Teaching and research training integrated intelligent education management system

Abstract

The invention belongs to the technical field of wisdom education, and particularly relates to a teaching, research and training integrated wisdom education management system, which comprises a teaching management module used for acquiring the learning progress of students and the teaching progress of teachers, and dynamically generating a teaching plan by adopting an artificial intelligence algorithm based on the learning progress of the students and the teaching progress of the teachers; the teaching and research management module is used for an online precise teaching and research teaching platform based on cloud computing to obtain an optimized teaching plan; the training management module is used for comprehensively analyzing teaching and teaching and research result data of teachers by adopting a multi-modal data fusion technology, identifying advantages and deficiencies of different teachers and automatically generating a personalized training course system by applying an intelligent algorithm; and the adjusting module is used for constructing all-around teaching effect evaluation ecology, accurately judging teaching intervention effect and feeding back the teaching intervention effect. Through modular design of teaching management, teaching and research management, training management and adjustment modules, intelligent education management integrating teaching, research and training is realized.

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent education, and particularly relates to an integrated intelligent education management system for teaching, research and training. Background Art

[0002] With the development of society, the education of students has become a key concern of society. With the wide application of information technology in the field of education, many schools have adopted education management systems for education management, and the education management system has improved the efficiency of teaching management to a certain extent.

[0003] Currently, most of the existing education management systems have the problems of single function and isolated modules, and often cannot integrate teaching, research and training together. This decentralized management mode makes it difficult to achieve the synergistic effect of mutual promotion among teaching, research and training, resulting in that teaching plans often cannot be adaptively adjusted based on teaching, research and training, restricting the overall improvement of education quality and the in-depth development of education innovation. Summary of the Invention

[0004] In order to overcome the above-mentioned defects of the prior art, the present invention provides an integrated intelligent education management system for teaching, research and training, and realizes the intelligent education management of teaching, research and training through the modular design of teaching management, research management, training management and adjustment modules.

[0005] The present invention provides the following technical solutions: An integrated intelligent education management system for teaching, research and training, comprising: A teaching management module, configured to obtain the learning progress of students and the teaching progress of teachers, and dynamically generate a teaching plan based on the learning progress of students and the teaching progress of teachers by using an artificial intelligence algorithm; A research management module, configured to support the teacher teams in multiple regions to conduct remote collaboration to study the teaching plan and optimize the teaching plan based on the online precise teaching and research platform based on cloud computing to obtain an optimized teaching plan; A training management module, configured to adopt a multi-modal data fusion technology to comprehensively analyze the teaching and research achievement data of teachers, identify the advantages and disadvantages of different teachers, automatically generate a personalized training curriculum system by using an intelligent algorithm, and during the training, establish an evaluation index system for training effects, determine the training effects, and adaptively adjust the optimized teaching plan according to the training effects; An adjustment module, configured to construct an all-round teaching effect evaluation ecosystem, collect various types of teaching feedback data, analyze the teaching feedback by using a machine learning method, accurately determine the effectiveness of teaching intervention and give feedback.

[0006] Preferably, in the teaching management module, an artificial intelligence algorithm is used to dynamically generate a teaching plan based on the learning progress of students and the teaching progress of teachers. The generation process is as follows: Set the student learning progress index as S and calculate S. The calculation formula is as follows: ; In the formula, n represents the number of different dimensions for measuring the learning progress of students, i represents the index variable of n, represents the weight coefficient of each dimension of the learning progress, and its value range is between 0 and 1, represents the quantified value of each dimension of the learning progress; Set the teacher teaching progress index as J and calculate J. The calculation formula is as follows: ; In the formula, m represents the number of different dimensions for measuring the teaching progress of teachers, represents the index variable of m, represents the weight coefficient of each dimension of the teaching progress, represents the quantified value of each dimension of the teaching progress; Through the artificial intelligence algorithm module receive S and J, clean S and J, normalize S and J to the same dimension, and dynamically generate the teaching plan P according to the difference between S and J. The generation formula is as follows: ; In the formula, GPG represents the priority of the teaching goal, CPC represents the complexity of the course content, , , and are the weight coefficients corresponding to S, J, GPG, and CPC respectively; Map the teaching plan P to the initial time allocation sequence , and its time allocation rule is defined by the following formula: ; In the formula, VVP represents the total number of teaching links, represents the complexity score of the vvp-th link in the teaching plan P, is the basic time weight of the vvp-th link; Analyze the teaching behavior of teachers and the classroom responses of students based on the object detection algorithm of deep learning.

