A smart education system based on educational big data

Abstract

The present invention relates to the field of smart education technology, and is used to solve the problem that traditional smart education is mostly limited to monitoring and analyzing a specific learning stage or a certain dimension. In particular, a smart education system based on education big data is disclosed, including a data acquisition module, a pre-class analysis module, an in-class analysis module, a post-class analysis module, a comprehensive analysis module, a management execution module, a cloud database and a display terminal. The present invention collects student learning data information, and monitors and calculates key indicators such as the student's preview preparation value, mastery value, resource usage value, body concentration value, eye concentration value, completion value, test value and habit value, thereby generating pre-class, in-class and post-class status type signals, and integrating and analyzing these type signals to generate corresponding management type signals, and displaying corresponding text information feedback, thereby providing accurate and comprehensive student learning status feedback and management suggestions for education managers.

Description

Technical Field

[0001] The present invention relates to the field of smart education technology, and in particular to a smart education system based on education big data. Background Art

[0002] With the rapid development of Internet technology and the widespread application of big data in the field of education, the education industry is rapidly entering a new era of smart education. More and more schools are adopting Internet-based teaching models, equipping each student with exclusive learning tools, such as tablets. Students can log in through their personal accounts and perform various learning tasks. Compared with traditional education methods, this model can more comprehensively and real-timely grasp the learning status of students, thereby improving the quality of education and teaching effectiveness.

[0003] However, current smart education systems are mostly limited to monitoring and analyzing a specific learning stage (such as classroom teaching) or a certain dimension (such as student performance). They often lack comprehensive analysis before, during and after class, and are unable to fully evaluate students' learning status. More importantly, existing systems usually lack effective data integration and analysis functions, and are unable to track students' entire learning process, nor provide personalized guidance and support. Therefore, these smart education systems are difficult to accurately reflect changes in students' status throughout the learning process, affecting the accuracy and effectiveness of education management.

[0004] In order to solve the above defects, a technical solution is now provided. Summary of the invention

[0005] The purpose of this invention is to solve the problem that traditional smart education is mostly limited to monitoring and analyzing a specific learning stage or a certain dimension, cannot achieve full tracking of students' learning process, and is difficult to provide personalized guidance and support, and proposes a smart education system based on educational big data.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] A smart education system based on education big data, including:

[0008] Data collection module, used to collect students' learning data information;

[0009] The pre-class analysis module is used to monitor and calculate the students' preparation value, mastery value and resource usage value, thereby analyzing the students' pre-class status and obtaining the pre-class evaluation index, so as to generate a pre-class status type signal and send it to the comprehensive analysis module;

[0010] The in-class analysis module is used to monitor and calculate the students' body concentration value and eye concentration value, thereby analyzing the students' in-class status and obtaining the in-class evaluation index, so as to generate an in-class status type signal and send it to the comprehensive analysis module;

[0011] The post-class analysis module is used to monitor and calculate the completion value, test value and habit value of the students, thereby analyzing the students' post-class status and obtaining the post-class evaluation index to generate a post-class status type signal and send it to the comprehensive analysis module;

[0012] A comprehensive analysis module is used to receive the pre-class status type signal, the in-class status type signal and the post-class status type signal, and perform comprehensive analysis and processing thereon to generate a management type signal and send it to the management execution module;

[0013] The management execution module is used to receive a management type signal to generate text words.

[0014] Furthermore, the specific solution process for the students' preparation value, mastery value, and resource usage value is as follows:

[0015] By obtaining the students' pre-class login records, each pre-class login is marked as a pre-class monitoring point;

[0016] Extract the initial login time and offline time of each pre-class monitoring point within a period of time, and calculate the difference between the initial login time and the offline time to obtain the login time Qds of each pre-class monitoring point within a period of time i ;

[0017] Extract the preview time Qys of each pre-class monitoring point within a period of time i ;

[0018] According to the formula: , we get the preview preparation value KQ1, where i represents the number of each pre-class monitoring point within a period of time, and i=1, 2, 3…n, n represents the total number of pre-class monitoring point numbers, Qys i * represents the reference preview duration, a1 and a2 represent the set weight factors respectively;

[0019] Extract the total answer value Qzt of each pre-class monitoring point within a period of time i 、Correct answer value Qdt i and the average time to answer the questions Qst i , according to the formula: , and obtain the master value KQ2, where a3 and a4 represent the set weight factors respectively;

[0020] Extract the video viewing time Qsp of each pre-class monitoring point within a period of time i 、Video views Qcp i , Courseware browsing time Qsk i And the number of courseware views Qck i , according to the formula: , get the resource usage value KQ3, where QS * Indicates the reference ratio of video viewing time to courseware browsing time, QC * It represents the reference ratio of video viewing to courseware browsing times, and a5 and a6 represent the set weight factors respectively.

