Classroom teaching effect evaluation method and system

Abstract

The invention discloses a classroom teaching effect evaluation method. The method comprises the following steps: acquiring classroom data of a teacher; acquiring classroom data of students; processing the teacher classroom data and the student classroom data to obtain a data set of each dimension; calculating the weight of each dimension by using an analytic hierarchy process, and calculating the score of the classroom effect according to the weight of the attribute; outputting classroom effect evaluation and suggestions according to the calculated score of the classroom effect; the invention also discloses a classroom teaching effect evaluation system. According to the invention, manual supervision evaluation is replaced by automatic data acquisition and processing, so that the labor cost is reduced; according to the method, the matching degree and the interaction quality of the teaching content of the teacher and the behavior data of the students are combined to construct the multi-dimensional quantitative model, so that the objectivity of evaluation is improved; according to the invention, the evaluation result can be output in real time, teaching strategies can be adjusted timely, and dynamic feedback can be realized.

Description

Technical Field

[0001] The present invention belongs to the technical field of classroom teaching effect evaluation, and specifically relates to a method and a system for classroom teaching effect evaluation. Background Art

[0002] The quality of the class determines the quality of teaching, which directly affects students' learning of the course. The evaluation of a good class requires not only the assessment of the teacher's behavior and teaching, but also a comprehensive evaluation based on the students' performance in the class.

[0003] At present, the evaluation of classroom effectiveness mainly relies on the evaluation and scoring of teaching supervisors based on their experience after random checks. This solution has the following disadvantages:

[0004] 1. High labor costs. Currently, the evaluation of teachers' classroom effects mainly relies on teaching supervisors to evaluate after attending classes, which invisibly increases the workload and number of teaching supervisors;

[0005] 2. The professional knowledge requirements for teaching supervisors are high, and the scores of classroom effectiveness given by assessors with different years of work experience vary greatly.

[0006] Therefore, the evaluation of classroom effectiveness also varies greatly, so a method of classroom teaching effectiveness evaluation is needed to accurately and efficiently evaluate the teaching classroom effectiveness scientifically. Summary of the invention

[0007] The purpose of the present invention is to provide a method for evaluating classroom teaching effects to solve the problems raised in the above background technology. The present invention provides a method for evaluating classroom teaching effects, which has the characteristics of being able to accurately and efficiently scientifically evaluate the teaching classroom effects.

[0008] Another object of the present invention is to provide a system for evaluating classroom teaching effectiveness.

[0009] To achieve the above object, the present invention provides the following technical solution: a method for evaluating classroom teaching effect, comprising the following steps:

[0010] S1. Obtain teacher classroom data;

[0011] S2, obtain students’ classroom data;

[0012] S3, processing the teacher's classroom data and the student's classroom data to obtain data sets of various dimensions;

[0013] S4. Use the hierarchical analysis method to calculate the weight of each dimension, and calculate the score of classroom effect according to the weight of the attribute;

[0014] S5. Output classroom effect evaluation and suggestions based on the calculated classroom effect scores.

[0015] Further in the present invention, in step S1, the teacher's classroom data includes but is not limited to one or more of courseware information, subject content, class start time, get out of class end time, teacher's lecture content, number of interactions with students, number of interactive students and classroom time occupied by interactions.

[0016] Further in the present invention, in step S2, the student classroom data includes but is not limited to one or more of the number of late students, the number of early leaving students, the number of absent students, the number of students asking for leave, the number of students taking notes and the number of non-disciplined students.

[0017] In the present invention, in step S3, each dimension data set of the teacher's classroom data includes whether the classroom content is consistent, whether the teacher is late or leaves early, and the interaction data with the students, wherein:

[0018] The method to determine whether the classroom content is consistent is to obtain the voice dataset V of the teacher's lecture in the classroom. t , and converted into a text dataset W t , by calling Baidu short text similarity to calculate the text dataset W t With the subject content dataset D theme The similarity value S text , output data results:

[0019]

[0020] The method to obtain whether the teacher is late or leaves early is: Get the teacher's class time t T-start and the teacher's end time t T-end , according to the class time T start and end of get out of class time T end For comparison, if t T-start <=T start If t T-end >=T end If the teacher leaves get out of class normally, it will be recorded as the teacher leaving class early.