[0007] Preferably, regarding the analysis of the teaching behavior of teachers and the classroom responses of students based on the object detection algorithm of deep learning, the analysis process is as follows: Obtain the real-time image data of the camera in the classroom, and set the image sequence at time t as , grayscale and denoise the acquired images to obtain a preprocessed image sequence ; Based on the image sequence , identify the types of the teacher's actions, the categories of expressions, as well as the attention level, emotional categories, and interaction behavior categories of the students; Calculate the activity and diversity quantification indexes of the teacher's actions at time t , and the calculation formula is as follows: ; In the formula, represents the number of times the v-th action appears within time t, represents the weight corresponding to the v-th action, and the number of action categories is C; Set the number of expression categories as E, and calculate the positive degree of the teacher's expression at time t , and the calculation formula is as follows: ; In the formula, represents the duration of the z-th expression within time t, represents the positive degree weight corresponding to the z-th expression; Integrate the teacher's action and expression information, and calculate the comprehensive index of the teacher's teaching behavior at time t , and the calculation formula is as follows: ; In the formula, represents the weight coefficient of the action and expression in the comprehensive evaluation; Calculate the overall attention proportion of the students at time t , and the calculation formula is as follows: ; In the formula, R represents the total number of students, represents the attention level of the r-th student at time t; Set the number of emotional categories as M, and calculate the proportion of students with different emotions at time t , and the calculation formula is as follows: ; In the formula, represents the indicator function, which judges whether the emotion of the r-th student at time t is the q-th emotion. If so, takes the value of 1, otherwise 0; Set the number of student interaction behavior categories as NF, and calculate the student interaction frequency at time t , and the calculation formula is as follows: ; ; Wherein, represents the number of occurrences of the l-th interaction behavior within time t, is the weight corresponding to the l-th interaction behavior, represents an indicator function. If the interaction behavior bSS belongs to the l-th interaction behavior, then , otherwise . By traversing each interaction behavior bSS in, and by judging its category and accumulating the count, the number of occurrences of the l-th interaction behavior within time t is finally obtained ; The comprehensive student overall concentration ratio , the ratio of students with different emotions and the student interaction frequency are combined to calculate the comprehensive index of students' classroom responses at time t , and the calculation formula is as follows: ; Wherein, represents the weight coefficient corresponding to the comprehensive student overall concentration ratio , represents the weight coefficient corresponding to the ratio of students with different emotions , and both have a value range from 0 to 1, and 1, represents the weight corresponding to the q-th emotion; Calculate the optimized time allocation sequence , and the calculation formula is as follows: ; Wherein, represents the adjustment coefficient, , respectively represent the thresholds of the comprehensive index of students' classroom responses , the comprehensive index of teachers' teaching behaviors , represents the adjustment vector of the time allocation for each teaching link; Feed the optimized time allocation sequence back to the artificial intelligence algorithm module to update the teaching plan P.

[0008] Preferably, in the teaching and research management module, the process of obtaining the optimized teaching plan is as follows: Set the comprehensive evaluation index EZ to measure the effectiveness of the entire online precise teaching and research platform, and calculate EZ. The calculation formula is as follows: ; Wherein, $I_a$ represents the interaction index, which is used to measure the frequency of interaction between teacher teams on the platform; $C_a$ represents the collaboration degree, which is used to reflect the closeness of collaboration between teacher teams; $O_a$ represents the optimization level, which is used to measure the degree of optimization and improvement of the teaching plan in terms of quality, applicability, etc. after remote collaboration among teacher teams. 、 and respectively represent the weight coefficients corresponding to $I_a$, $C_a$ and $O_a$; The interaction index $I_a$ is calculated, and the calculation formula is as follows: ; Wherein, represents the number of meaningful interactions that actually occur between teacher teams on the platform within a certain time period, represents the maximum number of interactions that occur within this time period, represents the weight corresponding to each interaction, represents the quality score of the $i_r$-th interaction; The collaboration degree $C_a$ is calculated, and the calculation formula is as follows: ; Wherein, represents the number of collaboration projects completed by teacher teams within time $t$, represents the total number of projects within time $t$, represents the number of teacher teams participating in collaboration, represents the quantitative value of the workload of the $j_q$-th teacher team investing resources in each collaboration project; The optimization level $O_a$ is calculated, and the calculation formula is as follows: ; Wherein, represents the number of teaching plans to be evaluated after collaboration, represents the difference in quality evaluation indicators of the $r_k$-th teaching plan before and after collaboration optimization; Through big data analysis technology, in-depth analysis is carried out on the education field to provide suggestions on research topic directions for teachers.

[0009] Preferably, for the in-depth analysis of the education field through big data analysis technology to provide suggestions on research topic directions for teachers, the analysis process is as follows: Set the recommended value evaluation index as $V_H$, which is used to measure the quality and practicality of the research topic direction suggestions provided for teachers. The calculation formula for $V_H$ is as follows: ; Wherein, $H_d$ represents the hot topic relevance, $D_d$ represents the difficulty overcoming ability, and $Y_d$ represents the trend consistency. , and represent the weight coefficients corresponding to Hd, Dd, and Yd respectively; Calculate the hotspot relevance Hd, and the calculation formula is as follows: ; In the formula, represents the number of hotspot issues in the education field, represents the degree of attention score of the mo-th hotspot issue, represents the correlation coefficient between the proposed research direction of the topic and the mo-th hotspot issue; Calculate the difficulty-solving ability Dd, and the calculation formula is as follows: ; In the formula, represents the number of difficult and painful issues in the statistical teaching practice, represents the influence weight of the md-th difficult and painful issue in the actual teaching, represents the effectiveness score of the research direction of the topic for solving the md-th difficult and painful issue; Calculate the trend consistency Yd, and the calculation formula is as follows: ; In the formula, represents the number of dimensions in terms of the discipline development trend, represents the importance coefficient of the mp-th discipline development trend dimension, represents the degree of compliance of the proposed research direction of the topic with the mp-th discipline development trend dimension; Set the comprehensive influence factor as XQ, which is used to measure the comprehensive effect of the comprehensive evaluation index EZ and the recommended value evaluation index VH on the optimization of the teaching plan P. Calculate the comprehensive influence factor XQ, and the calculation formula is as follows: ; In the formula, and represent the weight coefficients corresponding to EZ and VH respectively, and ; Optimize each element of the teaching plan P based on the comprehensive influence factor XQ.

[0010] Preferably, regarding the optimization of each element of the teaching plan P based on the comprehensive influence factor XQ, the optimization process is as follows: Set the teaching plan P to consist of four elements: teaching goal GP, teaching content CP, teaching method MP, and teaching evaluation EP, that is , and the optimized teaching plan is set as , The four elements are represented as ; For the optimized teaching objectives perform calculations, and the calculation formula is as follows: ; In the formula, represents the weight coefficient of the optimized teaching objectives, represents the potential value of the optimized teaching objectives; For the optimized teaching content perform calculations, and the calculation formula is as follows: ; In the formula, represents the weight coefficient of the optimized teaching content, represents the potential value of the optimized teaching objectives; For the optimized teaching methods perform calculations, and the calculation formula is as follows: ; In the formula, represents the weight coefficient of the optimized teaching methods, represents the potential value of the optimized teaching methods; For the optimized teaching evaluation perform calculations, and the calculation formula is as follows: ; In the formula, represents the weight coefficient of the optimized teaching evaluation, represents the potential value of the optimized teaching evaluation.

