Intelligent education management system and method based on artificial intelligence

Abstract

The invention discloses an intelligent education management system and method based on artificial intelligence, and relates to the technical field of education management, and the method comprises the steps: collecting classroom speech, distinguishing different student speech through employing a voiceprint recognition technology, building a classroom interaction round recognition algorithm, and analyzing the classroom student participation degree Cy; obtaining the answer correctness, the number of times of modification, the answering starting time and the submitting time of each question according to the homework answering record, calculating the total answering time of students and the answering time of single questions, constructing a student homework database, analyzing the difficulty degree # imgabs0 # of the single questions, marking the difficult questions, and storing the difficult questions in the homework database; counting the difficult question proportion, the average answer total duration and the average answer total accuracy of the after-class homework, and analyzing an after-class homework evaluation index Pj; and obtaining a classroom student participation degree Cy and an after-class homework evaluation index Pj, and introducing a time decay effect and a cooperative gain factor to analyze a classroom evaluation index Pz. And teaching evaluation is changed from static result guiding to dynamic process optimization.

Description

Technical Field

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

[0002] Under the traditional education management model, educational institutions generally face pain points such as insufficient quantification of student engagement, highly subjective homework evaluation, and a lack of data support for the allocation of teaching resources. Specifically, these pain points include: Limited student engagement assessment: Traditional classrooms rely on teachers' subjective observations to record student performance, making it difficult to accurately capture implicit data such as the frequency of classroom interactions and the quality of speech. This leads to a one-sided assessment of student engagement and fails to provide a basis for personalized instruction. The inefficiency and accuracy of homework evaluations conflict: Teachers spend a significant amount of time grading homework, but manual evaluation is susceptible to subjective factors and struggles with in-depth analysis of massive amounts of homework data, making it impossible to promptly identify student weaknesses. Data silos exist for teaching decisions: Data such as student classroom performance, homework completion, and test scores are stored in disparate locations, lacking an integrated analysis platform. This makes it difficult for teachers to fully grasp the student situation and results in delayed adjustments to teaching strategies. Mismatch between resource allocation and demand: Traditional teaching models rely on experience to develop teaching plans and lack dynamic optimization mechanisms for parameters such as course difficulty and homework load, which can easily lead to excessive student burdens and poor learning outcomes.

[0003] While existing technology solutions attempt to incorporate information technology to optimize education management, they still suffer from significant flaws: some systems only implement basic data recording functions, lacking intelligent analysis technologies and unable to unlock the deeper value of data. The few education management systems that have attempted to apply AI often focus on a single scenario and lack closed-loop management of the entire student learning process. Existing technologies fail to fully integrate multimodal data, resulting in a single assessment dimension and making it difficult to construct an accurate profile of student capabilities.

[0004] In the Chinese invention application with application publication number CN119918806A, a smart education management system and method based on AI and cloud computing are disclosed, including first obtaining classroom monitoring images on the terminal side and detecting student behaviors, such as raising hands, standing, walking and other common classroom behaviors. When student behaviors are detected, they are uploaded to the cloud computing platform, otherwise the next frame of classroom monitoring images is received. On the cloud computing platform, image analysis technology based on AI and cloud computing is used to detect student behavior regions of interest and perform multi-scale image extraction processing on the classroom monitoring images to obtain a set of local image coding features of student behaviors, and based on the student behavior feature aggregation representation of the local image coding feature set of student behaviors, the classroom behavior recognition results are intelligently obtained.

[0005] In the above invention application, the classroom behavior recognition results are obtained intelligently based on the aggregated representation of student behavior features based on the local image coding feature set of student behavior. However, only classroom behavior is analyzed, and the mastery of knowledge outside of class is not analyzed. Classroom behavior may reflect learning attitude or short-term attention, but it cannot be directly equated with knowledge internalization. For example, students may complete classroom tasks through mechanical repetition without truly understanding the logical relationship between concepts. Introverted students may make up for the lack of classroom interaction through self-study after class, while extroverted students may ignore independent review because they rely on immediate feedback in class. Single-dimensional analysis cannot capture this difference.