[0021] Output data results:

[0022]

[0023] The method for obtaining the interaction data with students is to retrieve the classroom monitoring data and count the number of teacher-student interactions in the classroom by calling Baidu EasyDL image statistics. student-pv , Number of interactive students student-uv And the duration of class time for interaction student-time , the statistical data is normalized, where the number of teacher-student interactions in the classroom is A student-pvAnd the number of interactive students A student-uv Perform discretization processing.

[0024] In the present invention, further, in step S3, each dimensional data set of the student classroom data includes the late rate, early leave rate, truancy rate, the proportion of students taking notes, and the proportion of non-disciplined students. Among them, the number of students who should attend the class S is obtained from the school educational administration system database all And the number of students on leave on that day S r , to obtain the number of students who should attend the class S y = S all - S r ;

[0025] The methods for obtaining the late rate, early leave rate, and truancy rate are as follows:

[0026] Retrieve the classroom monitoring data, and count the number of late students S late And the number of early leave students S out , and the number of truant students S noin , and calculate to obtain:

[0027] Late rate

[0028] Early leave rate

[0029] Truancy rate

[0030] Perform discretization processing on the late rate, early leave rate, and truancy rate, and output the results;

[0031] The methods for obtaining the proportion of students taking notes and the proportion of non-disciplined students are as follows: Retrieve the classroom monitoring data, and count the number of students taking notes S a-num And the number of non-disciplined students S b-num , and calculate to obtain:

[0032] Proportion of students taking notes

[0033] Proportion of non-disciplined students

[0034] Perform discretization processing on the proportion of students taking notes and the proportion of non-disciplined students, and output the results.

[0035] In the present invention, further, in step S4, the weight values of each dimension are calculated using the analytic hierarchy process, and the method for calculating the classroom effect score according to the weight values of the attributes includes the following steps:

[0036] S41. Establish a hierarchical structure model;

[0037] S42. Construct a judgment matrix;

[0038] S43. Conduct a hierarchical consistency test;

[0039] S44. Perform hierarchical ranking.

[0040] In the present invention, further, in step S41, establishing a hierarchical structure model includes the following steps:

[0041] S411. Establish an objective layer;

[0042] S412. Establish a criterion layer, where the criterion layer includes but is not limited to whether the teacher is late for class, whether the teacher leaves early after class, the similarity of the course content explained, the number of interactions with students, the number of students interacting, the duration of interactions in class, the early leaving rate of students, the late arrival rate of students, the absenteeism rate of students, the proportion of students taking notes, and the proportion of students not observing discipline;

[0043] S413. Establish a solution layer, and the solution layer includes teacher classroom data and student classroom data.

[0044] In the present invention, further, in step S43, the hierarchical consistency test includes the following steps:

[0045] S431. Calculate the in-row product value;

[0046] S432. Take the nth power of the in-row product value;

[0047] S433. Calculate the weight value WI;

[0048] S434. Calculate the sub-vector AWI;

[0049] S435. Calculate the consistency index CR;

[0050] In the present invention, further, in step S5, the method for outputting the classroom effect evaluation is as follows: According to the constructed classroom evaluation indicators and the corresponding weight values, retrieve the classroom data from the database and perform discretization processing to obtain a set D of the values of each attribute. After modeling with the AHP algorithm, obtain a set X of the weight values of each attribute. The evaluation value of the classroom teaching effect of each class is the product of the set D of the values of each attribute and the set X of the weight values of each attribute; perform normalization processing on the calculated result, and normalize the calculated value to the interval (0, 10), that is:

[0051]

[0052] Output the evaluation result for the normalized value.

[0053] Further in the present invention, a system for evaluating classroom teaching effects includes a terminal, a camera, and a cloud server. Among them, the terminal is used for teachers to upload courseware information and theme content; the camera is used to record audio and video data in the classroom; and the cloud server is used to process and store the data.

[0054] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0055] 1. The present invention replaces manual supervision and evaluation through automated data collection and processing, reducing labor costs.

[0056] 2. The present invention combines the matching degree of teachers' teaching content, interaction quality, and students' behavior data to construct a multi-dimensional quantification model, improving the objectivity of evaluation.

[0057] 3. The present invention can output evaluation results in real time, facilitating timely adjustment of teaching strategies and enabling dynamic feedback.