[0011] Preferably, in the training management module, an intelligent algorithm is used to automatically generate a personalized training course system, and the generation process is as follows: Define to represent the evaluation result of the training effect. The training effect evaluation index system includes the following indicators: teacher ability improvement, teaching effect improvement, student learning effect improvement, training process quality, training resource utilization efficiency, training satisfaction, external evaluation; After numerically quantifying the above indicators, normalize the quantified values, and take the average of the normalized values as the value of; Set two thresholds for the evaluation result of the training effect, namely the high threshold and the low threshold ; When it indicates that the training effect is excellent, and adjust the teaching plan : increase extracurricular knowledge, increase group interaction, and accelerate the teaching schedule arrangement; When it indicates that the training is excellent and medium, and the teaching plan is adjusted as follows: adding a special consolidation practice session and slowing down the teaching progress arrangement; When it indicates that the training is excellent and poor, and the teaching plan is adjusted as follows: re-planning the teaching order of knowledge points, re-replacing the teaching plan, and slowing down the teaching progress arrangement.

[0012] Preferably, in the adjustment module, the teaching intervention effect is accurately determined and fed back, and the determination process is as follows: Set the student learning effect evaluation function as Esk, and calculate Esk. The calculation method is as follows: ; In the formula, Njs represents the number of student teaching characteristics. Student teaching characteristics include: test score, homework completion score, class participation score, and learning time score. represents the weight of the rth student learning characteristic, represents the conversion function of the rth student learning characteristic data, represents the index variable; Set the teacher teaching effect evaluation function as Esj, and calculate Esj. The calculation method is as follows: ; In the formula, Macb represents the number of teacher teaching characteristics. Teacher teaching characteristics include: teaching method diversity score, teaching content integrity score, teaching progress compliance score, and teacher-student interaction frequency score. represents the weight of the nbcqth teacher teaching characteristic, IQc represents the index variable, represents the conversion function of the nbcqth teacher teaching characteristic data; Set the research result influence evaluation function as Ejy, and calculate Ejy. The calculation method is as follows: ; In the formula, Maut represents the number of research result characteristics. Research result characteristics include: research result application value score, research result innovation score, and research result influence degree score on the teaching plan. IQut represents the index variable, represents the weight of the nunfth research result characteristic, represents the conversion function of the nunfth research result characteristic data; Set the teaching intervention effect evaluation function as Effca, and calculate Effca. The calculation formula is as follows: ; Wherein, and and represent the weight coefficients corresponding to Esk, Esj, and Ejy respectively, and ; Set the evaluation threshold to ; When , it is determined that the teaching intervention effect is good, and teaching is carried out according to the adjusted teaching plan ; When , it indicates that the teaching intervention effect is poor, a reminder is sent, and the teacher is retrained and the teaching plan is adjusted again.

[0013] Technical effects and advantages of the present invention: Through the modular design of the teaching management, teaching and research management, training management, and adjustment modules, the present invention realizes the intelligent education management of the integration of teaching, research, and training. The division of labor between the modules is clear, the functions are clear, it can dynamically generate and optimize the teaching plan, support the remote collaboration of teacher teams in multiple regions, comprehensively analyze the teaching and teaching research achievement data of teachers, automatically generate a personalized training course system, and accurately determine the teaching intervention effect through the all-round teaching effect evaluation ecosystem.

[0014] The multi-modal data fusion technology is adopted in the present invention to comprehensively analyze the teaching and teaching research achievement data of teachers and identify the advantages and disadvantages of different teachers. Dynamically generate the teaching plan and personalized training course system through artificial intelligence algorithms, and analyze the teaching behaviors of teachers and the classroom responses of students using the object detection algorithm of deep learning to ensure the scientificity and adaptability of the teaching plan.

[0015] The present invention constructs an all-round teaching effect evaluation ecosystem, collects various types of teaching feedback data, analyzes the teaching feedback using machine learning methods, and accurately determines the teaching intervention effect. Through the comprehensive evaluation index system, comprehensively evaluate the teaching effect and training effect of teachers, provide a scientific basis for the continuous optimization of the teaching plan, and ensure the continuous improvement of teaching quality and training effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 is the principle block diagram of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0017] Figure 1 is the best embodiment of the present invention. The present invention will be further described below in conjunction with Figure 1

[0018] As Figure 1 shown, the intelligent education management system for the integration of teaching, research, and training includes: ​The teaching management module is used to obtain the learning progress of students and the teaching progress of teachers, and dynamically generate a teaching plan based on the learning progress of students and the teaching progress of teachers using artificial intelligence algorithms; The teaching and research management module is used for an online precise teaching and research teaching platform based on cloud computing, supporting teacher teams in multiple regions to conduct remote collaboration to research the teaching plan and optimize the teaching plan to obtain an optimized teaching plan; The training management module is used to adopt multi-modal data fusion technology, comprehensively analyze the teaching and teaching research achievement data of teachers, identify the strengths and weaknesses of different teachers, automatically generate a personalized training curriculum system using intelligent algorithms, and during the training, establish an evaluation index system for training effects, determine the training effects, and make adaptive adjustments to the optimized teaching plan according to the training effects; The adjustment module is used to build an all-round teaching effect evaluation ecosystem, collect various types of teaching feedback data, analyze the teaching feedback using machine learning methods, accurately determine the effectiveness of teaching interventions and provide feedback.

