Interactive teaching management system and method for smart classroom

Abstract

The present application discloses an interactive teaching management system and method for a smart classroom, which relates to the field of teaching management technology and solves the technical problem in the prior art that teaching activities are carried out through intelligent multimedia without considering the actual classroom situation of students, and it is difficult to set students' homework according to their actual situation; the application includes: a central processing module, a data acquisition module, a display control terminal and a database; the central processing module: generates acquisition instructions based on teacher's voice information; acquires student images according to the acquisition instructions, and inputs the student images into a concentration model to obtain the student's time period concentration; generates reminder signals and homework based on the time period concentration and the concentration threshold; facilitates teachers to adjust teaching methods and teaching progress according to students' class status; can set corresponding types of questions according to students' listening status of different knowledge points during class to better suit students' learning situation, so that students can have learning resources suitable for themselves.

Description

Technical Field

[0001] This application belongs to the field of teaching management technology, and relates to teaching management technology, specifically an interactive teaching management system and method for smart classrooms. Background Art

[0002] With the help of advanced information technology, smart classrooms have realized the digitization, intelligence and internetization of the teaching process, allowing teachers to monitor students' learning status in real time, accurately understand their classroom status, and then adjust the teaching content in a timely manner; smart classrooms provide a variety of interactive methods to make classroom content more vivid and interesting, which is conducive to improving students' concentration.

[0003] Existing technology (patent publication number CN109754653B) discloses a personalized teaching system that uses facial, gesture, or emotion recognition results to output corresponding personalized content through an output interface. Teachers then use this personalized content to provide personalized instruction in the classroom. However, this technology fails to consider students' actual performance in class, making it difficult to tailor homework assignments to their specific circumstances. This can potentially lead to student polarization. Therefore, the present invention provides an interactive teaching management system and method for smart classrooms. Summary of the Invention

[0004] This application aims to solve at least one of the technical problems existing in the prior art; to this end, this application proposes an interactive teaching management system and method for a smart classroom, which is used to solve the technical problem that the prior art uses intelligent multimedia to conduct teaching activities without considering the students' actual classroom situation, and it is difficult to set students' homework according to their actual situation. This application solves the above problems by obtaining students' real-time concentration during class to remind teachers to adjust their teaching methods and teaching progress, and at the same time generates teaching homework according to the concentration of time periods.

[0005] To achieve the above objectives, the first aspect of this application provides an interactive teaching management system and method for a smart classroom, including:

[0006] Central processing module, and the data acquisition module, display control terminal and database connected thereto;

[0007] Data acquisition module: obtain teacher voice information, video information and student images through data acquisition equipment;

[0008] Central processing module: Acquires teacher voice information and generates acquisition instructions based on the teacher voice information; acquires student images according to the acquisition instructions, inputs the student images into the concentration model to obtain the student's concentration level for each period; generates reminder signals and after-class homework based on the concentration level for each period and the concentration threshold; the concentration model is trained through an artificial intelligence model;

[0009] The display control terminal includes a teacher terminal and a student terminal; the teacher terminal is used to display the student status during class, etc.; the student terminal is used to display the teacher's teaching content, etc.

[0010] Database: used to store historical data, teaching content, student status, etc.

[0011] This application generates collection instructions based on the teacher's voice information, collects student images according to the collection instructions, obtains the student's concentration at that time period through the student image, and generates reminder signals in time according to the concentration of the time period, so that the teacher can adjust the teaching method and teaching progress according to the student's class status; at the same time, teaching homework is generated according to the concentration of the time period, and the corresponding type of questions can be set according to the student's listening status of different knowledge points during class to better suit the student's learning situation, so that students can have learning resources that suit them.

[0012] Preferably, the generating of the collection instruction based on the teacher's voice information includes:

[0013] Obtain teacher's voice information in real time and convert it into text information;

[0014] Detect whether key words appear in text information;

[0015] If yes, then generate important words of the text according to the text information and generate collection instructions at the same time;

[0016] If not, the teacher's voice information is continuously obtained, converted into text information, and the text information is tested; among them, the key vocabulary includes chapter keywords, knowledge point keywords and specific keywords.

[0017] During the actual class, when the teacher talks about some key knowledge points, the students' concentration in class may change. If student images are collected throughout the class, the subsequent image processing will require a huge amount of computation. In the actual class, it is only necessary to focus on some special times for image collection and processing. This application collects and analyzes the teacher's voice information, and generates collection instructions based on the key vocabulary in the teacher's voice information. The collection instructions are instructions for real-time collection of student images. This enables this application to flexibly collect student images according to the teacher's class situation, thereby reducing the real-time computational complexity.

[0018] Preferably, the central processing module is further used to expand teaching content through the cloud platform, including:

[0019] Extract chapter keywords and knowledge point keywords from important words in the text;

[0020] Search for corresponding chapter content and knowledge points in the cloud platform;

[0021] The chapter content and knowledge point content are sent to the teacher's terminal for screening to obtain teaching content; the teaching content is sent to the student terminal;

[0022] Screen chapter content and knowledge point content to obtain teaching content, including current affairs mode and classic mode;

[0023] The current affairs mode sorts chapter contents and / or knowledge points in chronological order, and selects the chapter contents and / or knowledge points with the highest rankings;

[0024] The classic mode is to sort the chapter content and / or knowledge point content in descending order according to the number of views and / or content ratings, and filter out the chapter content and / or knowledge point content with the highest ranking.

[0025] This application obtains chapter content and knowledge point content through the cloud platform, and sends the chapter content and knowledge point content to student terminals as extended content for teaching; it facilitates students to obtain content related to chapter keywords and knowledge point keywords, and thus facilitates students to understand and learn relevant chapters and knowledge points.