[0058] 4. The present invention uses the analytic hierarchy process to clarify the weights of each index, avoiding the untrustworthy problem of the "black box" algorithm. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 It is a schematic flow chart of the method of the present invention;

[0060] Figure 2 It is a schematic diagram of the hierarchical model built by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0061] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0062] Embodiment 1

[0063] Please refer to Figure 1 - Figure 2 , the present invention provides the following technical solutions: A method for evaluating classroom teaching effects includes the following steps:

[0064] S1. Obtain teachers' classroom data, where teachers' classroom data includes courseware information, theme content, start time of class, end time of class, teachers' lecture content, number of interactions with students, number of interactive students, and duration of interaction in the class. Among them, the courseware information and theme content are uploaded by teachers to the cloud server through the terminal;

[0065] S2. Obtain students' in-class data, which includes the number of students being late, leaving early, absent, on leave, taking notes, and non-compliant students. Non-compliant students include, but are not limited to, those playing with mobile phones, whispering to each other, or sleeping;

[0066] S3. Process the teachers' in-class data and the students' in-class data to obtain datasets for each dimension;

[0067] S4. Use the analytic hierarchy process to calculate the weights for each dimension, and calculate the score of the class effect based on the weights of the attributes;

[0068] S5. Output the class effect evaluation and suggestions according to the calculated score of the class effect.

[0069] By adopting the above technical solutions, the present invention replaces manual supervision and evaluation with automated data collection and processing, reducing the labor cost; the present invention constructs a multi-dimensional quantification model by combining the matching degree of the teacher's teaching content, the interaction quality, and the students' behavior data, improving the objectivity of the evaluation; the present invention can output the evaluation results in real time, facilitating the timely adjustment of teaching strategies and enabling dynamic feedback; the present invention uses the analytic hierarchy process to clarify the weights of each index, avoiding the untrustworthy problem of the "black box" algorithm.

[0070] Specifically, in step S3, the datasets for each dimension of the teachers' in-class data include whether the class content is consistent, whether the teacher is late or leaves early, and the interaction data with the students. Among them,

[0071] The method for obtaining whether the class content is consistent is: obtain the voice dataset V of the teacher's lecture in the class t , and convert it into a text dataset W t , and calculate the similarity value S between the text dataset W t and the theme content dataset D theme by calling the Baidu short text similarity, and the Baidu text similarity API link is: text ;

[0072] https: / / ai.baidu.com / tech / nlp_basic / simnet?track=cp:ainsem|pf:pc|pp: chanpin-NLP|pu:NLP-xiangguanjishu|ci:|kw:10007382 ;

[0073] Output data result:

[0074]

[0075] The method for obtaining whether the teacher is late or leaves early is: obtain the teacher's class start time t T-start and the teacher's class end time t T-end , and compare them with the class start time T start and the class end time T end . If t T-start <= T startIt is recorded as normal class, otherwise it is recorded as the teacher being late for class; if t T-end >= T end It is recorded as the teacher finishing class normally, otherwise it is recorded as the teacher leaving early;

[0076] Output data result:

[0077]

[0078] The method for obtaining the data of interaction with students is as follows: retrieve the classroom monitoring data, and count the number of interactions A between teachers and students in the classroom by calling Baidu EasyDL images student-pv , the number of interacting students A student-uv and the duration of interaction in the classroom A student-time , and perform normalization processing on the statistical data. Among them, perform discretization processing on the number of interactions A between teachers and students in the classroom student-pv and the number of interacting students A student-uv ; the Baidu EasyDL image recognition API link is:

[0079] https: / / ai.baidu.com / tech / easydl / cv?track=cp:Ainsem|pf:pc|pp:easyDL|pu:easyDL-tuxiang-tuozhan|ci:|kw:10525413.

[0080] The number of students A interacted by the teacher in the classroom student-uv and the number of interactions A student-pv are discretized as follows:

[0081]

[0082] Specifically, in step S3, the datasets of each dimension of the students' classroom data include the late rate, early leave rate, truancy rate, the proportion of students taking notes, and the proportion of students not observing discipline. Among them, obtain the number of students S who should attend the class from the school educational administration system database all and the number of students S who are absent due to leave on that day r , and obtain the number of students S who should attend the class y = S all - S r ;

[0083] The methods for obtaining the late rate, early leave rate, and truancy rate are as follows:

[0084] Retrieve the classroom monitoring data, and count the number of late students S late and the number of early leave students S out , the number of truant students S noin , and perform calculations to obtain:

[0085] Tardy rate

[0086] Early leave rate

[0087] Absence rate

[0088] Discretize the tardy rate, early leave rate, and absence rate, and output the results;

[0089] The tardy rate, early leave rate, and absence rate are discretized as follows:

[0090]

[0091] The method for obtaining the proportion of students taking notes and the proportion of students not observing discipline is: retrieve the classroom monitoring data, and count the number of students taking notes S a-num and the number of students not observing discipline S b-num in the classroom through Baidu EasyDL Image Statistics, and calculate to obtain:

[0092] Proportion of students taking notes

[0093] Proportion of students not observing discipline

[0094] Discretize the proportion of students taking notes and the proportion of students not observing discipline, and output the results.

