obtained, and according to the average basic ranking and ranking increment model of all students gathered by a certain student, the influence of the basic ranking on the ranking increment is obtained; using the original average ranking increment Subtract the influence of the basic ranking on the ranking increment to obtain the ranking increment after removing the basic influence; According to the ranking increment after removing the basic influence, the final evaluation result is obtained; The regression analysis method is used to determine the relationship between the ranking increment and the basic ranking, thereby eliminating the influence of the basic differences of the teaching classes on the incremental evaluation, and combining the psychological state evaluation results of each class of each student to achieve a more accurate and comprehensive evaluation. Teaching Evaluation; Among them, the method to obtain the ranking increment after removing the basic influence is: (1) Suppose there are n students in the grade, and the basic ranking of student i is recorded as x _i , i=1,2,...,n. After a period of teaching, the grade ranking of student i becomes x′ _i , i=1 ,2,...,n; (2) The ranking increment obtained by student i is y _i =xi _-x ′ _i , i=1,2,...,n; (3) ranking increment is related to basic ranking, assuming that the functional relationship is y=f(x), wherein x represents the basic ranking, and y represents the ranking increment; (4) The function y=f(x) contains unknown parameters, using regression analysis combined with the given data ( _xi , y _i ), i=1,2,...,n to determine the unknown parameters in the function; (5) Suppose there are m classes in the grade, the average basic rank of all students in class k is X _k , k=1,2,...,m, after a period of teaching, the original average rank obtained by all students in class k The increment is Y _k , k=1,2,...,m; (6) Substitute the average basic ranking X _k of class k into the function y=f(x), and calculate the influence of the basic ranking on the ranking increment Y _k ^* = f(X _k ), k=1,2,..., m; (7) Remove the influence Y _k ^* of the basic rank from the increment Y _k of the original ranking, record Z _k ＝ Y _k －Y _k ^* , k=1,2,...,m, then Z _k is to remove the basic influence of the students in the class Post-rank increments; (8) Use Z _k to evaluate the teaching quality of different teaching classes or teachers. 9. A teaching evaluation system based on regression analysis, characterized in that: comprising: The data acquisition module is configured to: acquire the two adjacent grades of the students to be evaluated and the panoramic video data of all students in the time period to be evaluated; The mental state recognition module is configured to: obtain the mental state recognition result of each student according to the obtained panoramic video data of all students and the preset student emotional expression recognition model; The ranking incremental acquisition module is configured to: obtain the ranking incremental model according to the two adjacent grade rankings and regression analysis, and obtain the basic ranking vs. Quantitative influence; use the original average ranking increment to subtract the influence of the basic ranking on the ranking increment, and obtain the ranking increment after removing the basic influence; The teaching evaluation module is configured to: perform a weighted summation of the obtained ranking increment after removing the basic influence and the identification result of the mental state of the student to obtain the final evaluation result; The regression analysis method is used to determine the relationship between the ranking increment and the basic ranking, thereby eliminating the influence of the basic differences of teaching classes on the incremental evaluation, and combining the psychological state evaluation results of each class of each student to achieve a more accurate and comprehensive evaluation. Teaching Evaluation; Among them, the method to obtain the ranking increment after removing the basic influence is: (1) Suppose there are n students in the grade, and the basic ranking of student i is recorded as x _i , i=1,2,...,n. After a period of teaching, the grade ranking of student i becomes x′ _i , i=1 ,2,...,n; (2) The ranking increment obtained by student i is y _i =xi _-x ′ _i , i=1,2,...,n; (3) ranking increment is related to basic ranking, assuming that the functional relationship is y=f(x), wherein x represents the basic ranking, and y represents the ranking increment; (4) The function y=f(x) contains unknown parameters, using regression analysis combined with the given data ( _xi , y _i ), i=1,2,...,n to determine the unknown parameters in the function; (5) Suppose there are m classes in the grade, the average basic rank of all students in class k is X _k , k=1,2,...,m, after a period of teaching, the original average rank obtained by all students in class k The increment is Y _k , k=1,2,...,m; (6) Substitute the average basic ranking X _k of class k into the function y=f(x), and calculate the influence of the basic ranking on the ranking increment Y _k ^* = f(X _k ), k=1,2,..., m; (7) Remove the influence Y _k ^* of the basic rank from the increment Y _k of the original ranking, record Z _k ＝ Y _k －Y _k ^* , k=1,2,...,m, then Z _k is to remove the basic influence of the students in the class Post-rank increments; (8) Use Z _k to evaluate the teaching quality of different teaching classes or teachers. 10. A computer-readable storage medium on which a program is stored, characterized in that, when the program is executed by a processor, the steps in the teaching evaluation method based on regression analysis as described in any one of claims 1-8 are realized . 11. An electronic device comprising a memory, a processor, and a program stored in the memory and operable on the processor, wherein the processor implements any one of claims 1-8 when executing the program Steps in the described teaching evaluation method based on regression analysis.

Images