[0006] To this end, the present invention provides an intelligent education management system and method based on artificial intelligence. Summary of the Invention

[0007] (1) Technical problems solved In response to the shortcomings of the existing technology, the present invention provides an intelligent education management system and method based on artificial intelligence. The present invention obtains classroom student participation Cy and after-class homework evaluation index Pj, introduces time decay effect and synergistic gain factor to analyze the classroom evaluation index Pz, so as to shift the teaching evaluation from "static result orientation" to "dynamic process optimization", significantly improve the timeliness and accuracy of educational intervention, and ultimately promote the dual improvement of teaching quality and student participation, provide data support for tiered homework arrangement and dynamic adjustment of teaching progress, thereby comprehensively improving teaching quality and student participation, thereby solving the technical problems recorded in the background technology.

[0008] (2) Technical solution To achieve the above objectives, the present invention is implemented through the following technical solutions: a smart education management method based on artificial intelligence, comprising the following steps: Collect classroom speeches, use voiceprint recognition technology to distinguish different students' speeches, build a classroom interaction turn recognition algorithm, and analyze the classroom student participation Cy; Based on the homework answer records, the correctness of the answer, the number of revisions, the start time of the answer, and the submission time of each question are obtained. The total time of the student answering and the time taken to answer a single question are calculated. The student homework database is constructed and the difficulty of a single question is analyzed. , mark difficult questions, and count the proportion of difficult questions, average total answering time, average total correct answer rate of this homework, and analyze the homework evaluation index Pj; Obtain classroom student participation Cy and homework evaluation index Pj , the time decay effect and synergistic gain factor are introduced to analyze the classroom evaluation index Pz.

[0009] Furthermore, we build a classroom interaction round recognition algorithm, initialize the round list chains = [], the current round current_chain = [], traverse the speech records, and if the interval between the current speech and the previous speech is less than the threshold and the speaker is different, then add it to the current round; if the interval is greater than the threshold or the speaker is repeated, then save the current round and reset it. Finally, we get all the rounds and their lengths, count the proportion of rounds of each length in the classroom, and mark the longest discussion chain length of each class.

[0010] Furthermore, 10-30 seconds of voice samples are collected from each student, the voiceprint features are extracted and stored in the database, classroom voice is collected in real time, the features are extracted and compared with the voiceprint database, the student ID with the highest similarity is output, and the number of student IDs speaking in the classroom is counted and recorded as the number of speaking students.

[0011] Furthermore, we can obtain the number of students who speak in class S and the length of all rounds. and the corresponding proportion , analyze the classroom student participation Cy, specifically:

[0012] in, i Indicates the sequence number of the class round, i =1, 2, ..., a , a Indicates the total number of class rounds.

[0013] Furthermore, we collect statistics on the accuracy rate, average number of revisions and average answering time of each question based on the student homework database, and analyze the difficulty of each question. :

[0014] in, is the correct rate of a single question, is the average number of revisions per question, The average time it takes to answer a single question.

[0015] Difficulty of a single question If the difficulty threshold is exceeded, the question will be marked as a difficult question, where the difficulty threshold is μ+1.5σ, μ is the mean of all single question difficulty data, and σ is the variance of all single question difficulty data.

[0016] Furthermore, we calculated the percentage of difficult questions, the average total time spent answering questions, and the average total correct rate of answers in the homework, and analyzed the homework evaluation index. Pj :

[0017] in, is the corrected accuracy rate, Rc=R×(1−0.2D), which penalizes the high accuracy rate under high difficulty, R is the original average total accuracy rate of answering questions, D is the proportion of difficult questions, T is the average total time of answering questions, As the upper limit of reasonable duration, we take the historical average total duration + 2 standard deviations, and Dw is the weighted proportion of difficult questions. , strengthen the influence of the number of questions, H is the number of difficult questions in the homework, is the variance of the correct rate of the student group, reflecting the degree of differentiation of grades, is the dynamic weight, , initial value =0.4, =0.3, =0.3, which can be adjusted dynamically according to the characteristics of the subject. The initial threshold for the proportion of difficult questions is 0.3. If the value exceeds this threshold, the score will be significantly reduced.