[0021] Furthermore, the students' pre-class status is analyzed. The specific analysis process is as follows:

[0022] By extracting the values ​​of the preview preparation value KQ1, the mastery value KQ2 and the resource utilization value KQ3, normalization is performed according to the formula: , we get the pre-class evaluation index KQP, where e represents a natural constant, η1, η2, η3 represent the weight coefficients of the pre-study preparation value, mastery value, and resource use value, respectively, and η1>η2>η3;

[0023] The student's pre-class evaluation index is compared and matched with the pre-class status determination table to obtain the student's pre-class status and generate a corresponding pre-class status type signal, and the pre-class status type signal includes an excellent pre-class status signal, a good pre-class status signal and a poor pre-class status signal.

[0024] Furthermore, the specific process of solving the student's body concentration value is as follows:

[0025] Extract the students' performance videos in class over a period of time and analyze them to obtain images of the students' performance in class over a period of time;

[0026] The student's body contour image is extracted from the performance image, thereby obtaining each body contour image of the student, and each body contour image of the student is matched and analyzed with the reference body contour image library to obtain the number of successfully matched images, and the ratio of the number of successfully matched images to the total number of matches is calculated to obtain the body focus value KZ1.

[0027] Furthermore, the specific process of solving the students' eye focus value is as follows:

[0028] Extracting the trainee's eye images from the performance images, and combining the trainee's eye images in sequence according to the time frame order, thereby obtaining each eye video of the trainee, extracting the duration of the interval between adjacent blinking actions in each eye video, thereby obtaining the duration of each blinking interval in each eye video, comparing and analyzing the duration of each blinking interval in each eye video with the set reference duration, if the duration of a blinking interval in each eye video is greater than the set reference duration, marking a blinking interval in each eye video as an abnormal interval period, thereby obtaining each abnormal interval period, extracting the eye image of each abnormal interval period, and matching it with the reference eye image, thereby obtaining each abnormal eye-opening period and each abnormal eye-closing period respectively;

[0029] Integrate each abnormal eye-closing period to obtain the eye-closing duration value Zby;

[0030] Extract the number of eyeball rotations in each abnormal eye-opening period, and compare and analyze it with the set reference number of eyeball rotations. If the number of eyeball rotations in a certain abnormal eye-opening period is less than the set reference number of eyeball rotations, then the abnormal eye-opening period is determined as a daze period, and the daze periods determined as such are integrated to obtain the daze duration value Zfd;

[0031] According to the formula: , and get the eye focus value KZ2, where Zzs represents the total duration of each eye video of the student, and b1 and b2 represent the set weight factors respectively.

[0032] Furthermore, the students' in-class status is analyzed. The specific analysis process is as follows:

[0033] By extracting the values ​​of body focus value KZ1 and eye focus value KZ2, normalization is performed according to the formula: , we get the in-class evaluation index KZP, where u represents a natural constant, η4 ​​and η5 represent the weight coefficients of body concentration value and eye concentration value, respectively, and η5>η4;

[0034] The student's in-class evaluation index is compared and matched with the in-class status determination table to obtain the student's in-class status and generate a corresponding in-class status type signal, and the in-class status type signal includes an excellent in-class status signal, a good in-class status signal and a poor in-class status signal.

[0035] Furthermore, the specific process of solving the students' completion value, test value and habit value is as follows:

[0036] By obtaining the students' after-class login records, each after-class login is marked as a after-class monitoring point;

[0037] Extract the total number of tasks and the number of completed tasks at each after-school monitoring point within a period of time, and calculate the ratio of the number of completed tasks to the total number of tasks to obtain the completion value KH1 j , where j represents the number of each after-school monitoring point within a period of time, and j=1, 2, 3…m, and m represents the total number of after-school monitoring point numbers;

[0038] Extract the average test score of each after-class monitoring point within a period of time as the test value KH2 of each after-class monitoring point within a period of time j ;

[0039] Extract the login time of each after-school monitoring point within a period of time, and calculate the difference between it and the set login time to obtain the habit value KH3 j, where setting the login time is specifically: extracting the login time of each after-class monitoring point within the student's historical unit time period, and arranging them in chronological order to obtain a login time sequence, and extracting the login time corresponding to the majority from the login time sequence as the set login time.