[0026] Preferably, the concentration model is obtained through artificial intelligence model training, including:

[0027] Obtain several student status images and their corresponding concentration scores; integrate the student status images and their corresponding concentration scores into training data and test data; use the training data to train an artificial intelligence model; use the test data to test the trained artificial intelligence model; obtain a concentration model whose input is the student image and output is the time period concentration score;

[0028] Among them, the time period concentration score is recorded as time period concentration; the artificial intelligence model includes a BP neural network model or an RBF neural network model.

[0029] Preferably, generating a reminder signal based on the student's concentration during a period of time includes:

[0030] Obtain the concentration of the time period corresponding to the student images collected under the same collection instruction for several students, as well as the concentration threshold corresponding to each student;

[0031] Obtain all students whose concentration is less than the corresponding concentration threshold and mark them as target personnel; mark the concentration of the target personnel in the time period as SZi, and mark the corresponding concentration threshold as YZi; where i is the target personnel number, and i=1, 2, 3, ..., n; n is the total number of target personnel in this collection instruction;

[0032] The signal value XH corresponding to this acquisition instruction is calculated by the formula XH=∑[(YZi-SZi) / YZi];

[0033] When the signal value XH is greater than the signal value threshold and is not equal to the total number of target personnel, an interactive instruction is generated;

[0034] When the signal value XH is greater than the signal value threshold and equal to the total number of target personnel, a rest instruction is generated;

[0035] When the signal value XH is less than the signal value threshold, marking information is generated. The reminder signal includes interactive instructions, rest instructions and marking information. The interactive instructions include small games and jokes related to the course, etc., which are used for teachers to interact with students to improve students' concentration.

[0036] This application generates reminder signals based on the concentration level of each time period, reminding teachers to make reasonable teaching arrangements, thereby improving students' learning efficiency. At the same time, it makes it easier for teachers to pay attention to a small number of students who are not focused in class, so as to solve the problem of poor learning effect of certain knowledge points caused by a small number of students being absent in class.

[0037] Preferably, generating homework based on the student's concentration level and concentration threshold during a period of time includes:

[0038] Obtain all key vocabulary in this lesson; extract chapter keywords and knowledge point keywords from the key vocabulary, and divide this lesson into several teaching periods based on the chapter keywords and knowledge point keywords; obtain the corresponding time period concentration and corresponding concentration threshold for each student in each teaching period;

[0039] Determine whether the concentration level during the time period is greater than the concentration threshold; if yes, mark the student's homework for the time period as a test homework; if no, mark the student's homework for the time period as a teaching homework; wherein the test homework includes homework items; the teaching homework includes teaching videos and guided homework, and the guided homework is a homework question that explains the knowledge points;

[0040] When a student's time period homework is a test homework, the homework items are set according to the difference between the concentration level of the time period and the concentration threshold, and the homework items are adjusted according to the corresponding student's historical homework items and historical accuracy rate;

[0041] When a student's time period homework is a teaching homework, it is determined whether the corresponding student's time period concentration is lower than the non-learning threshold. If yes, the corresponding student's teaching homework is set as a teaching video; if not, the corresponding student's teaching homework is set as a guidance homework;

[0042] Integrate all the time assignments for this class to generate after-class homework.

[0043] Preferably, the step of setting the homework item according to the difference between the concentration level in each time period and the concentration threshold value includes:

[0044] Obtain the time period concentration and concentration threshold; calculate the difference between the time period concentration and the concentration threshold; when the difference is less than the medium threshold, set the corresponding job type to a simple job; when the difference is greater than the medium threshold and less than the high threshold, set the corresponding job type to a medium job; when the difference is greater than the high threshold, set the corresponding job type to a difficult job; wherein, the job items include the job type and the number of jobs; the medium threshold and the high threshold are set by experience, and the high threshold is greater than the medium threshold, such as the medium threshold is 5% of the concentration threshold, and the high threshold is 10% of the concentration threshold, etc.

[0045] This application generates homework items based on the students' concentration in each time period, and adjusts the homework items based on the historical accuracy rate, so that the after-class homework is more suitable for the students' current status, which is convenient for students to review and reinforce the content of the class. Setting homework items of different levels according to the students' class status can stimulate students' learning motivation and avoid problems such as questions being too difficult, which will dampen students' enthusiasm, and questions being too simple, which will prevent students from in-depth learning.

[0046] Preferably, the concentration threshold is determined by comprehensively considering the difficulty of the teaching subject, the teaching quality of the teacher and the concentration of the student, including:

[0047] The teaching quality scores of each teacher obtained through the cloud platform are marked as JS; the difficulty scores of the teaching subjects are obtained as KM;

[0048] The database obtains the mode of several historical concentrations corresponding to the teaching subjects with the highest student test scores, and records it as the standard concentration, marked as BZ; the standard concentration can be understood as the optimal concentration, which can be understood as the highest learning efficiency of the corresponding student using this concentration; the difficulty score of the teaching subject with the highest student test score is marked as BK; the teaching quality score of the teacher of the corresponding subject is marked as BJ;

[0049] The concentration threshold ZY of the corresponding student under the teaching of the corresponding subject by the corresponding teacher is calculated by the formula ZY=[α1×exp((JS-BJ) / MJ)+α2×exp((KM-BK) / MK)]×BZ; among them, MJ is the full score of the teaching quality score. The higher the teaching quality score, the higher the teaching level of the teacher and the higher the student's concentration in class; MK is the full score of the difficulty score of the teaching subject. The higher the difficulty score of the teaching subject, the greater the learning difficulty of the corresponding teaching subject and the higher the student's concentration required; α1 and α2 are weight coefficients.

[0050] Preferably, the central processing module is further used to determine whether the student has abnormalities based on the concentration level during a period, including:

[0051] Obtain the concentration level of each time period and determine whether the concentration level is less than the abnormal concentration threshold. The abnormal concentration threshold is set based on experience, such as 5% of the corresponding student's concentration threshold.