[0019] Among them, the online precise teaching and research teaching platform can use existing precise teaching and research teaching platforms / precise teaching platforms / precise teaching cloud platforms / smart education cloud platforms, and use a suitable platform according to needs.

[0020] In the teaching management module, a teaching plan is dynamically generated using artificial intelligence algorithms based on the learning progress of students and the teaching progress of teachers. The generation process is as follows: Set the student learning progress index as S and calculate S. The calculation formula is as follows: ; In the formula, n represents the number of different dimensions for measuring the learning progress of students, i represents the index variable of n, represents the weight coefficient of each dimension of the learning progress, and the value range is between 0 and 1, represents the quantitative value of each dimension of the learning progress; The different dimensions of the student learning progress are set according to needs. For example, they can be set as: Exam score: Evaluate the student's mastery of knowledge points through exam scores; Homework completion score: including the accuracy, completion speed and quality of homework completion; Classroom exercise score: The results of instant exercises and quizzes in class; Completion of project homework: The performance of students in group projects or individual projects; Classroom participation: The frequency of students' speaking in class, interaction and enthusiasm for participating in discussions; Learning time: The time students spend on learning every day or week, including in-class and out-of-class; Learning frequency: The regularity of students' learning, such as whether they study every day; Use of learning methods: Whether students use effective learning methods (such as review, summary, preview, etc.); Depth of knowledge mastery: Students' understanding and application ability of core knowledge points; Problem-solving ability: Students' ability to solve problems when facing new problems; Self-directed learning ability: Students' ability to study independently and explore new knowledge; Time management ability: Students' ability to reasonably arrange their study time.

[0021] The quantitative values in this embodiment can adopt: Expert scoring method: By inviting domain experts (such as education experts, subject teachers, etc.) to score each dimension, so as to determine the quantitative values of each dimension. For example, classroom participation can be divided into "actively participate (90 - 100 points)", "generally participate (60 - 89 points)" and "do not participate (0 - 59 points)".

[0022] Fuzzy comprehensive evaluation method: An evaluation method based on fuzzy mathematics, suitable for dealing with evaluation indicators with strong fuzziness. Set membership functions for each evaluation indicator and map the indicator values to the interval [0, 1].

[0023] Analytic hierarchy process: By constructing a hierarchical structure model, decompose complex problems into multiple levels for analysis. For example, the goal layer is "learning progress evaluation", the criterion layer is "learning achievements", "learning behaviors", etc., and the index layer is specific dimensions (such as exam scores, classroom participation, etc.).

[0024] In addition, when calculating each formula in this embodiment, dimensionless processing can be performed because the purpose of each known formula is to calculate reference values, and the specific analysis process has no requirements for dimensions. In addition, time calculation uses minutes.

[0025] Set the teacher's teaching progress index as J, and calculate J. The calculation formula is as follows: ; In the formula, m represents the number of different dimensions for measuring the teacher's teaching progress, represents the index variable of m, represents the weight coefficient of each dimension of the teaching progress, represents the quantitative value of each dimension of the teaching progress; The different dimensions for measuring the teacher's teaching progress are set according to requirements. For example, they can be set as: Completion ratio of course content: Measure whether the teacher has completed the established course content according to the teaching plan; Achievement of teaching objectives: Evaluate whether the teacher has achieved the predetermined teaching objectives, such as knowledge transfer, skill cultivation, etc.; Coverage of knowledge points: Check whether the teacher has comprehensively covered the knowledge points required in the curriculum syllabus; Diversity of teaching methods: Evaluate whether the teacher has adopted a variety of teaching methods, such as lectures, discussions, experiments, etc.; Utilization rate of teaching resources: Measure whether the teacher has made full use of various teaching resources, such as textbooks, multimedia, network resources, etc.; Frequency of classroom interaction: Evaluate the interaction between the teacher and students, such as questions, discussions, feedback, etc.; Conformity of teaching progress: Measure whether the teacher's teaching progress is consistent with the teaching plan; Rationality of teaching rhythm: Evaluate whether the teacher reasonably adjusts the teaching rhythm according to the students' acceptance ability; Smoothness of connection between teaching links: Check whether the transition between teaching links is natural and smooth.