[0040] Furthermore, the students' after-class status is analyzed. The specific analysis process is as follows:

[0041] Complete the value KH1 by extracting j , test value KH2 j and custom value KH3 j , according to the formula: , and get the post-class evaluation index KHP, where KH1 j-1 、KH2 j-1 、KH3 j-1 They represent the completion value, test value and habit value of the j-1th post-class monitoring point, respectively, KH1 * 、KH2 * 、KH3 * denote reference completion value, reference test value and reference habit value respectively, c1, c2, c3, c4, c5 and c6 denote set weight factors respectively, η6, η7 and η8 denote weight coefficients of completion fluctuation degree, test fluctuation degree and habit fluctuation degree respectively, and η6>η7>η8;

[0042] The student's after-class evaluation index is compared and matched with the after-class status determination table to obtain the student's after-class status and generate a corresponding after-class status type signal, and the after-class status type signal includes an excellent after-class status signal, a good after-class status signal and a poor after-class status signal.

[0043] Furthermore, the specific analysis process of generating management type signals is as follows:

[0044] According to the pre-class state type signal, a set R is established, and the excellent pre-class state signal is marked as element r1, the good pre-class state signal is marked as element r2, and the poor pre-class state signal is marked as element r3, and element r1∈set R, element r2∈set R, and element r3∈set R;

[0045] According to the class status type signal, a set Y is established, the excellent class status signal is marked as element y1, the good class status signal is marked as element y2, the poor class status signal is marked as element y3, and the element y1∈set Y, the element y2∈set Y, and the element y3∈set Y;

[0046] According to the after-class status type signal, T is established, the excellent after-class status signal is marked as element t1, the good after-class status signal is marked as element t2, and the poor after-class status signal is marked as element t3, and element t1∈set T, element t2∈set T, element t3∈set T;

[0047] The sets R, Y and T are unioned. If R∪Y∪T={r3, y1, t1} or {r3, y2, t1} or {r3, y1, t2}, a first-order management signal is generated. If R∪Y∪T={r1, y3, t1} or {r2, y3, t1} or {r1, y3, t2}, a second-order management signal is generated. If R∪Y∪T={r1, y1, t3} or {r2, y1, t3} or {r1, y2, t3}, a third-order management signal is generated. If R∪Y∪T={r2, y2, t2} or {r3, y3, t3}, a fourth-order management signal is generated.

[0048] Furthermore, the specific analysis process of the generated text words is as follows:

[0049] If the first-order management signal is captured, a text message is generated, "The student's learning state before class is poor, showing low readiness and participation, and needs to strengthen supervision and guidance before class learning" and sent to the display terminal for display explanation;

[0050] If the second-order management signal is captured, a text message is generated, "The student's learning status in class is poor, showing a lack of concentration, and needs to be supervised and guided during class learning." This is sent to the display terminal for display;

[0051] If the third-order management signal is captured, a text message is generated, "The student's learning status after class is poor, indicating that he / she has not fully understood or mastered the content learned in class in a timely manner, and needs to strengthen supervision and guidance of after-class learning" and sent to the display terminal for display explanation;

[0052] If the fourth-order management signal is captured, the text "The overall learning status of the student is poor, and he shows certain deficiencies and difficulties before, during and after class. He needs to be paid special attention to, and the supervision and guidance of his learning process should be strengthened comprehensively" is generated and sent to the display terminal for display.

[0053] Compared with the known prior art, the technical solution provided by the present invention has the following beneficial effects:

[0054] 1. The present invention can comprehensively evaluate the learning progress of students by real-time monitoring and analysis of the learning status of students before, during and after class, avoiding the limitation of the traditional education model that only relies on test scores. The multi-dimensional evaluation before, during and after class helps education managers to accurately identify the strengths and weaknesses of students at each learning stage, discover problems in time and provide real-time feedback, thereby achieving better promotion of students' learning improvement;

[0055] 2. The present invention can generate management signals at different levels by comprehensively analyzing the data before, during and after class. Based on these signals, teachers and parents can receive personalized guidance suggestions in a timely manner. This data-driven tutoring method ensures that education managers can grasp the learning status of students in real time and adjust teaching strategies in a timely manner, providing students with more refined and targeted learning support, thereby effectively promoting students' learning effects and overall development; BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0057] Figure 1 It is the overall module block diagram of the present invention. DETAILED DESCRIPTION

[0058] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are 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 creative work are within the scope of protection of the present invention.

[0059] like Figure 1 As shown, a smart education system based on educational big data includes: a data acquisition module, a pre-class analysis module, an in-class analysis module, a post-class analysis module, a comprehensive analysis module, a management execution module, a cloud database and a display terminal.