[0052] If yes, the abnormal parameter is increased by one. When the abnormal parameter is greater than the abnormal parameter threshold, an abnormal instruction is generated and sent to the teacher terminal. The abnormal parameter represents the number of times the concentration level in a period is continuously less than the abnormal concentration threshold. The abnormal parameter threshold is a positive integer set by experience and can be 3. Otherwise, the concentration level in the next period is determined to be less than the concentration threshold.

[0053] If not, the abnormal parameter is cleared to zero; and the concentration in the next period is judged to be less than the concentration threshold.

[0054] In the actual teaching process, some students are not good at communication. If they have physical problems or are in a bad mood during class, the teacher cannot find out in time. Physical problems of students will affect their concentration in class. This application determines whether the student is abnormal based on the number of abnormal concentration times of the student during a period of time, and generates abnormal instructions in time to remind the teacher to pay attention to the student's status. This is conducive to establishing a good relationship between teachers and students.

[0055] A second aspect of the present application provides an interactive teaching management method for a smart classroom, comprising the following steps:

[0056] Step 1: Obtain the teacher's voice information and generate a collection instruction based on the teacher's voice information;

[0057] Step 2: Acquire student images according to acquisition instructions;

[0058] Step 3: Input the student image into the concentration model to obtain the student's concentration during the period;

[0059] Step 4: Generate a reminder signal based on the concentration level and concentration threshold during the time period;

[0060] Step 5: Generate homework based on the concentration level in each time period and the concentration threshold.

[0061] Compared with the prior art, the present invention has the following advantages:

[0062] 1. This application generates a collection instruction based on the teacher's voice information, collects student images according to the collection instruction, obtains the student's concentration at that time through the student image, and generates a reminder signal in time according to the concentration of the time period, so that the teacher can adjust the teaching method and teaching progress according to the student's class status; at the same time, it generates teaching homework according to the concentration of the time period, and can set corresponding types of questions according to the student's listening status of different knowledge points during class to better suit the student's learning situation, so that students can have learning resources that suit them.

[0063] 2. This application determines whether a student is abnormal based on the number of abnormal concentration levels during a period of time, and generates abnormal instructions in a timely manner to remind teachers to pay attention to the student's status; this facilitates the establishment of a good relationship between teachers and students. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0065] Figure 1 This is a schematic diagram of the principles of this application;

[0066] Figure 2 A diagram showing the steps of the method of this application;

[0067] Figure 3 This is a single student status bar chart in this application;

[0068] Figure 4 This is a single student status bar chart in this application;

[0069] Figure 5 This is a single student status bar chart in this application. DETAILED DESCRIPTION

[0070] The following will clearly and completely describe the technical solutions of this application in conjunction with the embodiments. Obviously, the embodiments described are only a part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0071] See also Figure 1-Figure 2 The first embodiment of the present application provides an interactive teaching management system and method for a smart classroom, including:

[0072] Central processing module, and the data acquisition module, display control terminal and database connected thereto;

[0073] Data acquisition module: obtains teacher voice information, video information and student images through data acquisition equipment; data acquisition equipment includes recorders and high-definition cameras, etc. Teacher voice information is the content of the teacher's speech recorded by the recorder in the classroom; video information includes teaching video information and student video information. Teaching video information is the teacher's class video recording; student video information is the status video of each student in class;

[0074] Central processing module: Acquires teacher voice information and generates acquisition instructions based on the teacher voice information; acquires student images according to the acquisition instructions, inputs the student images into the concentration model to obtain the student's concentration level for each period; generates reminder signals and after-class homework based on the concentration level for each period and the concentration threshold; the concentration model is trained through an artificial intelligence model;

[0075] The display control terminal includes a teacher terminal and a student terminal; the teacher terminal is used to display the student status and teaching content during class; the student terminal is used to display the teacher's teaching content;

[0076] like Figure 3-5 , is a single student status bar graph, which includes: student name, student time period concentration, concentration threshold and abnormal status; among them, when the student's time period concentration is 0, the corresponding student status bar will be displayed in red; when the student's time period concentration is less than the corresponding concentration threshold, the corresponding student status bar will be displayed in yellow; when the student's time period concentration is greater than or equal to the corresponding concentration threshold, the corresponding student status bar will be displayed in green; when the time period concentration is less than the concentration threshold, the abnormal status of the corresponding student is judged to be yes, otherwise, the abnormal status of the student is judged to be no;

[0077] It is worth noting that the student status overview displayed on the teaching terminal will be updated in real time according to the concentration of each student in each time period obtained by the collection instruction; and the update time is marked;

[0078] It is understandable that when a student's physical condition has problems, the value of their concentration during the period will decrease, which will be displayed to the corresponding teacher through the student status bar;

[0079] Database: used to store historical data, teaching content, student status, etc.

[0080] This application generates collection instructions based on the teacher's voice information, collects student images according to the collection instructions, obtains the student's concentration at that time period through the student image, and generates reminder signals in time according to the concentration of the time period, so that the teacher can adjust the teaching method and teaching progress according to the student's class status; at the same time, teaching homework is generated according to the concentration of the time period, and the corresponding type of questions can be set according to the student's listening status of different knowledge points during class to better suit the student's learning situation, so that students can have learning resources that suit them.