[0026] Through the artificial intelligence algorithm module Receive S and J, clean S and J, normalize S and J to the same dimension, and dynamically generate the teaching plan P according to the differences between S and J. The generation formula is as follows: ; In the formula, GPG represents the priority of the teaching objective (which can be set in advance), CPC represents the complexity of the course content (which can be manually evaluated), , , and are the weight coefficients corresponding to S, J, GPG, and CPC respectively; Map the teaching plan P to the initial time allocation sequence , and its time allocation rule is defined by the following formula: ; In the formula, VVP represents the total number of teaching links, represents the complexity score of the vvp-th link in the teaching plan P, is the basic time weight of the vvp-th link; Analyze the teacher's teaching behavior and students' classroom reactions based on the object detection algorithm of deep learning. The analysis process is as follows: Obtain the real-time image data of the camera in the classroom. The image sequence at time t is , perform grayscale and noise reduction processing on the obtained images, and obtain the preprocessed image sequence ; Based on the image sequence , identify and obtain the types of teachers' actions, categories of expressions, as well as the attention levels, emotional categories, and interaction behavior categories of students; identification can be carried out based on well-known image recognition models; Quantify the activity and diversity metrics of teachers' actions at time t Perform calculations, and the calculation formula is as follows: ; In the formula, represents the number of times the v-th type of action appears within time t, represents the weight corresponding to the v-th type of action, and the number of action categories is C; Set the number of expression categories to E, and calculate the positivity level of the teacher's expression at time t , and the calculation formula is as follows: ; In the formula, represents the duration of the z-th type of expression within time t, represents the positivity weight corresponding to the z-th type of expression; Integrate the information of teachers' actions and expressions, and calculate the comprehensive index of teachers' teaching behaviors at time t , and the calculation formula is as follows: ; In the formula, represents the weight coefficient of actions and expressions in the comprehensive evaluation; Calculate the overall attention proportion of students at time t , and the calculation formula is as follows: ; In the formula, R represents the total number of students, represents the attention level of the r-th student at time t; Set the number of emotional categories to M, and calculate the proportion of students with different emotions at time t , and the calculation formula is as follows: ; In the formula, represents the indicator function, which determines whether the emotion of the r-th student at time t is the q-th emotion. If so, takes a value of 1, otherwise 0; Set the number of students' interaction behavior categories to NF, and calculate the interaction frequency of students at time t , and the calculation formula is as follows: ; ; In the formula, represents the number of times the l-th type of interaction behavior appears within time t, is the weight corresponding to the l-th interaction behavior represents an indicator function. If the interaction behavior bSS belongs to the l-th interaction behavior, then , otherwise . By traversing each interaction behavior bSS in, by judging its category and accumulating the count, the number of occurrences of the l-th interaction behavior within time t is finally obtained ; The comprehensive student concentration ratio , the ratio of students with different emotions and the student interaction frequency are combined to calculate the comprehensive index of students' classroom responses at time t . The calculation formula is as follows: ; In the formula, represents the weight coefficient corresponding to the comprehensive student concentration ratio , represents the weight coefficient corresponding to the ratio of students with different emotions , and both have a value range of 0 to 1, and 1, represents the weight corresponding to the q-th emotion; Calculate the optimized time allocation sequence . The calculation formula is as follows: ; In the formula, represents the adjustment coefficient, , respectively represent the thresholds of the comprehensive index of students' classroom responses , the comprehensive index of teachers' teaching behaviors , represents the adjustment vector of the time allocation for each teaching link; Feed the optimized time allocation sequence back to the artificial intelligence algorithm module to update the teaching plan P.

[0027] In the teaching and research management module, the process of obtaining the optimized teaching plan is as follows: Set the comprehensive evaluation index EZ to measure the effectiveness of the entire online precise teaching and research platform, and calculate EZ. The calculation formula is as follows: ; In the formula, $I_a$ represents the interaction index, which is used to measure the frequency of interaction between teacher teams on the platform; $C_a$ represents the collaboration degree, which is used to reflect the tightness of collaboration between teacher teams; $O_a$ represents the optimization level, which is used to measure the degree of optimization and improvement of the teaching plan in terms of quality, applicability, etc. after remote collaboration among teacher teams. 、 and respectively represent the weight coefficients corresponding to $I_a$, $C_a$, and $O_a$; The interaction index $I_a$ is calculated, and the calculation formula is as follows: ; In the formula, represents the number of meaningful interactions that actually occur between teacher teams on the platform within a certain time period, represents the maximum number of interactions that occur within this time period, represents the weight corresponding to each interaction, represents the quality score of the $i_r$-th interaction; The collaboration degree $C_a$ is calculated, and the calculation formula is as follows: ; In the formula, represents the number of collaboration projects completed by teacher teams within time $t$, represents the total number of projects within time $t$, represents the number of teacher teams participating in collaboration, represents the quantified value of the workload of resources invested by the $j_q$-th teacher team in each collaboration project; The optimization level $O_a$ is calculated, and the calculation formula is as follows: ; In the formula, represents the number of teaching plans to be evaluated after collaboration, represents the difference in quality evaluation indicators of the $r_k$-th teaching plan before and after collaboration optimization; Through big data analysis technology, in-depth analysis is carried out on the education field to provide suggestions on the research direction of topics for teachers. The analysis process is as follows: Set the recommended value evaluation index as $V_H$, which is used to measure the quality and practicality of the topic research direction suggestions provided for teachers. The calculation formula for $V_H$ is as follows: ; In the formula, $H_d$ represents the hot topic relevance, $D_d$ represents the difficulty-solving ability, $Y_d$ represents the trend consistency, 、 and respectively represent the weight coefficients corresponding to $H_d$, $D_d$, and $Y_d$; Calculate the hotspot relevance Hd, and the calculation formula is as follows: ; In the formula, represents the number of hotspot issues in the education field, represents the attention score of the mo-th hotspot issue, represents the correlation coefficient between the provided research direction suggestions for the topic and the mo-th hotspot issue; Calculate the difficulty conquest Dd, and the calculation formula is as follows: ; In the formula, represents the number of difficult and painful issues in teaching practice statistics, represents the influence weight of the md-th difficult and painful issue in actual teaching, represents the effectiveness score of the research direction of the topic for solving the md-th difficult and painful issue; Calculate the trend consistency Yd, and the calculation formula is as follows: ; In the formula, represents the number of dimensions in terms of the discipline development trend, represents the importance coefficient of the mp-th discipline development trend dimension, represents the degree of compliance of the research direction suggestions for the topic in the mp-th discipline development trend dimension; Set the comprehensive influence factor as XQ, which is used to measure the comprehensive effect of the comprehensive evaluation index EZ and the recommended value evaluation index VH on the optimization of the teaching plan P. Calculate the comprehensive influence factor XQ, and the calculation formula is as follows: ; In the formula, and represent the weight coefficients corresponding to EZ and VH respectively, and ; Optimize each element of the teaching plan P based on the comprehensive influence factor XQ.

