Online student learning behavior evaluation method and system based on multi-modal feature analysis
Through the online student learning behavior evaluation method based on multimodal feature analysis, a multimodal information database and evaluation system was established, and the problem of the failure of existing technology to effectively analyze learning efficiency is solved, and the accurate and personalized evaluation of students' learning behavior is realized, and teaching quality and students' learning motivation are improved.
Patent Information
- Application Number
- CN202510146975.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-02-08
- Publication Date
- 2025-05-30
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The existing technology only monitors the acquisition and utilization of students' learning materials, and fails to effectively analyze learning efficiency, making it difficult to provide students with personalized learning support and guidance, and the inability to adjust learning strategies in a timely manner, which affects the improvement of learning results.
The online student learning behavior evaluation method based on multimodal feature analysis is adopted. Through the steps of feature extraction, generation model, attendance detection analysis, student interest detection analysis, learning progress monitoring analysis and learning behavior evaluation, a multimodal information database for students' learning process is established, and a theoretical course evaluation system for multimodal information fusion is established using machine learning technology.
It has achieved accurate and personalized real-time evaluation of students' learning behavior, helping schools and teachers optimize teaching management, timely discover student problems, and take effective measures to improve teaching quality and students' learning motivation.
Smart Images
Figure CN120070118A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of educational analysis, and specifically to an online student learning behavior evaluation method and system based on multi-modal feature analysis. Background Art
[0002] The analysis and evaluation of students' teaching and learning behaviors refer to the systematic analysis and evaluation of the behaviors demonstrated by students during the learning process, aiming to deeply understand the students' learning status, behavior habits, and learning effects, provide scientific basis and effective reference for educators, and further optimize teaching methods, improve teaching effects, and promote the all-round development of students.
[0003] In the field of education, students' learning behaviors are closely related to their academic achievements. Understanding the characteristics and performances of students' learning behaviors helps to discover problems, provide support and guidance, help students better adapt to the learning environment, improve their learning initiative and participation. The analysis and evaluation of students' teaching and learning behaviors are of great significance, which can promote personalized teaching, improve the quality of education, and achieve the goal of high-quality education.
[0004] However, currently, only the acquisition and utilization of students' learning materials are monitored, but the learning efficiency is not further analyzed. This may make it difficult for the system to provide personalized learning support and guidance for students, unable to adjust learning strategies and methods in a timely manner, and affect the improvement of students' learning effects. Summary of the Invention
[0005] (I) Technical Problems to be Solved
[0006] Aiming at the deficiencies of the prior art, the present invention provides an online student learning behavior evaluation method and system based on multi-modal feature analysis, mainly to solve the problem that only the acquisition and utilization of students' learning materials are monitored currently, but the learning efficiency is not further analyzed, which may make it difficult for the system to provide personalized learning support and guidance for students, unable to adjust learning strategies and methods in a timely manner, and affect the improvement of students' learning effects.
[0007] (II) Technical Solutions
[0008] To achieve the above object, the present invention provides the following technical solutions:
[0009] An online student learning behavior evaluation method based on multi-modal feature analysis, including the following steps:
[0010] S1: Feature extraction. Extract the effective regions of each student's face for feature collection, use the Maximal Information Coefficient (MIC) for feature screening to remove irrelevant factors, calculate the correlation coefficients between the feature space X and the performance space S column by column, select the feature corresponding to the maximum value of the correlation coefficient as the first feature, calculate the MIC values between the feature f1 and other features, and select the feature corresponding to Lmax as the second feature f 2 ; Remove the first feature and repeat the above steps until enough feature quantities are obtained;
[0011] S2: Model generation. After using MIC analysis for feature screening, reorganize the screened features into a feature space and use random forest for regression to obtain the final evaluation model;
[0012] S3: Attendance rate detection and analysis. Detect the attendance rate data of each student in each monitoring period, and obtain the attendance rate of students in each monitoring period based on the attendance data of each student in each monitoring period;
[0013] S4: Student interest detection and analysis. Detect the learning interest data of each student in class in each monitoring period, and obtain the learning interest coefficient of students in each monitoring period based on the learning interest data of each student in class in each monitoring period;
[0014] S5: Learning progress monitoring and analysis. Monitor the learning progress data of each student in each monitoring period, and obtain the learning efficiency of students in each monitoring period based on the learning progress data of each student in each monitoring period;
[0015] S6: Learning behavior evaluation. Adopt a method that combines learning analysis technology and data mining algorithms. Based on the attendance rate, student interest, and learning progress generated by students' online learning on the theoretical course learning platform, establish an evaluation model for the learning effect of theoretical online courses, evaluate the learning effect of students, and use visualization technology to output the evaluation results in the form of charts and numbers.