[0081] Generate collection instructions based on teacher voice information, including:

[0082] Obtain teacher's voice information in real time and convert it into text information;

[0083] Detect whether key words appear in text information;

[0084] If yes, then generate important words of the text according to the text information and generate collection instructions at the same time;

[0085] If not, the teacher's voice information is continuously obtained, the teacher's voice information is converted into text information, and the text information is tested; among them, key vocabulary includes chapter keywords, knowledge point keywords and specific keywords; chapter keywords, knowledge point keywords and specific keywords are all set by experience; for example, chapter keywords of mathematics subjects can be "Section 1: Unit of Length", "Section 1", "Unit of Length", etc.; knowledge point keywords can be "What is a unit", "What is a unit of length", etc.; specific keywords can be "Next is the key point", "Frequently tested points", "Difficult points" and "Easy to make mistakes", etc.; among them, chapter keywords and knowledge point keywords vary with different chapters or subjects of the class, and chapter keywords, knowledge point keywords and specific keywords can also be obtained through big data statistics;

[0086] Specifically, obtain course data of the same chapter taught by different teachers, identify the words that appear more frequently in different course data, and classify these words into chapter keywords, knowledge point keywords and specific keywords; among them, chapter keywords are words related to this chapter; knowledge point keywords are words related to each knowledge point in this chapter; specific keywords are words that are not related to this chapter and each knowledge point in the chapter, but appear more frequently.

[0087] During the actual class, when the teacher talks about some key knowledge points, the students' concentration in class may change. If student images are collected throughout the class, the subsequent image processing will require a huge amount of computation. In the actual class, it is only necessary to focus on some special times for image collection and processing. This application collects and analyzes the teacher's voice information, and generates collection instructions based on the key vocabulary in the teacher's voice information. The collection instructions are instructions for real-time collection of student images. This enables this application to flexibly collect student images according to the teacher's class situation, thereby reducing the amount of computation.

[0088] The central processing module is also used to expand teaching content through the cloud platform, including:

[0089] Extracting chapter keywords and knowledge point keywords from important words in the text; uploading the chapter keywords and knowledge point keywords to the cloud platform; searching the cloud platform for chapter content corresponding to the corresponding chapter keywords and knowledge point content corresponding to the corresponding knowledge point keywords; chapter content can be text information, video information, image information, etc.;

[0090] The chapter content and knowledge point content are sent to the teacher's terminal for screening to obtain teaching content; the teaching content is sent to the student terminal;

[0091] The teaching content is obtained by screening the chapter content and the knowledge point content, including the current affairs mode and the classic mode;

[0092] The current affairs mode is to sort the chapter contents and / or knowledge point contents obtained from the cloud platform in chronological order from recent to remote, and filter out the chapter contents and / or knowledge point contents with the highest ranking;

[0093] The classic mode is to sort the chapter contents and / or knowledge point contents obtained from the cloud platform in descending order according to the number of views and / or according to the content ratings, and filter out the top-ranked chapter contents and / or knowledge point contents; it can be understood that the top ranking is set by the teacher, and the chapter contents and / or knowledge point contents ranked first, third, fifth, etc. can be obtained.

[0094] It is worth noting that in this embodiment, the teacher can also conduct a secondary screening of the chapter content and / or knowledge point content that has been screened out with high rankings, obtain the chapter content and / or knowledge point content that is finally approved by the teacher, and then integrate the corresponding chapter content and / or knowledge point content to generate teaching content; this is convenient for selecting chapter content and / or knowledge point content that are suitable for the teacher's teaching for recommendation, and increases the fit between this system and the corresponding teacher's teaching; at the same time, this system will record the publisher of the chapter content and / or knowledge point content corresponding to the teacher's secondary screening; when expanding the teaching content later, the chapter content and / or knowledge point content that has been screened out with high rankings will be sorted again, and the relevant chapter content and / or knowledge point content of the corresponding publisher will be ranked first; this is convenient for the teacher to conduct a secondary screening.

[0095] This application obtains chapter content and knowledge point content through the cloud platform, and sends the chapter content and knowledge point content to student terminals as extended content for teaching; it facilitates students to obtain content related to chapter keywords and knowledge point keywords, and thus facilitates students to understand and learn relevant chapters and knowledge points.

[0096] The focus model is obtained through artificial intelligence model training, including:

[0097] Obtain several images of student status and their corresponding concentration scores; the concentration scores are scored by professionals based on the student status images. For example, if the student lies down to rest, the concentration score is recorded as 0;

[0098] The student status images and their corresponding concentration scores are integrated into a number of training data and a number of test data; the ratio of the number of training data to the number of test data is 7:3;

[0099] Use training data to train the artificial intelligence model; use test data to test the trained artificial intelligence model;

[0100] Specifically, the student status image in the test data is input into the trained artificial intelligence model to obtain the time period concentration score, and the time period concentration score is compared with the corresponding concentration score in the test data. When the difference between the two is within the threshold, the threshold is obtained based on experience, and there is no need to adjust the parameters, and the next set of test data is tested; if it is not within the threshold, the corresponding parameters are adjusted until the output time period concentration score is within the threshold of the concentration score, and then the next set of test data is tested, where the corresponding threshold is set by experience; when the number of test data with the output time period concentration score within the threshold obtained from all test data accounts for 90% or more of the total test data, that is, when the number of test data that pass the test reaches 90%; a concentration model with student images as input and time period concentration scores as output is obtained; this model supports multi-input and multi-output mode, that is, multiple student images are input at the same time, and the corresponding multiple time period concentration scores can be output at the same time;

[0101] Among them, the time period concentration score is recorded as time period concentration; the artificial intelligence model includes a BP neural network model or an RBF neural network model.