[0095] The proportion of students taking notes and the proportion of students not observing discipline are discretized as follows:

[0096]

[0097] Embodiment 2

[0098] Specifically, in step S4, the method for calculating the weight values of each dimension using the Analytic Hierarchy Process and calculating the classroom effect score based on the weight values of the attributes includes the following steps:

[0099] S41. Establish a hierarchical structure model;

[0100] S411. Establish an objective layer. The main objective of this model is to evaluate the classroom effect of teachers;

[0101] S412. Establish a criterion layer. Among them, the criterion layer includes but is not limited to whether the teacher is late for class, whether the teacher leaves early after class, the similarity of the course content explained, the number of interactions with students, the number of students involved in the interaction, the duration of the interaction in the classroom, the early leave rate of students, the tardy rate of students, the absence rate of students, the proportion of students taking notes, and the proportion of students not observing discipline;

[0102] S413. Establish the solution layer, which includes teacher classroom data and student classroom data;

[0103] S42. Construct a judgment matrix;

[0104] When determining the weights of factors at each layer, if only qualitative results are obtained, they are often not easily accepted by others. Therefore, the consistent matrix method is adopted, that is, instead of comparing all factors together, they are compared pairwise. A relative scale is used in the comparison to minimize the difficulty of comparing factors with different natures as much as possible and improve the accuracy. For example, for a certain criterion, the solutions under it are compared pairwise, and their importance levels are rated. The proportional scale table of the importance between factors is as follows:

[0105] Element i is compared with element j Quantification value Equally important 1 Slightly important 3 Strongly important 5 Very important 7 Extremely important 9 Intermediate value between two adjacent judgments 2，4，6，8

[0106] Construct a teacher classroom data matrix as follows:

[0107]

[0108] Construct a student classroom data matrix as follows:

[0109]

[0110]

[0111] S43. Hierarchical consistency test;

[0112] The main purpose of the consistency test is to check whether there is a large difference between the judgment matrix we constructed and the consistent matrix. Usually, if the consistency index CR < 0.1, it is considered that the consistency test passes, otherwise it fails. The specific calculation process is as follows:

[0113] S431. Calculate the in-row product value;

[0114] The in-row product mainly refers to the result obtained by multiplying the data in a single row. Taking the data in the teacher classroom data matrix as an example, the in-row product value of the classroom content consistency index = 1 * 3 * 3 * 2 * 2 * 2 = 72;

[0115] S432. Take the nth power of the in-row product value;

[0116] Take the nth root of the calculated in-row product value to obtain an nth root value. The calculation logic is:

[0117] The value after taking the nth root = power(i, 1 / n);

[0118] S433. Calculate the weight value WI;

[0119] For the weight values of each attribute, use the proportion after taking the nth root, that is, the weight value of each attribute where i is the label of the attribute, and n i is the value after taking the nth root of attribute i;

[0120] Taking the construction of a student classroom data matrix as an example, the weight WI of each attribute i is as follows:

[0121] (1) Calculate the sum ∑n of the values after taking the nth root of each attribute i ∑n i ,

[0122] ∑n i = 1 + 1 + 1 + 0.561 + 1.782 = 5.343;

[0123] (2) Calculate the weight WI value of each attribute i For example, the weight of the late arrival rate = 1 / 5.343 = 0.187.

[0124] S434. Calculate the sub - vector AWI;

[0125] The sub - vector AWI = D * WI T , where D represents the constructed data set. Taking the construction of a student classroom data matrix as an example, D is the constructed student classroom data matrix, and WI is the set of weight values of each attribute. Then the sub - vector AWI of the late arrival rate is calculated as follows:

[0126] AWI = 1 * 0.178 + 1 * 0.178 + 1 * 0.178 + 2 * 1.050 + 0.5 * 0.333 = 0.93.

[0127] S435. Calculate the consistency index CR.