[0018] Furthermore, we can obtain the classroom student participation Cy and the homework evaluation index Pj , introducing the time decay effect and synergistic gain factor to analyze the classroom evaluation index Pz:

[0019] in, is a synergistic balance item, is the nonlinear sensitivity, , Sensitive attenuation, , Enhanced sensitivity, 1+ is the discrete correction term, is the discrete penalty coefficient, , is the covariance of historical Cy / Pj, is the time decay term, Δt is the time difference between class and homework, is the memory decay rate, =0.1 means daily attenuation of 10%, which needs to be calibrated according to the subject. is the gain correction factor.

[0020] Furthermore, the classroom evaluation index Pz is obtained, and different adjustment and early warning processing strategies are selected for the current classroom according to the classroom evaluation index Pz, specifically: Obtain the classroom evaluation index Pz, and select different adjustment and early warning processing strategies for the current classroom based on the classroom evaluation index Pz, specifically: when When teaching is currently in a state of sustained excellence, challenging tasks are introduced and student-led inquiry is initiated.

[0021] when At this time, the current teaching is in an optimized iterative state, the classroom interaction weight is adjusted, and the homework is designed in layers.

[0022] when When the teaching is in a state of reconstruction intervention, it is necessary to reorganize the teaching content and reset the homework structure.

[0023] when At this time, the current teaching is in an emergency rescue state, the teaching method is switched, and daily micro-assessments are implemented.

[0024] An artificial intelligence-based smart education management system, comprising: The classroom participation analysis module collects classroom speeches, uses voiceprint recognition technology to distinguish different students' speeches, builds a classroom interaction turn recognition algorithm, and analyzes the classroom student participation Cy; The after-school homework analysis module obtains the correctness of the answer, number of revisions, answering start time, and submission time of each question based on the homework answering record, calculates the total time of students answering questions and the time taken to answer each question, builds a student homework database, and analyzes the difficulty of each question , mark difficult questions, and count the proportion of difficult questions, average total answering time, average total correct answer rate of this homework, and analyze the homework evaluation index Pj; Classroom evaluation adjustment module, obtains classroom student participation Cy and after-class homework evaluation index Pj , the time decay effect and synergistic gain factor are introduced to analyze the classroom evaluation index Pz.

[0025] (3) Beneficial effects The present invention provides an artificial intelligence-based smart education management system and method, which has the following beneficial effects: 1. Collect classroom speeches, use voiceprint recognition technology to distinguish different students' speeches, build a classroom interaction turn recognition algorithm, analyze classroom student participation Cy, and quantitatively evaluate the depth of student participation. This can not only help grasp classroom dynamics in real time, but also generate participation reports, and can adjust teaching strategies in a targeted manner to stimulate the enthusiasm of more students. This helps to build a more active and inclusive classroom atmosphere and significantly improve students' overall participation.

[0026] 2. Obtain the correctness of the answer, number of revisions, start time, and submission time of each question based on the homework answer record, calculate the total time students spend answering questions and the time it takes to answer each question, build a student homework database, and analyze the difficulty of each question , mark difficult questions, and count the proportion of difficult questions, average total answering time, average total correct answer rate of this homework, and analyze the homework evaluation index Pj, can accurately identify students' knowledge weaknesses and common error patterns, helping to adjust teaching strategies in a timely manner, such as focusing on high-frequency errors or breaking down and reconstructing difficult questions to improve classroom efficiency. Furthermore, it can quantitatively assess the quality of homework, avoiding overly difficult or overly easy questions and ensuring that the difficulty of the homework matches the teaching objectives.

[0027] 3. Obtain the classroom student participation Cy and after-class homework evaluation index Pj , introducing the time decay effect and synergistic gain factor to analyze the classroom evaluation index Pz, so that the teaching evaluation will shift from "static result orientation" to "dynamic process optimization", significantly improving the timeliness and accuracy of educational intervention, and ultimately promoting the dual improvement of teaching quality and student participation, providing data support for stratified homework assignments and dynamic adjustment of teaching progress, thereby comprehensively improving teaching quality and student participation. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 This is a flow chart of an artificial intelligence-based smart education management method of the present invention; Figure 2 This is a structural diagram of an artificial intelligence-based smart education management system of the present invention. DETAILED DESCRIPTION

[0029] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0030] See also Figure 1 The present invention provides an artificial intelligence-based smart education management method, comprising the following steps: Step 1: Collect classroom speeches, use voiceprint recognition technology to distinguish different students' speeches, build a classroom interaction turn recognition algorithm, and analyze the classroom student participation Cy.