[0102] Generate reminder signals based on students' concentration during the period, including:

[0103] Obtain the concentration of the time period corresponding to the student images collected under the same collection instruction for several students, as well as the concentration threshold corresponding to each student;

[0104] Obtain all students whose concentration is less than the corresponding concentration threshold and mark them as target personnel; mark the concentration of the target personnel in the time period as SZi, and mark the corresponding concentration threshold as YZi; where i is the target personnel number, and i=1, 2, 3, ..., n; n is the total number of target personnel in this collection instruction;

[0105] The signal value XH corresponding to this acquisition instruction is calculated by the formula XH=∑[(YZi-SZi) / YZi];

[0106] When the signal value XH is greater than the signal value threshold and is not equal to the total number of target people, it means that most students are not listening carefully, and an interactive instruction is generated; the signal value threshold is greater than zero and less than the total number of target people;

[0107] When the signal value XH is greater than the signal value threshold and equal to the total number of target people; that is, when the concentration of multiple students in a period is zero, the signal value XH will be greater than the signal value threshold and equal to the total number of target people. This indicates that multiple students are lying down to rest and not attending the class. At this time, a rest instruction is generated; the signal value threshold is set by experience, and the value of the signal value threshold is less than the value of the total number of target people;

[0108] When the signal value XH is less than the signal value threshold, it means that a small number of students are not listening carefully, and marking information is generated. The reminder signal includes interactive instructions, rest instructions and marking information. The interactive instructions include small games and jokes related to the course, etc., which are used for teachers to interact with students and improve students' concentration. Rest instructions are used to let students rest for a certain period of time to relieve fatigue, which helps to improve the concentration of students in subsequent learning. The marking information is used to mark the corresponding students to facilitate subsequent teaching.

[0109] This application generates reminder signals based on the concentration level of each time period, reminding teachers to make reasonable teaching arrangements, thereby improving students' learning efficiency. At the same time, it makes it easier for teachers to pay attention to a small number of students who are not focused in class, so as to solve the problem of poor learning effect of certain knowledge points caused by a small number of students being absent in class.

[0110] Generate homework based on students' concentration level and concentration threshold, including:

[0111] Obtain all key vocabulary in this lesson; extract chapter keywords and knowledge point keywords from the key vocabulary, and divide this lesson into several teaching periods based on the chapter keywords and knowledge point keywords; obtain the corresponding time period concentration and corresponding concentration threshold for each student in each teaching period;

[0112] Determine whether the concentration level during the time period is greater than the concentration threshold; if yes, mark the student's homework for the time period as a test homework; if no, mark the student's homework for the time period as a teaching homework; wherein the test homework includes homework items; the teaching homework includes teaching videos and guided homework, and the guided homework is a homework question that explains the knowledge points;

[0113] When a student's time period homework is a test homework, the homework items are set according to the difference between the concentration level of the time period and the concentration threshold, and the homework items are adjusted according to the corresponding student's historical homework items and historical accuracy rate;

[0114] When a student's time period homework is a teaching homework, it is determined whether the corresponding student's time period concentration is lower than the non-learning threshold. If yes, the corresponding student's teaching homework is set as a teaching video; if not, the corresponding student's teaching homework is set as a guidance homework;

[0115] Integrate all the time assignments for this class to generate after-class homework.

[0116] It is worth noting that, in this embodiment, the teaching video information is divided according to the teaching time period to generate teaching videos of different time periods, which are then integrated to generate the teaching video.

[0117] Set up tasks based on the difference between the concentration level in a given time period and the concentration threshold, including:

[0118] Obtain the concentration level and concentration threshold for the time period; calculate the difference between the concentration level for the time period and the concentration threshold; when the difference is less than the medium threshold, set the corresponding assignment type to simple assignment; when the difference is greater than the medium threshold and less than the high threshold, set the corresponding assignment type to medium assignment; when the difference is greater than the high threshold, set the corresponding assignment type to difficult assignment; where the assignment items include the assignment type and the number of assignments; the medium and high thresholds are set based on experience, and the high threshold is greater than the medium threshold, e.g., the medium threshold is 5% of the concentration threshold, and the high threshold is 10% of the concentration threshold, etc.;

[0119] This application generates homework items based on the students' concentration in each time period, and adjusts the homework items based on the historical accuracy rate, so that the after-class homework is more suitable for the students' current status, which is convenient for students to review and reinforce the content of the class. Setting homework items of different levels according to the students' class status can stimulate students' learning motivation and avoid problems such as questions being too difficult, which will dampen students' enthusiasm, and questions being too simple, which will prevent students from in-depth learning.

[0120] Among them, simple assignments are assignments that test knowledge points; medium assignments are assignments that test the expansion of knowledge points; difficult assignments are assignments that test the integration of knowledge points; the assignment type corresponding to the question can be set based on experience, and the assignment types include simple assignments, medium assignments, and difficult assignments.

[0121] Obtain the historical data of the last homework through the database; extract the historical homework items in the historical data, and the historical accuracy rate corresponding to the historical homework type in each historical homework item;

[0122] When the job type in the current left and right projects is a simple job, determine whether the historical jobs meet the following conditions:

[0123] The historical homework items include simple homework, and the historical accuracy rate of the simple homework is greater than the upgrade threshold 1;

[0124] The historical work items include medium-level work, and the historical accuracy rate of the medium-level work is greater than the upgrade threshold 2;

[0125] The historical assignments include difficult assignments, and the historical accuracy rate of the difficult assignments is greater than the upgrade threshold of three;

[0126] When the historical homework meets any of the above three conditions: the set number of questions in this simple homework will be upgraded to medium homework; otherwise, the homework type of this time will remain simple homework; the set number is set by experience, and the number of homework is set by experience; the set number of questions can be 50% of the number of homework.

[0127] When the job type in the current left and right projects is medium job; determine whether the historical job meets the following conditions:

[0128] The historical work items include medium-level work, and the historical accuracy rate of the medium-level work is greater than the upgrade threshold 2;

[0129] The historical homework project includes difficult homework, and the historical accuracy rate of the difficult homework is greater than the upgrade threshold of three;

[0130] When the historical homework meets any one of the above two conditions: a set number of questions in this medium homework will be upgraded to difficult homework; wherein, the set number of questions can be 30% of the total number of homework; otherwise, when the historical accuracy rate corresponding to the medium homework is less than the upgrade threshold of four, a set number of questions in this medium homework will be downgraded to simple homework; wherein, the set number of questions can be 30% of the total number of homework.