[0016] Further, the multimodal-based networked teaching data analysis method uses the Maximal Information Coefficient (MIC) for feature screening to remove irrelevant factors. The specific steps are as follows: First, calculate the correlation coefficients P = (p1, p2,..., pn) between the feature space X and the performance space S column by column, and select the feature corresponding to the maximum value of the correlation coefficient Pmax as the first feature, assumed to be f1 = Xk; then calculate the MIC values M = (m1,..., mk-1, mk+1,..., mn) between the feature f1 and other features, let L = 0.5 * P(i≠k) + 0.5 * (1 - M), and select the feature corresponding to Lmax as the second feature f2; remove the first feature and repeat the above steps until enough feature quantities are obtained, or the maximum value of the current MIC is less than a certain threshold.
[0017] Based on the foregoing solution, the attendance data includes the class attendance rate and the assignment submission rate.
[0018] As a further solution of the present invention, the learning interest data includes the degree of classroom participation and the completeness of course playback.
[0019] Furthermore, the learning progress data includes the progress of assignment completion and the progress of material reading.
[0020] The present invention also proposes an online student learning behavior evaluation system based on multi-modal feature analysis, including a terminal platform. The terminal platform includes a feature extraction and screening module, an evaluation model acquisition module, an attendance rate detection and analysis module, a student interest detection and analysis module, a learning progress monitoring and analysis module, and an evaluation result output module. The feature extraction and screening module is connected to the evaluation model acquisition module to transmit the extracted and screened feature data to the evaluation model acquisition module to generate a model. The attendance rate detection and analysis module, the student interest detection and analysis module, and the learning progress monitoring and analysis module are all connected to the evaluation result output module to transmit the data obtained by the attendance rate detection and analysis module, the student interest detection and analysis module, and the learning progress monitoring and analysis module to the evaluation result output module to evaluate the learning behavior of students.
[0021] Based on the foregoing solution, the attendance rate detection and analysis module is used to detect the attendance rate data of each student in each monitoring period, and obtain the attendance rate of students in each monitoring period according to the attendance data of each student in each monitoring period; the student interest detection and analysis module is used to detect the learning interest data of each student in class in each monitoring period, and obtain the learning interest coefficient of students in each monitoring period according to the learning interest data of each student in class in each monitoring period; the learning progress monitoring and analysis module is used to monitor the learning progress data of each student in each monitoring period, and obtain the learning efficiency of each student in each monitoring period according to the learning progress data of each student in each monitoring period.
[0022] As a further solution of the present invention, the evaluation result output module is used to integrate and analyze the learning ability data, physiological data, and learning behavior data generated by students in the online learning platform of theoretical courses by combining learning analysis technology and data mining algorithms, establish an evaluation model for the learning effect of theoretical online courses, evaluate the learning effect of students, and output the evaluation results in the form of charts and numbers by applying visualization technology.
[0023] (III) Beneficial Effects
[0024] Compared with the prior art, the present invention provides an online student learning behavior evaluation method and system based on multi-modal feature analysis, having the following beneficial effects:
[0025] 1. The present invention uses various learning behavior data provided by the theoretical course learning platform, integrates physiological signals collected by intelligent bracelets and students' learning ability information, and establishes a multi-modal information database for the students' learning process; uses machine learning technology to establish a theoretical course evaluation system for multi-modal information fusion, providing theoretical and technical method support for the precise and personalized real-time evaluation of the online course learning process and the dynamic adjustment of teaching plans.