[0060] It should be pointed out that the data acquisition module, the pre-class analysis module, the in-class analysis module, and the post-class analysis module are respectively connected to the cloud database, the pre-class analysis module, the in-class analysis module, the post-class analysis module, and the management execution module are respectively connected to the comprehensive analysis module, and the management execution module is connected to the display terminal;

[0061] The data collection module is used to collect students' learning data information, including pre-class status information, in-class status information and post-class status information, and send them all to the cloud database;

[0062] The cloud database is used to store pre-class status information, in-class status information, and post-class status information, as well as the recommended preview time for each pre-class monitoring point, the reference body contour image library, the reference eye image, the pre-class status determination table, the in-class status determination table, and the post-class status determination table.

[0063] The pre-class analysis module is used to monitor the students' pre-class status information. The specific monitoring process is as follows:

[0064] By obtaining the students' pre-class login records on the mobile terminal, each pre-class login is marked as a pre-class monitoring point, and the initial login time and offline time of each pre-class monitoring point within a period of time are extracted. The difference between the initial login time and the offline time is calculated to obtain the login duration of each pre-class monitoring point within a period of time, which is recorded as Qds i ;

[0065] Extract the preview time of each pre-class monitoring point within a period of time, and obtain the preview time of each pre-class monitoring point within a period of time, recorded as Qys i ;

[0066] Extract the recommended preview duration of each pre-class monitoring point within a period of time from the cloud database as the reference preview duration Qys of each pre-class monitoring point within a period of time i * ;

[0067] According to the formula: , get the preview preparation value KQ1, where i represents the number of each pre-class monitoring point within a period of time, and i=1, 2, 3…n, n represents the total number of pre-class monitoring point numbers, and a1 and a2 represent the set weight factors respectively;

[0068] It should be pointed out that the mobile terminal refers to the smart devices used by students, such as computers and tablets. Students log in to the smart education platform website through smart devices to conduct learning activities such as course preview, study, and homework submission.

[0069] Extract the total answer value, correct answer value and average answer time of each pre-class monitoring point within a period of time, and obtain the total answer value, correct answer value and average answer time of each pre-class monitoring point within a period of time, and mark them as Qzt i 、Qdt i and Qst i , according to the formula: , and obtain the grasp value KQ2, where a3 and a4 represent the set weight factors respectively.

[0070] It should be pointed out that the total number of questions answered refers to the total number of questions answered by the students, the number of correct answers refers to the number of questions answered correctly by the students, and the average time spent on answering questions refers to the average time spent by the students on each question;

[0071] Extract the video viewing time, video viewing times, courseware browsing time and courseware browsing times of each pre-class monitoring point within a period of time, and mark them as Qsp i 、Qcp i , Qsk i and Qck i , according to the formula: , get the resource usage value KQ3, where QS * It indicates the reference ratio of video viewing time to courseware browsing time. It is used to indicate the ideal ratio between video viewing time and courseware browsing time of students in the learning process. It helps to evaluate whether students reasonably allocate time to different learning resources (videos and courseware). * It represents the reference ratio of video viewing to courseware browsing times. It is used to indicate the ideal ratio between video viewing times and courseware browsing times during the learning process. It helps evaluate whether students have reasonably arranged the frequency of accessing videos and courseware. a5 and a6 represent the set weight factors respectively.

[0072] Based on this, the students’ pre-class status is analyzed. The specific analysis process is as follows:

[0073] By extracting the values ​​of the preview preparation value KQ1, the mastery value KQ2 and the resource utilization value KQ3, normalization is performed according to the formula: , we get the pre-class evaluation index KQP, where e represents a natural constant, η1, η2, η3 represent the weight coefficients of the pre-study preparation value, mastery value, and resource use value, respectively, and η1>η2>η3;

[0074] The pre-class evaluation index of the student is compared and matched with the pre-class status determination table stored in the cloud database, thereby obtaining the pre-class status of the student, and each pre-class evaluation index obtained by the student corresponds to a pre-class status, and a corresponding pre-class status type signal is generated, and the pre-class status type signal includes an excellent pre-class status signal, a good pre-class status signal and a poor pre-class status signal;

[0075] The generated pre-class status type signal is sent to the comprehensive analysis module.

[0076] In a specific embodiment, the present invention analyzes students' pre-class learning activities, including pre-class login records, preview time, answering status, and resource usage (such as video viewing and courseware browsing), so as to evaluate students' pre-class learning preparation and learning attitude, provide teachers and parents with accurate learning status feedback, and provide data support for subsequent teaching and tutoring decisions.