[0131] When the historical homework includes difficult homework, and the historical accuracy rate of the difficult homework is lower than the upgrade threshold of five, a set number of questions in this difficult homework will be downgraded to medium homework; the set number of questions can be 50% of the total number of homework.

[0132] Specifically, upgrade threshold 1, upgrade threshold 2, upgrade threshold 3, upgrade threshold 4, and upgrade threshold 5 are all set based on experience; upgrade threshold 1 is greater than upgrade threshold 2, upgrade threshold 2 is greater than upgrade threshold 3; upgrade threshold 5 is less than upgrade threshold 4; for example, upgrade threshold 1 can be set to 90%, upgrade threshold 2 to 80%, upgrade threshold 3 to 60%, upgrade threshold 4 to 20%, and upgrade threshold 5 to 10%;

[0133] It is worth noting that in this embodiment, obtaining the time period concentration of any student in each time period includes:

[0134] Obtain the duration of each teaching period, set an interception instruction according to the duration, and intercept the student image of the corresponding student in the video information according to the interception instruction;

[0135] Input the student image into the concentration model to obtain the corresponding concentration of the corresponding student; number the corresponding concentration in the order of the time of intercepting the instruction, and mark the numbered concentration as ZZj; where j is the number, and j=1, 2, 3, ..., m

[0136] The time period concentration DZ is calculated by the formula DZ=∑(αj×ZZj); where j=1, 2, 3, ..., m; m is the total number of numbers for the corresponding concentrations according to the time sequence of the intercepted instructions; αj is the weight coefficient; αj=j / ∑(j).

[0137] Specifically, the interception instruction can be set according to the duration. The duration can be divided into several interception periods, and the interception instruction is generated at the end of each interception period. For example, if the teaching period is 10 minutes long, it can be divided into 10 interception periods of 1 minute each. The interception instruction is generated at the last second of the first minute of the corresponding teaching period, the last second of the second minute, and so on. The interception instruction is generated at the last second of the tenth minute.

[0138] In this embodiment, three interception instructions are generated within a certain teaching period. The concentration corresponding to the first interception instruction is ZZ1=69, the concentration corresponding to the second interception instruction is ZZ2=60, and the concentration corresponding to the third interception instruction is ZZ3=64. The period concentration DZ=63.5 is calculated by the formula DZ=∑(αj×ZZj).

[0139] In this embodiment, the time period concentration can also be generated based on the teacher's voice information, including:

[0140] Acquire several segments of teacher voice information during the entire teaching period, generate interception instructions based on the teacher voice information; intercept student images of corresponding students in the video information according to the interception instructions;

[0141] Input the student image into the concentration model to obtain the corresponding concentration of the corresponding student; number the corresponding concentration according to the time sequence of the intercepted instructions or according to the duration of the teacher's voice information, and mark the numbered concentration as ZZk; where k is the number; generate the time period concentration based on the duration and characteristic vocabulary of the teacher's voice information and the concentration level;

[0142] The time period concentration DZ is calculated by the formula DZ=∑(βk×ZZk); where k=1, 2, 3, ..., q; q is the total number of k; βk is the weight coefficient;

[0143] The duration of the teacher's voice information is obtained and marked as YSk; each segment of the teacher's voice information is converted into text information, and the number of characteristic words in the text information is obtained and counted and marked as Tsk; the weight coefficient βk numbered k is calculated using the formula βk=γ1×YSk / ∑(YSk)+γ2×TSk / ∑(TSk); where k=1, 2, 3, ..., q; γ1 and γ2 are weight coefficients set by experience;

[0144] In this embodiment, γ1=0.4, γ2=0.6 are set; three interception instructions are generated according to the teacher's voice information during a certain teaching period. The first interception instruction corresponds to a voice duration YS1=5 minutes, and the number of characteristic words TSk=4; the second interception instruction corresponds to a voice duration YS1=10 minutes, and the number of characteristic words TSk=5; the first interception instruction corresponds to a voice duration YS1=6 minutes, and the number of characteristic words TSk=2;

[0145] The first interception command corresponds to a concentration level of ZZ1 = 69; the second interception command corresponds to a concentration level of ZZ2 = 60; the third interception command corresponds to a concentration level of ZZ3 = 64;

[0146] The time period concentration DZ≈63.71 is calculated using the formula DZ=∑(βk×ZZk).

[0147] The concentration threshold is determined by the difficulty of the teaching subject, the teaching quality of the teaching staff and the concentration of the students, including:

[0148] The teaching quality scores of each teacher obtained through the cloud platform are marked as JS; the difficulty scores of the teaching subjects are obtained as KM;

[0149] The database obtains the mode of several historical concentrations corresponding to the teaching subjects with the highest student test scores, and records it as the standard concentration, marked as BZ; the standard concentration can be understood as the optimal concentration, which can be understood as the highest learning efficiency of the corresponding student using this concentration; the difficulty score of the teaching subject with the highest student test score is marked as BK; the teaching quality score of the teacher of the corresponding subject is marked as BJ;

[0150] The concentration threshold ZY of the corresponding student under the teaching of the corresponding subject by the corresponding teacher is calculated by the formula ZY=[α1×exp((JS-BJ) / MJ)+α2×exp((KM-BK) / MK)]×BZ; among them, MJ is the full score of the teaching quality score. The higher the teaching quality score, the higher the teaching level of the teacher and the higher the student's concentration in class; MK is the full score of the difficulty score of the teaching subject. The higher the difficulty score of the teaching subject, the greater the learning difficulty of the corresponding teaching subject and the higher the student's concentration required; α1 and α2 are weight coefficients.