[0028] Optimize each element of the teaching plan P based on the comprehensive influence factor XQ, and the optimization process is as follows: Set the teaching plan P to consist of four elements: teaching objective GP, teaching content CP, teaching method MP, and teaching evaluation EP, that is , and the optimized teaching plan is set as , The four elements of are represented as For the optimized teaching objective Perform calculations, and the calculation formula is as follows: ; In the formula, represents the optimized teaching objective weight coefficient, represents the optimized teaching objective potential value (which can be obtained based on the absolute value of the difference between the ideal value and the current value of the teaching objective, the same below); Perform calculations on the optimized teaching content The calculation formula is as follows: ; In the formula, represents the optimized teaching content weight coefficient, represents the optimized teaching objective potential value; Perform calculations on the optimized teaching method The calculation formula is as follows: ; In the formula, represents the optimized teaching method weight coefficient, represents the optimized teaching method potential value; Perform calculations on the optimized teaching evaluation The calculation formula is as follows: ; In the formula, represents the optimized teaching evaluation weight coefficient, represents the optimized teaching evaluation potential value.

[0029] In the training management module, use intelligent algorithms to automatically generate a personalized training course system, and the generation process is as follows: Define to represent the evaluation results of training effects. The training effect evaluation index system includes the following indicators: teacher ability improvement, teaching effect improvement, student learning effect improvement, training process quality, training resource utilization efficiency, training satisfaction, external evaluation; After numerically quantifying the above indicators, normalize the quantified values, and take the average of the normalized values as 's value; Set two thresholds for the evaluation results of training effects, namely the high threshold and the low threshold ; When it means that the training effect is excellent, and adjust the teaching plan : increase extracurricular knowledge, increase group interaction, and speed up the teaching schedule arrangement; When When it indicates excellent or medium training, adjust the teaching plan as follows: Add a special consolidation exercise session and slow down the teaching progress arrangement; When it indicates poor training, adjust the teaching plan as follows: Re-plan the order of knowledge point explanations, replace the teaching plan again, and slow down the teaching progress arrangement.

[0030] In the adjustment module, accurately determine the effectiveness of teaching intervention and give feedback. The determination process is as follows: Set the student learning effect evaluation function as Esk and calculate Esk. The calculation method is as follows: ; In the formula, Njs represents the number of student teaching characteristics. Student teaching characteristics include: exam score, homework completion score, class participation score, and learning time score. represents the weight of the r-th student learning characteristic. represents the conversion function of the data of the r-th student learning characteristic. represents the index variable; Set the teacher teaching effect evaluation function as Esj and calculate Esj. The calculation method is as follows: ; In the formula, Macb represents the number of teacher teaching characteristics. Teacher teaching characteristics include: teaching method diversity score, teaching content integrity score, teaching progress compliance score, and teacher-student interaction frequency score. represents the weight of the nbcq-th teacher teaching characteristic, IQc represents the index variable, represents the conversion function of the data of the nbcq-th teacher teaching characteristic; Here and in the subsequent conversion functions, the original data is converted into a form suitable for analysis, including methods such as normalization, standardization, and fuzzification, to ensure the consistency and comparability of the data.

[0031] Set the research and teaching achievement influence evaluation function as Ejy and calculate Ejy. The calculation method is as follows: ; In the formula, Maut represents the number of research and teaching achievement characteristics. Research and teaching achievement characteristics include: research and teaching achievement application value score, research and teaching achievement innovation score, and research and teaching achievement influence degree score on the teaching plan. IQut represents the index variable, represents the weight of the nunf-th research and teaching achievement characteristic. represents the conversion function of the data of the nunf-th research and teaching achievement characteristic; Set the teaching intervention effectiveness evaluation function as Effca and calculate Effca. The calculation formula is as follows: ; In the formula, , and represent the weight coefficients corresponding to Esk, Esj, and Ejy respectively, and ; Set the evaluation threshold as ; When , it is determined that the teaching intervention effectiveness is good, and teaching is carried out according to the adjusted teaching plan ; When , it means that the teaching intervention effectiveness is poor, a reminder is sent, and the teacher is retrained and the teaching plan is adjusted again.

Claims

1. An integrated teaching, research and training intelligent education management system, characterized in that Including: A teaching management module, which is used to obtain the learning progress of students and the teaching progress of teachers, and dynamically generate a teaching plan based on the learning progress of students and the teaching progress of teachers by using artificial intelligence algorithms; A teaching and research management module, which is an online precise teaching and research platform based on cloud computing, supports teacher teams in multiple regions to conduct remote collaboration to study the teaching plan, and optimize the teaching plan to obtain an optimized teaching plan; A training management module, which uses multi-modal data fusion technology to comprehensively analyze the teaching and teaching and research achievement data of teachers, identify the advantages and disadvantages of different teachers, automatically generate a personalized training course system by using intelligent algorithms. At the same time, during the training, establish an evaluation index system for training effects, determine the training effects, and make adaptive adjustments to the optimized teaching plan according to the training effects; An adjustment module, which is used to build an all-round teaching effect evaluation ecosystem, collect various types of teaching feedback data, use machine learning methods to analyze the teaching feedback, accurately determine the effectiveness of teaching interventions and give feedback.