[0077] The in-class analysis module is used to monitor the in-class status information of students. The specific monitoring process is as follows:

[0078] Use the high-definition camera on the mobile terminal to shoot the student's performance video in class, and obtain the student's performance video in class;

[0079] Extract the student's performance video in class over a period of time for analysis, obtain the student's performance image in class over a period of time, and extract the student's body contour image from the performance image, thereby obtaining each body contour image of the student, and match and analyze each body contour image of the student with the reference body contour image library stored in the cloud database, thereby obtaining the number of successfully matched images, and calculate the ratio of the number of successfully matched images to the total number of matches to obtain the body concentration value, which is recorded as KZ1;

[0080] It should be pointed out that the reference body contour image library contains standard body contour images, which are used as a benchmark for evaluation. The standard body contour images include but are not limited to the following: sitting posture images: the standard body contour of a student sitting normally in class, usually with both hands on the table or on the thighs and the body naturally relaxed; hand-raising images: the standard body contour of a student raising his or her hand to ask or answer questions in class; attentive listening images: students concentrate in class, with their eyes fixed on the screen, and remain still for a long time.

[0081] Extracting the trainee's eye images from the performance images, and combining the trainee's eye images in sequence according to the time frame order, thereby obtaining each eye video of the trainee, extracting the duration of the interval between adjacent blinking actions in each eye video, thereby obtaining the duration of each blinking interval in each eye video, comparing and analyzing the duration of each blinking interval in each eye video with the set reference duration, if the duration of a blinking interval in each eye video is greater than the set reference duration, marking a blinking interval in each eye video as an abnormal interval period, thereby obtaining each abnormal interval period, extracting the eye image of each abnormal interval period, and matching it with the reference eye image stored in the cloud database, the reference eye image includes an open eye image and a closed eye image, thereby obtaining each abnormal open eye period and each abnormal closed eye period respectively;

[0082] Integrate each abnormal eye-closing period to obtain the eye-closing duration value Zby;

[0083] Extract the number of eyeball rotations in each abnormal eye-opening period, and compare and analyze it with the set reference number of eyeball rotations. If the number of eyeball rotations in a certain abnormal eye-opening period is less than the set reference number of eyeball rotations, then the abnormal eye-opening period is determined as a daze period, and the daze periods determined as such are integrated to obtain the daze duration value Zfd;

[0084] According to the formula: , and get the eye focus value KZ2, where Zzs represents the total duration of each eye video of the student, and b1 and b2 represent the set weight factors respectively.

[0085] Based on this, the students' in-class status is analyzed. The specific analysis process is as follows:

[0086] By extracting the values ​​of body focus value KZ1 and eye focus value KZ2, normalization is performed according to the formula: , and the in-class evaluation index KZP is obtained, where u represents a natural constant, η4 ​​and η5 represent the weight coefficients of body concentration value and eye concentration value respectively, and η5>η4.

[0087] The in-class evaluation index of the student is compared and matched with the in-class status determination table stored in the cloud database, thereby obtaining the in-class status of the student, and each in-class evaluation index obtained by the student corresponds to an in-class status, and a corresponding in-class status type signal is generated, and the in-class status type signal includes an excellent in-class status signal, a good in-class status signal and a poor in-class status signal;

[0088] The generated in-class status type signal is sent to the comprehensive analysis module.

[0089] In a specific embodiment, the present invention comprehensively evaluates the students' in-class status by analyzing their performance in class, including body concentration and eye concentration, thereby comprehensively evaluating the students' concentration in class, helping teachers identify whether students have inattention, and providing data support for subsequent teaching and tutoring decisions.

[0090] The post-class analysis module is used to monitor the students' post-class status information. The specific monitoring process is as follows:

[0091] By obtaining the students' after-class login records on the mobile terminal, each after-class login is marked as a after-class monitoring point;

[0092] Extract the total number of tasks and the number of completed tasks at each after-school monitoring point within a period of time, and calculate the ratio of the number of completed tasks to the total number of tasks to obtain the completion value, which is recorded as KH1 j , where j represents the number of each after-class monitoring point within a period of time, and j=1, 2, 3…m, and m represents the total number of after-class monitoring point numbers.