[0031] The step 1 includes the following: Step 101: Collect classroom speeches using a microphone array, use voiceprint recognition technology to distinguish between different students, build a classroom interaction turn identification algorithm, initialize the turn list chains = [], and the current turn current_chain = [], traverse the speech records, and if the interval between the current and previous speeches is less than the threshold and the speaker is different, add them to the current round. If the interval is greater than the threshold or the speaker is the same, save the current round and reset it. Finally, obtain all rounds and their lengths, calculate the proportion of rounds of each length in the class, and mark the length of the longest discussion chain in each class.

[0032] The specific percentage of rounds of different lengths is as follows: In a certain lesson, length 1 accounts for 42%, 2-3 accounts for 35%, and > 3 accounts for 23%. The specific length of a round is: "A→B→C→D→E", so the length is 4.

[0033] Step 102: Collect 10-30 seconds of voice samples from each student, extract voiceprint features and store them in the database. Collect classroom voice in real time, extract features and compare them with the voiceprint database, output the student ID with the highest similarity, count the number of student IDs that speak in class, and record them as the number of speaking students.

[0034] Step 103: Get the number of students who spoke in class S and the length of all rounds and the corresponding proportion , analyze the classroom student participation Cy, specifically:

[0035] in, i Indicates the sequence number of the class round, i =1, 2, ..., a , a Indicates the total number of class rounds.

[0036] When using, combine the contents in steps 101 to 103: Collecting classroom speeches, using voiceprint recognition technology to distinguish different students' speeches, building a classroom interaction turn recognition algorithm, analyzing classroom student participation Cy, and quantitatively evaluating the depth of student participation can not only help grasp classroom dynamics in real time, but also generate participation reports, and can adjust teaching strategies in a targeted manner to stimulate the enthusiasm of more students. This helps to build a more active and inclusive classroom atmosphere and significantly improve students' overall participation.

[0037] Step 2: Obtain the correctness of the answer, number of revisions, start time, and submission time of each question based on the homework answer record, calculate the total time the student takes to answer the questions and the time it takes to answer each question, build a student homework database, and analyze the difficulty of each question , mark difficult questions, and count the proportion of difficult questions, average total answering time, average total correct answer rate of this homework, and analyze the homework evaluation index Pj.

[0038] The second step includes the following: Step 201: Obtain the correctness of the answer, number of revisions, answering start time, and submission time of each question based on the homework answering record, calculate the total time the student takes to answer the questions and the time it takes to answer a single question, and build a student homework database.

[0039] Step 202: Analyze the difficulty of each question by collecting statistics on the correct rate, average number of revisions and average answering time of each question based on the student homework database. :

[0040] in, is the correct rate of a single question, is the average number of revisions per question, The average time it takes to answer a single question.

[0041] Difficulty of a single question If the difficulty threshold is exceeded, the question will be marked as a difficult question, where the difficulty threshold is μ+1.5σ, μ is the mean of all single question difficulty data, and σ is the variance of all single question difficulty data.

[0042] Step 203: Count the percentage of difficult questions, average total answering time, and average total correct answer rate of this homework, and analyze the homework evaluation index. Pj :

[0043] in, is the corrected accuracy rate, Rc=R×(1−0.2D), which penalizes the high accuracy rate under high difficulty, R is the original average total accuracy rate of answering questions, D is the proportion of difficult questions, T is the average total time of answering questions, As the upper limit of reasonable duration, we take the historical average total duration + 2 standard deviations, and Dw is the weighted proportion of difficult questions. , strengthen the influence of the number of questions, H is the number of difficult questions in the homework, is the variance of the correct rate of the student group, reflecting the degree of differentiation of grades, is the dynamic weight, , initial value =0.4, =0.3, =0.3, which can be adjusted dynamically according to the characteristics of the subject. The initial threshold for the proportion of difficult questions is 0.3. If the value exceeds this threshold, the score will be significantly reduced.