2. The integrated teaching, research and training intelligent education management system according to claim 1, wherein Regarding the teaching management module, a teaching plan is dynamically generated by using artificial intelligence algorithms based on the learning progress of students and the teaching progress of teachers. The generation process is as follows: Set the student learning progress index as S, and calculate S. The calculation formula is as follows: ； Wherein, n represents the number of different dimensions for measuring the learning progress of students, i represents the index variable of n, represents the weight coefficient of each dimension of the learning progress, and its value range is between 0 and 1, represents the quantified value of each dimension of the learning progress; Set the teacher teaching progress index as J, and calculate J. The calculation formula is as follows: ； Wherein, m represents the number of different dimensions for measuring the teaching progress of teachers, represents the index variable of m, represents the weight coefficient of each dimension of the teaching progress, represents the quantified value of each dimension of the teaching progress; Through the artificial intelligence algorithm module Receive S and J, clean S and J, normalize S and J to the same dimension, and dynamically generate the teaching plan P according to the differences between S and J. The generation formula is as follows: ； Wherein, GPG represents the priority of teaching objectives, and CPC represents the complexity of course content, 、 、 and are the weight coefficients corresponding to S, J, GPG, and CPC respectively; Map the teaching plan P to the initial time allocation sequence , and its time allocation rule is defined by the following formula: ； Wherein, VVP represents the total number of teaching sessions, represents the complexity score of the vvp-th session in the teaching plan P, is the basic time weight of session vvp; Analyze the teacher's teaching behavior and students' classroom responses based on the object detection algorithm of deep learning.

3. The integrated teaching, research and training intelligent education management system according to claim 2, characterized in that, Regarding the analysis of the teacher's teaching behavior and students' classroom responses based on the object detection algorithm of deep learning, the analysis process is as follows: Obtain the real-time image data of the camera in the classroom, and set the image sequence at time t as , perform grayscale conversion and noise reduction processing on the obtained images to obtain the preprocessed image sequence ; Based on an image sequence , identify and obtain the types of the teacher's actions, expression categories, as well as the students' attentiveness, emotion categories, and interaction behavior categories; Quantification indicators of the activity and diversity of the teacher's actions at time t are calculated, and the calculation formula is as follows: ； In the formula, represents the number of occurrences of the v-th type of action within time t, represents the weight corresponding to the v-th type of action, and the number of action categories is C; Set the number of expression categories as E, and calculate the positivity of the teacher's expression at time t , and the calculation formula is as follows: ； In the formula, represents the duration of the z-th expression appearing within time t, represents the positive degree weight corresponding to the z-th expression; Integrate the teacher's action and expression information to calculate the comprehensive index of the teacher's teaching behavior at time t , and the calculation formula is as follows: ； In the formula, represents the weight coefficient of actions and expressions in the comprehensive evaluation; Calculate the overall concentration ratio of students at time t , and the calculation formula is as follows: ； where R represents the total number of students, represents the concentration of the r-th student at time t; Set the number of emotion categories as M, and calculate the proportion of students with different emotions at time t , and the calculation formula is as follows: ； In the formula, represents an indicator function that determines whether the emotion of the r-th student at time t is the q-th emotion. If so, takes the value of 1, otherwise 0; Set the number of categories of students' interaction behaviors as NF, and calculate the interaction frequency of students at time t , and the calculation formula is as follows: ； ； Wherein, represents the number of occurrences of the l-th interaction behavior within time t, is the weight corresponding to the l-th interaction behavior, represents an indicator function. If the interaction behavior bSS belongs to the l-th interaction behavior, then , otherwise . By traversing each interaction behavior bSS therein, judging its category and accumulating the count, the number of occurrences of the l-th interaction behavior within time t is finally obtained ; Overall student concentration ratio 、Ratio of students with different emotions and student interaction frequency , calculate the comprehensive index of students' classroom responses at time t , and the calculation formula is as follows: ； In the formula, represents the weight coefficient of the overall concentration ratio of the corresponding students , represents the weight coefficient of the proportion of students with different emotions . And both have a value range of 0 to 1, and 1, represents the weight corresponding to the q-th emotion; Calculated optimized time allocation sequence , and the calculation formula is as follows: ； In the formula, represents the adjustment coefficient, , respectively represent the comprehensive index of students' classroom responses , the comprehensive index of teachers' teaching behaviors threshold values, represents the adjustment vector of the time allocation for each teaching link; Feed the optimized time allocation sequence back to the artificial intelligence algorithm module to update the teaching plan P.

4. The integrated teaching, research and training intelligent education management system according to claim 1, characterized in that Regarding the teaching and research management module, the process of obtaining the optimized teaching plan is as follows: Set the comprehensive evaluation index EZ to measure the effectiveness of the entire online precise teaching and research platform, and calculate EZ. The calculation formula is as follows: ； In the formula, Ia represents the interaction index, which is used to measure the frequency of interaction between teacher teams on the platform; Ca represents the degree of collaboration, which is used to reflect the closeness of collaboration between teacher teams; Oa represents the optimization level, which is used to measure the degree to which the teaching plan is optimized and improved in terms of quality and applicability after remote collaboration of the teacher team. , and Respectively represent the weight coefficients corresponding to Ia, Ca and Oa; Calculate the interaction index Ia. The calculation formula is as follows: ； In the formula, represents the number of meaningful interactions that actually occur between teacher teams on the platform within a certain time period, represents the maximum number of interactions that occur within this time period, represents the weight corresponding to each interaction, represents the quality score of the ir-th interaction; Calculate the collaboration degree Ca. The calculation formula is as follows: ； In the formula, represents the number of collaborative projects completed by the teacher team within time t, represents the total number of projects within time t, represents the number of teacher teams participating in collaboration, represents the quantitative value of the workload of the jq-th teacher team investing resources in each collaborative project; Calculate the optimization level Oa. The calculation formula is as follows: ； In the formula, represents the number of teaching plans evaluated after collaboration, represents the difference in the quality evaluation index of the rk-th teaching plan before and after collaborative optimization; Conduct in-depth analysis of the education field through big data analysis technology, and provide suggestions on topic research directions for teachers.