[0026] 2. The present invention obtains the attendance rate of students in each monitoring period based on the attendance data analysis of each student in each monitoring period, obtains the learning interest coefficient of students in each monitoring period based on the learning interest data analysis of each student in class in each monitoring period, obtains the learning efficiency of students in each monitoring period based on the learning progress data analysis of each student in each monitoring period, and then comprehensively analyzes to obtain the learning behavior evaluation index of students in each monitoring period, which can help schools and teachers better optimize teaching management, timely discover students' problems, take effective measures, and improve teaching quality and students' learning motivation.
[0027] 3. The present invention can timely discover the learning status and problems of students by counting the number of late arrivals and absences of each student, help adjust the teaching plan and intervene in students' learning situations in a timely manner, can timely discover possible learning problems of students and give guidance by evaluating the completeness of course playback of students, improve students' learning effects, and can also monitor the progress of students' homework completion and material reading in real time, help teachers understand the learning progress of students, and timely adjust teaching content and methods. BRIEF DESCRIPTION OF THE DRAWINGS
[0028] Figure 1 is a schematic flow chart of the online student learning behavior evaluation method based on multi-modal feature analysis proposed by the present invention;
[0029] Figure 2 is a schematic diagram of the structure of the online student learning behavior evaluation system based on multi-modal feature analysis proposed by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS
[0030] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.
[0031] Embodiment 1
[0032] Refer to Figure 1 - Figure 2, An online student learning behavior evaluation method based on multimodal feature analysis, including the following steps:
[0033] S1: Feature extraction. Extract the effective regions of each student's face for feature collection, use the Maximal Information Coefficient (MIC) for feature screening to remove irrelevant factors, calculate the correlation coefficient between the feature space X and the performance space S column by column, select the feature corresponding to the maximum value of the correlation coefficient as the first feature, calculate the MIC values between the feature f1 and other features, and select the feature corresponding to Lmax as the second feature f 2 ; Remove the first feature and repeat the above steps until enough feature quantities are obtained;
[0034] S2: Model generation. After using MIC analysis for feature screening, reorganize the screened features into a feature space, and use random forest for regression to obtain the final evaluation model;
[0035] S3: Attendance rate detection and analysis. Detect the attendance rate data of each student in each monitoring period, and obtain the attendance rate of students in each monitoring period based on the attendance data of each student in each monitoring period;
[0036] S4: Student interest detection and analysis. Detect the learning interest data of each student in class in each monitoring period, and obtain the learning interest coefficient of students in each monitoring period based on the learning interest data of each student in class in each monitoring period;
[0037] S5: Learning progress monitoring and analysis. Monitor the learning progress data of each student in each monitoring period, and obtain the learning efficiency of students in each monitoring period based on the learning progress data of each student in each monitoring period;
[0038] S6: Learning behavior evaluation. Adopt a method combining learning analysis technology and data mining algorithms. Based on the attendance rate, student interest, and learning progress generated by students' online learning on the theoretical course online learning platform, establish an evaluation model for the learning effect of theoretical online courses, evaluate the learning effect of students, and output the evaluation results in the form of charts and numbers through visualization technology. Through the various learning behavior data provided by the theoretical course learning platform, fuse the physiological signals collected by intelligent bracelets and the student learning ability information to establish a multimodal information database for the student learning process; use machine learning technology to establish a theoretical course evaluation system for multimodal information fusion, providing theoretical and technical method support for the precise and personalized real-time evaluation of the online course learning process and the dynamic adjustment of teaching plans.
[0039] In particular, in the present invention, the multi-modal networked teaching data analysis method uses the maximum information coefficient MIC for feature screening to remove irrelevant factors. The specific steps are as follows: First, calculate the correlation coefficient P=(p1, p2,..., pn) between the feature space X and the performance space S column by column, and select the feature corresponding to the maximum value Pmax of the correlation coefficient as the first feature, assumed to be f1 = Xk; then calculate the MIC values M=(m1,..., mk-1, mk+1,..., mn) between the feature f1 and other features, and let L = 0.5*P(i≠k)+0.5*(1 - M), and select the feature corresponding to Lmax as the second feature f2; remove the first feature and repeat the above steps until the sufficient number of features is obtained, or the maximum value of the current MIC is less than a certain threshold. The attendance data includes the class attendance rate and the assignment submission rate, the learning interest data includes the class participation degree and the course playback completion degree, and the learning progress data includes the assignment completion progress and the material reading progress.