[0093] It should be pointed out that the total number of tasks refers to the sum of all review tasks that students need to complete at a certain post-class monitoring point, and the number of completed tasks refers to the number of review tasks that students have actually completed at a certain post-class monitoring point;

[0094] Extract the average test score of each after-class monitoring point within a period of time as the test value of each after-class monitoring point within a period of time, recorded as KH2 j ;

[0095] It should be noted that the average test score refers to the average of the scores obtained by students in the tests taken at a post-class monitoring point;

[0096] Extract the login time of each after-school monitoring point within a period of time, and calculate the difference between it and the set login time to obtain the habit value, which is recorded as KH3 j The specific method for setting the login time is as follows: extract the login time of each after-class monitoring point within the historical unit time period of the students, and arrange the login time of each after-class monitoring point within the historical unit time period of the students in chronological order to obtain a login time sequence, and extract the login time corresponding to the majority from the login time sequence as the set login time. The login time corresponding to the majority represents the time point when the students log in most frequently within the historical unit time period.

[0097] Based on this, we analyze the students’ after-class status. The specific analysis process is as follows:

[0098] Complete the value KH1 by extracting j , test value KH2 j and custom value KH3 j , according to the formula: , and get the post-class evaluation index KHP, where KH1 j-1 、KH2 j-1 、KH3 j-1 They represent the completion value, test value and habit value of the j-1th post-class monitoring point, respectively, KH1 * 、KH2 * 、KH3 * They respectively represent reference completion value, reference test value and reference habit value, c1, c2, c3, c4, c5 and c6 respectively represent the set weight factors, η6, η7, η8 respectively represent the weight coefficients of completion fluctuation degree, test fluctuation degree and habit fluctuation degree, and η6>η7>η8.

[0099] The after-class evaluation index of the students is compared and matched with the after-class status determination table stored in the cloud database, thereby obtaining the after-class status of the students, and each after-class evaluation index obtained by the students corresponds to a after-class status, and a corresponding after-class status type signal is generated, and the after-class status type signal includes an excellent after-class status signal, a good after-class status signal and a poor after-class status signal;

[0100] The generated after-class status type signal is sent to the comprehensive analysis module.

[0101] In a specific embodiment, the present invention comprehensively evaluates the students' after-class status by analyzing their after-class performance, including the number of completed tasks, average test scores, and login time, thereby comprehensively evaluating the students' after-class review situation, providing teachers and parents with accurate after-class review status feedback, and providing data support for subsequent teaching and tutoring decisions.

[0102] The comprehensive analysis module is used to receive the pre-class state type signal, the in-class state type signal and the post-class state type signal, and perform comprehensive analysis and processing. The specific process is as follows:

[0103] According to the pre-class state type signal, a set R is established, and the excellent pre-class state signal is marked as element r1, the good pre-class state signal is marked as element r2, and the poor pre-class state signal is marked as element r3, and element r1∈set R, element r2∈set R, and element r3∈set R;

[0104] According to the class status type signal, a set Y is established, the excellent class status signal is marked as element y1, the good class status signal is marked as element y2, the poor class status signal is marked as element y3, and the element y1∈set Y, the element y2∈set Y, and the element y3∈set Y;

[0105] According to the after-class status type signal, T is established, the excellent after-class status signal is marked as element t1, the good after-class status signal is marked as element t2, and the poor after-class status signal is marked as element t3, and element t1∈set T, element t2∈set T, element t3∈set T;

[0106] The sets R, Y and T are unioned. If R∪Y∪T={r3, y1, t1} or {r3, y2, t1} or {r3, y1, t2}, a first-order management signal is generated. If R∪Y∪T={r1, y3, t1} or {r2, y3, t1} or {r1, y3, t2}, a second-order management signal is generated. If R∪Y∪T={r1, y1, t3} or {r2, y1, t3} or {r1, y2, t3}, a third-order management signal is generated. If R∪Y∪T={r2, y2, t2} or {r3, y3, t3}, a fourth-order management signal is generated.

[0107] The management type signal is sent to the management execution module, wherein the management type signal includes a first-order management signal, a second-order management signal, a third-order management signal and a fourth-order management signal.

[0108] The management execution module is used to receive the management type signal and perform management execution analysis and processing. The specific process is as follows:

[0109] If the captured management type signal is a first-order management signal, a text message "The student's learning state before class is poor, showing low readiness and participation, and needs to strengthen supervision and guidance before class learning" is generated and sent to the display terminal for display explanation;

[0110] If the captured management type signal is a second-order management signal, a text message "The student's learning status in class is poor, showing a lack of concentration, and needs to strengthen supervision and guidance of learning in class" is generated and sent to the display terminal for display;

[0111] If the captured management type signal is a third-order management signal, a text message "The student's learning status after class is poor, indicating that he / she has not fully understood or mastered the content learned in class in a timely manner, and needs to strengthen supervision and guidance of after-class learning" is generated and sent to the display terminal for display explanation;

[0112] If the captured management type signal is a fourth-order management signal, a text message is generated, "The overall learning status of the student is poor, and he has shown certain deficiencies and difficulties before, during and after class. We need to pay special attention to him and strengthen the supervision and guidance of his learning process in an all-round way" and sent to the display terminal for display;

[0113] It should be pointed out that the display terminal is a mobile terminal of parents and teachers, which is used to display the feedback content of the management module in real time so that they can understand the learning status of the students in time and take corresponding intervention measures.