[0151] In this embodiment, the full score of the teaching quality score is set to MJ=100; the full score of the teaching subject difficulty score is set to MK=100; α1=0.6, α2=0.4; the value range of the teaching quality score and the teaching difficulty score is 0 to 100;

[0152] Obtain the standard concentration BZ=80 corresponding to the teaching subject with the highest test score of a student. The corresponding teaching subject is mathematics, the corresponding difficulty score BK=70, and the corresponding teaching teacher's teaching quality score BJ=70;

[0153] The student's current subject is physics, the difficulty score of the subject is KM=85, and the teacher's teaching quality score is JS=80;

[0154] The concentration threshold ZY≈90.22 of the corresponding student in the corresponding subject of physics taught by the corresponding teacher is calculated through the formula ZY=[α1×exp((JS-BJ) / MJ)+α2×exp((KM-BK) / MK)]×BZ.

[0155] The central processing module is also used to determine whether students have abnormalities based on their concentration during a certain period of time, including:

[0156] Obtain the concentration level of each time period and determine whether the concentration level is less than the abnormal concentration threshold. The abnormal concentration threshold is set based on experience, such as 5% of the corresponding student's concentration threshold.

[0157] If yes, the abnormal parameter is increased by one. When the abnormal parameter is greater than the abnormal parameter threshold, an abnormal instruction is generated and sent to the teacher terminal. The abnormal parameter represents the number of times the concentration level in a period is continuously less than the abnormal concentration threshold. The abnormal parameter threshold is a positive integer set by experience and can be 3. Otherwise, the concentration level in the next period is determined to be less than the concentration threshold.

[0158] If not, the abnormal parameter is cleared to zero; and the concentration in the next period is judged to be less than the concentration threshold.

[0159] In the actual teaching process, some students are not good at communication. If they have physical problems or are in a bad mood during class, the teacher cannot find out in time. Physical problems of students will affect their concentration in class. This application determines whether the student is abnormal based on the number of abnormal concentration times of the student during a period of time, and generates abnormal instructions in time to remind the teacher to pay attention to the student's status. This is conducive to establishing a good relationship between teachers and students.

[0160] The central processing module is also used to generate reinforcement operations, including:

[0161] Obtain the special case questions that the teacher explains in class, search for questions similar to the special case questions in the database, and divide them into simple questions, medium questions and difficult questions according to the accuracy rate; and according to the homework type in the student's homework project, extract a set number and type of questions from them to generate reinforcement homework; specifically, if the homework type in the homework project is simple homework, extract a set number of questions from the simple questions as reinforcement homework; if the homework type in the homework project is medium homework, extract a set number of questions from the medium questions as reinforcement homework; if the homework type in the homework project is difficult homework, extract a set number of questions from the difficult questions as reinforcement homework; the question type of the reinforcement homework can also be adjusted according to the corresponding student's historical homework project and historical accuracy rate, and the specific method is the same as the method of adjusting the homework project according to the corresponding student's historical homework project and historical accuracy rate.

[0162] A second aspect of the present application provides an interactive teaching management method for a smart classroom, comprising the following steps:

[0163] Step 1: Obtain the teacher's voice information and generate a collection instruction based on the teacher's voice information;

[0164] Step 2: Acquire student images according to acquisition instructions;

[0165] Step 3: Input the student image into the concentration model to obtain the student's concentration during the period;

[0166] Step 4: Generate a reminder signal based on the concentration level and concentration threshold during the time period;

[0167] Step 5: Generate homework based on the concentration level in each time period and the concentration threshold.

[0168] Some of the data in the above formula are calculated by removing the dimensions and taking their numerical values. The formula is a formula that is closest to the actual situation obtained by software simulation of a large amount of collected data; the preset parameters and preset thresholds in the formula are set by technical personnel in this field according to actual conditions or obtained through simulation of a large amount of data.

[0169] How this application works:

[0170] This embodiment obtains teacher voice information, generates collection instructions based on the teacher voice information, and collects student images based on the collection instructions; inputs the student image into the concentration model to obtain the student's time period concentration; sets the concentration threshold based on the difficulty of the teaching subject, the teaching quality of the teaching teacher and the student's concentration; generates a reminder signal based on the time period concentration and the concentration threshold; generates homework based on the time period concentration and the concentration threshold; and can set corresponding types of questions based on the students' listening status of different knowledge points during class to better suit the students' learning situation, so that students can have learning resources that suit them.

[0171] The above embodiments are only used to illustrate the technical method of the present application and are not intended to limit it. Although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present application can be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present application.

Claims

1. The interactive teaching management system of the smart classroom is characterized by: include: Central processing module, data acquisition module connected to the central processing module, display control terminal and database; Among them, the data acquisition module: obtains teacher voice information, video information and student images through data acquisition equipment; Central processing module: used to obtain teacher voice information and generate acquisition instructions based on the teacher voice information; collect student images according to the acquisition instructions, input the student images into the concentration model to obtain the student's concentration level during the period; generate reminder signals and after-class homework based on the concentration level during the period and the concentration threshold; wherein the concentration model is obtained through artificial intelligence model training; Generating the after-class homework based on the time period concentration and the concentration threshold includes: obtaining all key vocabulary in the current lesson; extracting chapter keywords and knowledge point keywords from the key vocabulary, and dividing the current lesson into several teaching periods according to the chapter keywords and knowledge point keywords; obtaining the time period concentration and the corresponding concentration threshold corresponding to each student in each teaching period; judging whether the time period concentration is greater than the concentration threshold; if so, marking the time period homework of the corresponding student as an examination homework; if not, marking the time period homework of the corresponding student as a teaching homework; wherein the examination homework includes homework items; and the teaching homework includes teaching videos and guidance homework; When a student's time period homework is a test homework, the homework items are set according to the difference between the concentration level of the time period and the concentration threshold, and the homework items are adjusted according to the corresponding student's historical homework items and historical accuracy rate; When a student's time period homework is a teaching homework, it is determined whether the corresponding student's time period concentration is lower than the non-learning threshold. If yes, the corresponding student's teaching homework is set as a teaching video; if not, the corresponding student's teaching homework is set as a guidance homework; Integrate all the period assignments of this class to generate after-class assignments; The display control terminal includes a teacher terminal and a student terminal; the teacher terminal is used to display the student's class status; the student terminal is used to display the teacher's teaching content; Database: used to store historical data, teaching content and student status.