[0044] When using, combine the contents in steps 201 to 203: Based on the homework answer records, the correctness of the answer, the number of revisions, the start time of the answer, and the submission time of each question are obtained. The total time of the student answering and the time taken to answer a single question are calculated. The student homework database is constructed and the difficulty of a single question is analyzed. , mark difficult questions, and count the proportion of difficult questions, average total answering time, average total correct answer rate of this homework, and analyze the homework evaluation index Pj, can accurately identify students' knowledge weaknesses and common error patterns, helping to adjust teaching strategies in a timely manner, such as focusing on high-frequency errors or breaking down and reconstructing difficult questions to improve classroom efficiency. Furthermore, it can quantitatively assess the quality of homework, avoiding overly difficult or overly easy questions and ensuring that the difficulty of the homework matches the teaching objectives.

[0045] Step 3: Obtain the classroom student participation Cy and homework evaluation index Pj , the time decay effect and synergistic gain factor are introduced to analyze the classroom evaluation index Pz.

[0046] The step three includes the following: Step 301: Obtain the class student participation Cy and the after-class homework evaluation index Pj , introducing the time decay effect and synergistic gain factor to analyze the classroom evaluation index Pz:

[0047] in, is a synergistic balance item, is the nonlinear sensitivity, , Sensitive attenuation, , Enhanced sensitivity, 1+ is the discrete correction term, is the discrete penalty coefficient, , is the covariance of historical Cy / Pj, is the time decay term, Δt is the time difference between class and homework, is the memory decay rate, =0.1 means daily attenuation of 10%, which needs to be calibrated according to the subject. is the gain correction factor.

[0048] Step 302: Obtain the classroom evaluation index Pz, and select different adjustment and early warning processing strategies for the current classroom based on the classroom evaluation index Pz, specifically: when When teaching is currently in a state of sustained excellence, challenging tasks are introduced and student-led inquiry is initiated.

[0049] when At this time, the current teaching is in an optimized iterative state, the classroom interaction weight is adjusted, and the homework is designed in layers.

[0050] when When the teaching is in a state of reconstruction intervention, it is necessary to reorganize the teaching content and reset the homework structure.

[0051] when At this time, the current teaching is in an emergency rescue state, the teaching method is switched, and daily micro-assessments are implemented.

[0052] When using, combine the contents in steps 301 and 302: Obtain classroom student participation Cy and homework evaluation index Pj , introducing the time decay effect and synergistic gain factor to analyze the classroom evaluation index Pz, so that the teaching evaluation will shift from "static result orientation" to "dynamic process optimization", significantly improving the timeliness and accuracy of educational intervention, and ultimately promoting the dual improvement of teaching quality and student participation, providing data support for stratified homework assignments and dynamic adjustment of teaching progress, thereby comprehensively improving teaching quality and student participation.

[0053] See also Figure 2 The present invention provides an artificial intelligence-based smart education management system, comprising: The classroom participation analysis module collects classroom speeches, uses voiceprint recognition technology to distinguish different students' speeches, builds a classroom interaction turn recognition algorithm, and analyzes the classroom student participation Cy; The after-school homework analysis module obtains the correctness of the answer, number of revisions, answering start time, and submission time of each question based on the homework answering record, calculates the total time of students answering questions and the time taken to answer each question, builds a student homework database, and analyzes the difficulty of each question , mark difficult questions, and count the proportion of difficult questions, average total answering time, average total correct answer rate of this homework, and analyze the homework evaluation index Pj; Classroom evaluation adjustment module, obtains classroom student participation Cy and after-class homework evaluation index Pj , the time decay effect and synergistic gain factor are introduced to analyze the classroom evaluation index Pz.

[0054] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. Those skilled in the art will appreciate that the units and algorithm steps of the various examples 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.

[0055] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0056] The above is only a specific implementation method of the present application, but the scope of protection of the present application is not limited thereto. Any technician familiar with this technical field can easily think of changes or replacements within the technical scope disclosed in this application, which should be covered by the scope of protection of the present application.