[0114] A computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements any of the above-mentioned systems;

[0115] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A smart education system based on educational big data, comprising: The data collection module is used to collect the learning data information of the students, and is characterized by further comprising: The pre-class analysis module is used to monitor and calculate the students' preparation value, mastery value and resource usage value, thereby analyzing the students' pre-class status and obtaining the pre-class evaluation index to generate a pre-class status type signal and send it to the comprehensive analysis module. The specific process is as follows: By obtaining the students' pre-class login records, each pre-class login is marked as a pre-class monitoring point; Extract the initial login time and offline time of each pre-class monitoring point within a period of time, and calculate the difference between the initial login time and the offline time to obtain the login time Qds of each pre-class monitoring point within a period of time i ; Extract the preview time Qys of each pre-class monitoring point within a period of time i ; According to the formula: , get the preview preparation value KQ1, where i represents the number of each pre-class monitoring point within a period of time, and i=1, 2, 3…n, n represents the total number of pre-class monitoring point numbers, Qys i * represents the reference preview duration, a1 and a2 represent the set weight factors respectively; Extract the total answer value Qzt of each pre-class monitoring point within a period of time i 、Correct answer value Qdt i and the average time to answer the questions Qst i , according to the formula: , and obtain the master value KQ2, where a3 and a4 represent the set weight factors respectively; Extract the video viewing time Qsp of each pre-class monitoring point within a period of time i 、Video views Qcp i , Courseware browsing time Qsk i And the number of courseware views Qck i , according to the formula: , get the resource usage value KQ3, where QS * Indicates the reference ratio of video viewing time to courseware browsing time, QC * It represents the reference ratio of video viewing to courseware browsing times, and a5 and a6 represent the set weight factors respectively; By extracting the values ​​of the preview preparation value KQ1, the mastery value KQ2 and the resource utilization value KQ3, normalization is performed according to the formula: , we get the pre-class evaluation index KQP, where e represents a natural constant, η1, η2, η3 represent the weight coefficients of the pre-study preparation value, mastery value, and resource use value, respectively, and η1>η2>η3; The student's pre-class evaluation index is compared and matched with the pre-class state determination table, thereby obtaining the student's pre-class state and generating a corresponding pre-class state type signal, and the pre-class state type signal includes an excellent pre-class state signal, a good pre-class state signal and a poor pre-class state signal; The in-class analysis module is used to monitor and calculate the students' body concentration value and eye concentration value, thereby analyzing the students' in-class status and obtaining the in-class evaluation index, so as to generate an in-class status type signal and send it to the comprehensive analysis module; The post-class analysis module is used to monitor and calculate the completion value, test value and habit value of the students, thereby analyzing the students' post-class status and obtaining the post-class evaluation index to generate a post-class status type signal and send it to the comprehensive analysis module. The specific process is as follows: By obtaining the students' after-class login records, each after-class login is marked as a after-class monitoring point; Extract the total number of tasks and the number of completed tasks at each after-school monitoring point within a period of time, and calculate the ratio of the number of completed tasks to the total number of tasks to obtain the completion value KH1 j , where j represents the number of each after-school monitoring point within a period of time, and j=1, 2, 3…m, and m represents the total number of after-school monitoring point numbers; Extract the average test score of each after-class monitoring point within a period of time as the test value KH2 of each after-class monitoring point within a period of time j ; Extract the login time of each after-school monitoring point within a period of time, and calculate the difference between it and the set login time to obtain the habit value KH3 j , wherein setting the login time is specifically as follows: extracting the login time of each after-class monitoring point in the student's historical unit time period, and arranging them in chronological order to obtain a login time sequence, and extracting the login time corresponding to the mode from the login time sequence as the set login time; Complete the value KH1 by extracting j , test value KH2 j and custom value KH3 j , according to the formula: , and get the post-class evaluation index KHP, where KH1 j-1 、KH2 j-1 、KH3 j-1 They represent the completion value, test value and habit value of the j-1th post-class monitoring point, respectively, KH1 * 、KH2 * 、KH3 * denote reference completion value, reference test value and reference habit value respectively, c1, c2, c3, c4, c5 and c6 denote set weight factors respectively, η6, η7 and η8 denote weight coefficients of completion fluctuation degree, test fluctuation degree and habit fluctuation degree respectively, and η6>η7>η8; The after-class evaluation index of the students is compared and matched with the after-class status determination table, thereby obtaining the after-class status of the students, and generating a corresponding after-class status type signal, and the after-class status type signal includes an excellent after-class status signal, a good after-class status signal and a poor after-class status signal; A comprehensive analysis module is used to receive the pre-class status type signal, the in-class status type signal and the post-class status type signal, and perform comprehensive analysis and processing thereon to generate a management type signal and send it to the management execution module; The management execution module is used to receive a management type signal to generate text words.