[0128] Perform a consistency test on the matrix, where where RI is the average random consistency index, and the specific values are as follows:

[0129] Matrix order 1 2 3 4 5 6 7 8 9 10 RI 0 0 0.58 0.90 1.12 1.24 1.32 1.41 1.45 1.49

[0130] The consistency index value CI is calculated as CI = (n - 1) * (Avg n - n), where n is the number of dimensions of the attribute, and Avg n is the mean value of

[0131] Taking the construction of a student classroom data matrix as an example, calculate the CR value of the consistency index, and the calculation process is as follows:

[0132] (1) Calculate the Avg n value,

[0133] (2) Calculate the consistency index value CI,

[0134] CI = (n - 1) * (Avg n - n) = (5 - 1) * (5.027 - 5) = 0.107;

[0135] (3) Calculate the consistency ratio CR,

[0136] Since CR = 0.095 < 0.1, the matrix passes the consistency test.

[0137] The hierarchical consistency test table of the teacher's classroom data matrix is as follows:

[0138]

[0139] The hierarchical consistency test table of the student's classroom data matrix is as follows:

[0140]

[0141] S44. Conduct hierarchical sorting.

[0142] Since this method constructs a two - layer model, the first layer mainly has two indicators, namely teacher's classroom data and student's classroom data respectively. Since this patent mainly evaluates the teaching effect of the teacher's classroom, there is an inclination in the attribute weights of the teacher's classroom data. The specific attribute construction and attribute weights are as follows:

[0143]

[0144] Example 3

[0145] Specifically, in step S5, the method for outputting the classroom effect evaluation is: according to the constructed classroom evaluation indicators and the corresponding weight values, retrieve the classroom data from the database and perform discretization processing to obtain the set D of the values of each attribute. After modeling with the AHP algorithm, the set X of the weight values of each attribute is obtained. The evaluation value of the classroom teaching effect of each class is the product of the set D of the values of each attribute and the set X of the weight values of each attribute; perform normalization processing on the calculated result, and normalize the calculated value to the interval (0, 10), that is:

[0146]

[0147] Output the evaluation result for the normalized value as follows:

[0148] Value after normalization Rating [8–10] Excellent [7–8) Good [6–7) Pass [4–6) To be improved [0–4) Poor

[0149] Example 4

[0150] The present embodiment provides the following technical solutions: Specifically, a system for evaluating the effect of classroom teaching includes a terminal, a camera, and a cloud server. Among them, the terminal is used for teachers to upload courseware information and theme content; the camera is used to record the audio-visual data in the classroom; the cloud server is used to process and store the data.

[0151] In summary, the present invention replaces manual supervision and evaluation through the automated collection and processing of data, reducing the labor cost; the present invention combines the matching degree of the teacher's teaching content, the interaction quality, and the behavior data of students to construct a multi-dimensional quantification model, improving the objectivity of evaluation; the present invention can output the evaluation results in real time, facilitating the timely adjustment of teaching strategies and enabling dynamic feedback; the present invention uses the analytic hierarchy process to clarify the weights of each index, avoiding the untrustworthy problem of the "black box" algorithm.

[0152] Although the embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for evaluating the effect of classroom teaching, characterized in that: It includes the following steps: S1. Obtain teachers' classroom data; S2. Obtain students' classroom data; S3. Process the teachers' classroom data and the students' classroom data to obtain datasets for each dimension; S4. Use the Analytic Hierarchy Process to calculate the weights for each dimension, and calculate the score of the classroom effect based on the weights of the attributes; S5. Output the classroom effect evaluation and suggestions according to the calculated score of the classroom effect.

2. The method for evaluating the effect of classroom teaching according to claim 1, wherein: In step S1, the teachers' classroom data includes, but is not limited to, one or more of courseware information, theme content, start time of class, end time of class, teachers' lecture content, number of interactions with students, number of interacting students, and the duration of interactions in the class.

3. The method for evaluating the effect of classroom teaching according to claim 1, wherein: In step S2, the students' classroom data includes, but is not limited to, one or more of the number of students being late, the number of students leaving early, the number of students absent from class, the number of students on leave, the number of students taking notes, and the number of students not observing discipline.