Claims

1. A smart education management method based on artificial intelligence, characterized by: The steps include: Collect classroom speeches, use voiceprint recognition technology to distinguish different students' speeches, build a classroom interaction turn recognition algorithm, and analyze the classroom student participation Cy; Based on the homework answer records, the correctness of the answer, the number of revisions, the start time of the answer, and the submission time of each question are obtained. The total time of the student answering and the time taken to answer a single question are calculated. The student homework database is constructed and the difficulty of a single question is analyzed. , mark difficult questions, and count the proportion of difficult questions, average total answering time, average total correct answer rate of this homework, and analyze the homework evaluation index Pj; Obtain classroom student participation Cy and homework evaluation index Pj , the time decay effect and synergistic gain factor are introduced to analyze the classroom evaluation index Pz.

2. The method of intelligent education management based on artificial intelligence according to claim 1, characterized in that: Build a classroom interaction turn recognition algorithm, initialize the turn list chains = [], the current turn current_chain = [], traverse the speech records, if the interval between the current speech and the previous speech is less than the threshold and the speaker is different, then add it to the current round; if the interval is greater than the threshold or the speaker is repeated, save the current round and reset it, and finally obtain all rounds and their lengths, count the proportion of rounds of each length in the classroom, and mark the longest discussion chain length of each class.

3. The method for intelligent education management based on artificial intelligence according to claim 1, characterized in that: Collect 10-30 seconds of voice samples from each student, extract voiceprint features and store them in the database, collect classroom voice in real time, extract features and compare them with the voiceprint database, output the student ID with the highest similarity, count the number of student IDs speaking in class, and record it as the number of speaking students.

4. The method for intelligent education management based on artificial intelligence according to claim 1, characterized in that: Get the number of students speaking in class S and the length of all rounds and the corresponding proportion , analyze the classroom student participation Cy. Among them, i Indicates the sequence number of the class round, i =1, 2, ..., a , a Indicates the total number of class rounds.

5. The method of intelligent education management based on artificial intelligence according to claim 1, characterized in that: Difficulty of a single question If the difficulty threshold is exceeded, the question will be marked as a difficult question, where the difficulty threshold is μ+1.5σ, μ is the mean of all single question difficulty data, and σ is the variance of all single question difficulty data.

6. The method for intelligent education management based on artificial intelligence according to claim 1, characterized in that: Count the percentage of difficult questions, average total answering time, average total correct answer rate of this homework, and analyze the homework evaluation index Pj, is the corrected accuracy rate, Rc=R×(1−0.2D), which penalizes the high accuracy rate under high difficulty, R is the original average total accuracy rate of answering questions, D is the proportion of difficult questions, T is the average total time of answering questions, As the upper limit of reasonable duration, we take the historical average total duration + 2 standard deviations, and Dw is the weighted proportion of difficult questions. , H is the number of difficult questions in the homework, is the variance of the correct rate of the student group, is the dynamic weight, , initial value =0.4, =0.3, =0.3, is the threshold for the proportion of difficult questions, and the initial threshold for the proportion of difficult questions is 0.

3.

7. The method of intelligent education management based on artificial intelligence according to claim 1, characterized in that: Obtain the classroom evaluation index Pz, and select different adjustment and early warning processing strategies for the current classroom based on the classroom evaluation index Pz, specifically: when When the current teaching is in a state of excellence maintenance; when When , the current teaching is in the optimization and iteration state; when When , the current teaching is in the reconstruction intervention state; when At this time, the current teaching is in an emergency rescue state.

8. An artificial intelligence-based smart education management system, used to implement the method according to any one of claims 1 to 7, characterized in that: include: The classroom participation analysis module collects classroom speeches, uses voiceprint recognition technology to distinguish different students' speeches, builds a classroom interaction turn recognition algorithm, and analyzes the classroom student participation Cy; The after-school homework analysis module obtains the correctness of the answer, number of revisions, answering start time, and submission time of each question based on the homework answering record, calculates the total time of students answering questions and the time taken to answer each question, builds a student homework database, and analyzes the difficulty of each question , mark difficult questions, and count the proportion of difficult questions, average total answering time, average total correct answer rate of this homework, and analyze the homework evaluation index Pj; Classroom evaluation adjustment module, obtains classroom student participation Cy and after-class homework evaluation index Pj , the time decay effect and synergistic gain factor are introduced to analyze the classroom evaluation index Pz.

Images