2. The interactive teaching management system for smart classrooms according to claim 1, characterized in that: The generating of the collection instruction based on the teacher's voice information includes: Obtain teacher's voice information in real time and convert it into text information; Detect whether key words appear in text information; If yes, then generate important words of the text according to the text information and generate collection instructions at the same time; If not, the teacher's voice information is continuously obtained, converted into text information, and the text information is tested; among them, the key vocabulary includes chapter keywords, knowledge point keywords and specific keywords.

3. The interactive teaching management system for smart classrooms according to claim 2, characterized in that: The central processing module is also used to expand teaching content through the cloud platform, including: Extract chapter keywords and knowledge point keywords from important words in the text; Search for corresponding chapter content and knowledge points in the cloud platform; Send chapter content and knowledge point content to the teacher's terminal for screening to obtain teaching content; Send teaching content to student terminals.

4. The interactive teaching management system for smart classrooms according to claim 1, characterized in that: The concentration model is obtained through artificial intelligence model training, including: Obtain several student status images and the concentration scores corresponding to each student status image; The student status image and the concentration score corresponding to the student status image are integrated into training data and test data; the training data is used to train the artificial intelligence model; the trained artificial intelligence model is tested using the test data, and when the test data that passes the test reaches 90%, a concentration model is obtained whose input is the student image and whose output is the time period concentration score; wherein the time period concentration score is recorded as the time period concentration; the artificial intelligence model includes a BP neural network model or an RBF neural network model.

5. The interactive teaching management system for smart classrooms according to claim 1, characterized in that: The reminder signal includes an interaction instruction, a rest instruction and a marking information; Generating the reminder signal based on the concentration level during the time period and the concentration level threshold includes: Obtain the concentration of the time period corresponding to the student images collected under the same collection instruction for several students, as well as the concentration threshold corresponding to each student; Mark all students whose concentration in each time period is less than the corresponding concentration threshold as target personnel; mark the concentration of the target personnel in each time period as SZi, and mark the corresponding concentration threshold as YZi; where i is the number of the target personnel, and i=1, 2, 3, ..., n; n is the total number of target personnel in this collection instruction; The signal value XH corresponding to this acquisition instruction is calculated by the formula XH=∑[(YZi-SZi) / YZi]; When the signal value XH is greater than the signal value threshold and is not equal to the total number of target personnel, an interactive instruction is generated; When the signal value XH is greater than the signal value threshold and equal to the total number of target personnel, a rest instruction is generated; When the signal value XH is less than the signal value threshold, marking information is generated; wherein the interactive instructions include mini-games and jokes related to the course.

6. The interactive teaching management system for smart classrooms according to claim 1, characterized in that: The task items are set according to the difference between the concentration level in a time period and the concentration threshold, including: Obtain the time period concentration and concentration threshold; calculate the difference between the time period concentration and the concentration threshold; when the difference is less than the medium threshold, set the job type to simple job; when the difference is greater than the medium threshold and less than the high threshold, set the job type to medium job; when the difference is greater than the high threshold, set the job type to difficult job; where the job items include the job type and the number of jobs; the high threshold is greater than the medium threshold.

7. The interactive teaching management system for smart classrooms according to claim 1, characterized in that: The concentration threshold is determined by the teaching subject, the teaching teacher and the student, including: Obtain the teaching quality scores of each teacher through the cloud platform and mark them as JS; obtain the difficulty scores of the teaching subjects and mark them as KM; The database obtains the mode of several historical concentrations corresponding to the teaching subject with the highest student test score, and records it as the standard concentration and marks it as BZ; marks the difficulty score of the teaching subject with the highest student test score as BK; marks the teaching quality score of the teacher of the corresponding subject as BJ; The concentration threshold ZY of the corresponding student under the teaching of the corresponding subject by the corresponding teacher is calculated by the formula ZY=[α1×exp((JS-BJ) / MJ)+α2×exp((KM-BK) / MK)]×BZ; among them, MJ is the full mark of the teaching quality score; MK is the full mark of the difficulty score of the teaching subject; α1 and α2 are weight coefficients.

8. The interactive teaching management system for smart classrooms according to claim 1, characterized in that: The central processing module is also used to determine whether the student has abnormalities based on the concentration level during a period, including: Obtain the concentration of the time period and determine whether the concentration of the time period is less than the abnormal concentration threshold; If yes, the abnormal parameter is increased by one. When the abnormal parameter is greater than the abnormal parameter threshold, an abnormal instruction is generated and sent to the teacher terminal. Otherwise, the concentration level in the next period is determined to be less than the concentration threshold. The abnormal parameter represents the number of times the concentration level in the period is continuously less than the abnormal concentration threshold. The abnormal parameter threshold is a set positive integer. If not, the abnormal parameter is cleared to zero; and the concentration in the next period is judged to be less than the concentration threshold.

9. A method for managing interactive teaching in a smart classroom, based on the interactive teaching management system of a smart classroom according to any one of claims 1 to 8, characterized in that: The following steps are involved: Step 1: Obtain the teacher's voice information and generate a collection instruction based on the teacher's voice information; Step 2: Collect student images according to the collection instructions; Step 3: Input the student image into the concentration model to obtain the student's concentration during the period; Step 4: Generate a reminder signal based on the concentration level and concentration threshold during the time period; Step 5: Generate homework based on the concentration level in each time period and the concentration threshold.

Images