2. According to claim 1, a smart education system based on educational big data is characterized in that: The specific process of calculating the student's body concentration value is as follows: Extract the students' performance videos in class over a period of time and analyze them to obtain images of the students' performance in class over a period of time; The student's body contour image is extracted from the performance image, thereby obtaining each body contour image of the student, and each body contour image of the student is matched and analyzed with the reference body contour image library to obtain the number of successfully matched images, and the ratio of the number of successfully matched images to the total number of matches is calculated to obtain the body focus value KZ1.

3. The smart education system based on educational big data according to claim 1 is characterized in that: The specific process of solving the students' eye focus value is as follows: Extracting the trainee's eye images from the performance images, and combining the trainee's eye images in sequence according to the time frame order, thereby obtaining each eye video of the trainee, extracting the duration of the interval between adjacent blinking actions in each eye video, thereby obtaining the duration of each blinking interval in each eye video, comparing and analyzing the duration of each blinking interval in each eye video with the set reference duration, if the duration of a blinking interval in each eye video is greater than the set reference duration, marking a blinking interval in each eye video as an abnormal interval period, thereby obtaining each abnormal interval period, extracting the eye image of each abnormal interval period, and matching it with the reference eye image, thereby obtaining each abnormal eye-opening period and each abnormal eye-closing period respectively; Integrate each abnormal eye-closing period to obtain the eye-closing duration value Zby; Extract the number of eyeball rotations in each abnormal eye-opening period, and compare and analyze it with the set reference number of eyeball rotations. If the number of eyeball rotations in a certain abnormal eye-opening period is less than the set reference number of eyeball rotations, then the abnormal eye-opening period is determined as a daze period, and the daze periods determined as such are integrated to obtain the daze duration value Zfd; According to the formula: , and get the eye focus value KZ2, where Zzs represents the total duration of each eye video of the student, and b1 and b2 represent the set weight factors respectively.

4. The smart education system based on educational big data according to claim 1 is characterized in that: Analyze the students' in-class status. The specific analysis process is as follows: By extracting the values ​​of body focus value KZ1 and eye focus value KZ2, normalization is performed according to the formula: , we get the in-class evaluation index KZP, where u represents a natural constant, η4 ​​and η5 represent the weight coefficients of body concentration value and eye concentration value, respectively, and η5>η4; The student's in-class evaluation index is compared and matched with the in-class status determination table to obtain the student's in-class status and generate a corresponding in-class status type signal, and the in-class status type signal includes an excellent in-class status signal, a good in-class status signal and a poor in-class status signal.

5. The smart education system based on educational big data according to claim 1 is characterized in that: The specific analysis process for generating management type signals is as follows: According to the pre-class state type signal, a set R is established, and the excellent pre-class state signal is marked as element r1, the good pre-class state signal is marked as element r2, and the poor pre-class state signal is marked as element r3, and element r1∈set R, element r2∈set R, and element r3∈set R; According to the class status type signal, a set Y is established, the excellent class status signal is marked as element y1, the good class status signal is marked as element y2, the poor class status signal is marked as element y3, and the element y1∈set Y, the element y2∈set Y, and the element y3∈set Y; According to the after-class status type signal, T is established, the excellent after-class status signal is marked as element t1, the good after-class status signal is marked as element t2, and the poor after-class status signal is marked as element t3, and element t1∈set T, element t2∈set T, element t3∈set T; The sets R, Y and T are unioned. If R∪Y∪T={r3, y1, t1} or {r3, y2, t1} or {r3, y1, t2}, a first-order management signal is generated. If R∪Y∪T={r1, y3, t1} or {r2, y3, t1} or {r1, y3, t2}, a second-order management signal is generated. If R∪Y∪T={r1, y1, t3} or {r2, y1, t3} or {r1, y2, t3}, a third-order management signal is generated. If R∪Y∪T={r2, y2, t2} or {r3, y3, t3}, a fourth-order management signal is generated.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the system according to any one of claims 1 to 5 is implemented.

Images