5. The integrated teaching, research and training intelligent education management system according to claim 4, characterized in that Regarding the in-depth analysis of the education field through big data analysis technology and providing suggestions on topic research directions for teachers, the analysis process is as follows: Set the suggestion value evaluation index as VH, which is used to measure the quality and practicality of the topic research direction suggestions provided for teachers, and calculate VH. The calculation formula is as follows: ； Wherein, Hd represents the hotspot correlation, Dd represents the difficulty conquest, Yd represents the trend consistency, 、 and represent the weight coefficients corresponding to Hd, Dd, and Yd respectively; Calculate the hotspot relevance Hd. The calculation formula is as follows: ； Wherein, represents the number of hot issues in the field of education, represents the degree of attention score of the mo -th hot issue, represents the correlation coefficient between the provided research direction suggestions for the topic and the mo -th hot issue; Calculate the difficulty overcoming degree Dd. The calculation formula is as follows: ； In the formula, represents the number of difficult and painful problems in the statistical teaching practice, represents the influence weight of the md-th difficult and painful problem in the actual teaching, represents the effectiveness score of the research direction of the subject for solving the md-th difficult and painful problem; Calculate the trend consistency Yd. The calculation formula is as follows: ； In the formula, represents the number of dimensions in terms of the discipline development trend, represents the importance coefficient of the mp-th discipline development trend dimension, represents the degree of compliance of the research direction suggestion of the subject with the mp-th discipline development trend dimension; Set the comprehensive influence factor as XQ, which is used to measure the comprehensive effect of the comprehensive evaluation index EZ and the suggestion value evaluation index VH on the optimization of the teaching plan P, and calculate the comprehensive influence factor XQ. The calculation formula is as follows: ； Wherein, and represent the respective weight coefficients of EZ and VH, and ; Optimize each element of the teaching plan P based on the comprehensive influence factor XQ.

6. The integrated teaching, research and training intelligent education management system according to claim 5, characterized in that, Regarding the optimization of each element of the teaching plan P based on the comprehensive influence factor XQ, the optimization process is as follows: The set teaching plan P consists of four elements: teaching objective GP, teaching content CP, teaching method MP, and teaching evaluation EP, that is , the optimized teaching plan is set as , The four elements of are expressed as ; For the optimized teaching objectives Perform calculations, and the calculation formula is as follows: ； In the formula, represents the optimized teaching objective weight coefficient, represents the optimized teaching objective potential value; For the optimized teaching content Perform calculations, and the calculation formula is as follows: ； In the formula, represents the weight coefficient of the optimized teaching content, represents the potential value of the optimized teaching objective; For the optimized teaching method perform calculations. The calculation formula is as follows: ； In the formula, represents the weight coefficient of the optimized teaching method, represents the potential value of the optimized teaching method; For the optimized teaching evaluation perform calculations, and the calculation formula is as follows: ； Wherein, represents the optimized teaching evaluation weight coefficient, represents the optimized teaching evaluation potential value.

7. The integrated teaching, research and training intelligent education management system according to claim 1, characterized in that Regarding the training management module, a personalized training course system is automatically generated by using intelligent algorithms. The generation process is as follows: Definition is used to represent the evaluation results of training effects. The training effect evaluation index system includes the following indicators: improvement of teachers' capabilities, improvement of teaching effects, improvement of students' learning effects, quality of the training process, utilization efficiency of training resources, training satisfaction, and external evaluation; After numerically quantifying the above indicators, normalize the quantified values, calculate the average of the normalized values, and use it as the value; Set two thresholds for the evaluation results of training effects as the high threshold and the low threshold ; When it indicates that the training effect is excellent, and the teaching plan is adjusted: extracurricular knowledge is increased, group interaction is increased, and the teaching schedule is accelerated; When it indicates that the training is excellent or medium, and the teaching plan is adjusted as follows: add a special consolidation practice session and slow down the teaching progress arrangement; When it indicates that the training is excellent or poor, and the teaching plan shall be adjusted: re-plan the explanation order of knowledge points, re-replace the teaching plan, and slow down the teaching progress arrangement.

8. The integrated teaching, research and training intelligent education management system according to claim 7, characterized in that, In the adjustment module, the teaching intervention effectiveness is accurately determined and feedback is provided. The determination process is as follows: Set the student learning effect evaluation function as Esk and calculate Esk. The calculation method is as follows: ； Wherein, Njs represents the number of students' teaching characteristics, and the students' teaching characteristics include: test score, homework completion score, class participation score, and study time score. represents the weight of the r-th student learning characteristic. represents the conversion function for the data of the r-th student learning characteristic. represents the index variable; Set the teacher teaching effect evaluation function as Esj and calculate Esj. The calculation method is as follows: ； Wherein, Macb represents the quantity of teachers' teaching characteristics, and the teachers' teaching characteristics include: the diversity score of teaching methods, the integrity score of teaching content, the compliance score of teaching progress, and the frequency score of teacher-student interaction. represents the weight of the nbcq-th teachers' teaching characteristic, and IQc represents an index variable. represents the conversion function of the data of the nbcq-th teachers' teaching characteristic. Set the research result influence evaluation function as Ejy and calculate Ejy. The calculation method is as follows: ； In the formula, Maut represents the number of characteristics of teaching and research achievements, and the characteristics of teaching and research achievements include: the scoring of the application value of teaching and research achievements, the scoring of the innovation of teaching and research achievements, and the scoring of the impact degree of teaching and research achievements on the teaching plan. IQut represents the index variable, represents the weight of the nunf-th characteristic of teaching and research achievements, represents the conversion function of the data of the nunf-th characteristic of teaching and research achievements; Set the teaching intervention effectiveness evaluation function as Effca and calculate Effca. The calculation formula is as follows: ； In the formula, , and represent the weight coefficients corresponding to Esk, Esj, and Ejy respectively, and ; Set the evaluation threshold to ; When , it is determined that the effect of teaching intervention is good, and teaching is carried out according to the adjusted teaching plan . When , it indicates that the effect of teaching intervention is poor, a reminder is sent, and the teacher is retrained and the teaching plan is adjusted.

Images