4. A method for evaluating the effect of classroom teaching according to claim 1, characterized in that: In step S3, the datasets for each dimension of the teachers' classroom data include whether the classroom content is consistent, whether the teacher is late or leaves early, and the interaction data with students. Among them, The method for obtaining whether the classroom content is consistent is as follows: obtain the voice dataset V of the teacher's lecture in the classroom t and convert it into a text dataset W t By calling the short text similarity of Baidu, calculate the similarity value S of the text dataset W t and the theme content dataset D theme Output the data result: text ​ The method for obtaining whether a teacher is late or leaves early is as follows: obtain the class start time t of the teacher T-start and the class end time t of the teacher T-end , and compare with the class start time T start and the class end time T end . If t T-start <= T start , it is recorded as normal class attendance, otherwise it is recorded that the teacher is late for class; if t T-end >= T end , it is recorded that the teacher ends the class normally, otherwise it is recorded that the teacher leaves early; Output data results: The method for obtaining student interaction data is as follows: retrieve classroom monitoring data, and count the number of interactions A between teachers and students in the classroom by calling Baidu EasyDL images student-pv , the number of interacting students A student-uv , and the duration of interaction A in the classroom student-time . Normalize the statistically obtained data. Among them, discretize the number of interactions A student-pv between teachers and students in the classroom and the number of interacting students A student-uv .

5. The method for evaluating the classroom teaching effect according to claim 1, wherein: In the step S3, the data sets of each dimension of the students' classroom data include the late rate, the early leave rate, the absenteeism rate, the proportion of students taking notes, and the proportion of students not observing discipline. Among them, the number of students S who should attend the class is obtained from the school educational administration system database all and the number of students S who ask for leave on the same day r , and the number of students S who should attend the class is obtained y =S all -S r ; The methods for obtaining the late rate, early leave rate, and absenteeism rate are as follows: Retrieve classroom monitoring data, and count the number of late students S by calling Baidu EasyDL images late and the number of early-leaving students S out , the number of absent students S noin , and calculate to obtain: Late arrival rate Early departure rate Absence rate Discretize the late rate, early leave rate, and absenteeism rate, and output the results; The method for obtaining the proportion of students taking notes and the proportion of students not observing discipline is as follows: retrieve the classroom monitoring data, and call Baidu EasyDL Image to count the number of students S a-num taking notes and the number of students S b-num not observing discipline in the classroom, and calculate to obtain: Percentage of students taking notes Percentage of students not complying with discipline Discretize the proportion of students taking notes and the proportion of students not observing discipline, and output the results.

6. The method for evaluating the effect of classroom teaching according to claim 1, characterized in that: In step S4, the method of using the Analytic Hierarchy Process to calculate the weights for each dimension and calculating the classroom effect score based on the weights of the attributes includes the following steps: S41. Establish a hierarchical structure model; S42. Construct a judgment matrix; S43. Conduct a hierarchical consistency test; S44. Conduct a hierarchical ranking.

7. A method for evaluating the effect of classroom teaching according to claim 6, characterized in that: In step S41, establishing a hierarchical structure model includes the following steps: S411. Establish an objective layer; S412. Establish a criterion layer, where the criterion layer includes, but is not limited to, whether the teacher is late for class, whether the teacher leaves early after class, similarity of the course content explained, number of interactions with students, number of interacting students, duration of interactions in the class, early leave rate of students, late rate of students, absenteeism rate of students, proportion of students taking notes, and proportion of students not observing discipline; S413. Establish a solution layer, and the solution layer includes teachers' classroom data and students' classroom data.

8. A method for evaluating the effect of classroom teaching according to claim 6, characterized in that: In step S43, the hierarchical consistency test includes the following steps: S431. Calculate the product value within each row; S432. Take the nth power of the product value within each row; S433. Calculate the weight value WI; S434. Calculate the sub-vector AWI; S435. Calculate the consistency index CR.

9. A method for evaluating the effect of classroom teaching according to claim 1, characterized in that: In step S5, the method for outputting the classroom effect evaluation is as follows: According to the constructed classroom evaluation indicators and the corresponding weight values, retrieve the classroom data from the database and conduct discretization processing to obtain a set D of values for each attribute. After modeling with the AHP algorithm, obtain a set X of weight values for each attribute. The evaluation value of the classroom teaching effect for each class is the product of the set D of values for each attribute and the set X of weight values for each attribute; Normalize the calculated results, and normalize the calculated values to the interval (0, 10), that is: Output the evaluation results for the normalized values.

10. A system for evaluating the effect of classroom teaching according to any one of claims 1-9, characterized in that: It includes a terminal, a camera, and a cloud server. Among them, the terminal is used for teachers to upload courseware information and theme content; the camera is used to record the audio-visual data in the classroom; the cloud server is used to process and store the data.

Images