[0040] The present invention also proposes an online student learning behavior evaluation system based on multi-modal feature analysis, including a terminal platform. The terminal platform includes a feature extraction and screening module, an evaluation model acquisition module, an attendance rate detection and analysis module, a student interest detection and analysis module, a learning progress monitoring and analysis module, and an evaluation result output module. The feature extraction and screening module is connected to the evaluation model acquisition module to transmit the extracted and screened feature data to the evaluation model acquisition module to generate a model. The attendance rate detection and analysis module, the student interest detection and analysis module, and the learning progress monitoring and analysis module are all connected to the evaluation result output module to transmit the data obtained by the attendance rate detection and analysis module, the student interest detection and analysis module, and the learning progress monitoring and analysis module to the evaluation result output module to evaluate the learning behavior of students. According to the attendance data of each student in each monitoring period, the attendance rate of students in each monitoring period is obtained. According to the learning interest data of each student in class in each monitoring period, the learning interest coefficient of students in each monitoring period is obtained. According to the learning progress data of each student in each monitoring period, the learning efficiency of students in each monitoring period is obtained. Furthermore, through comprehensive analysis, the learning behavior evaluation index of students in each monitoring period is obtained, which can help schools and teachers better optimize teaching management, timely discover student problems, take effective measures, and improve teaching quality and student learning motivation.
[0041] It should be specifically noted that the attendance rate detection and analysis module is used to detect the attendance rate data of each student in each monitoring period, and obtain the attendance rate of students in each monitoring period based on the attendance data of each student in each monitoring period. By counting the number of late arrivals and absences of each student, the learning status and problems of students can be discovered in a timely manner, which helps to adjust the teaching plan and intervene in the learning situation of students in a timely manner. The student interest detection and analysis module is used to detect the learning interest data of each student in class in each monitoring period, and obtain the learning interest coefficient of students in each monitoring period based on the learning interest data of each student in class in each monitoring period. By evaluating the completeness of course playback of students, possible learning problems of students can be discovered in a timely manner and guidance can be given to improve the learning effect of students. The learning progress monitoring and analysis module is used to monitor the learning progress data of each student in each monitoring period, and obtain the learning efficiency of each student in each monitoring period based on the learning progress data of each student in each monitoring period. It can monitor the progress of students' homework completion and material reading in real time, helping teachers understand the learning progress of students and adjust teaching content and methods in a timely manner. The evaluation result output module is used to integrate and analyze the learning ability data, physiological data and learning behavior data generated by students in the online learning platform of theoretical courses by combining learning analysis technology and data mining algorithms, establish an evaluation model for the learning effect of theoretical online courses, evaluate the learning effect of students, and output the evaluation results in the form of charts and numbers by applying visualization technology.
[0042] Embodiment 2
[0043] Refer to Figure 1 - Figure 2 , an online student learning behavior evaluation method based on multi-modal feature analysis, includes the following steps:
[0044] S1: Feature extraction, extract the effective area of each student's face for feature collection, use the maximum information coefficient MIC for feature screening to remove irrelevant factors, calculate the correlation coefficient between the feature space X and the score space S column by column, and select the feature corresponding to the maximum value of the correlation coefficient as the first feature. Calculate the MIC value between the feature f1 and other features, and select the feature corresponding to Lmax as the second feature f 2 ; Remove the first feature and repeat the above steps until enough feature quantities are obtained;
[0045] S2: Generate a model. After using MIC analysis for feature screening, reorganize the screened features into a feature space, and use random forest for regression to obtain the final evaluation model;
[0046] S3: Attendance rate detection and analysis, detect the attendance rate data of each student in each monitoring period, and obtain the attendance rate of students in each monitoring period based on the attendance data of each student in each monitoring period;
[0047] S4: Student interest detection and analysis, detecting the learning interest data of each student in each monitoring period in the classroom, and obtaining the learning interest coefficient of students in each monitoring period according to the learning interest data analysis of each student in each monitoring period;
[0048] S5: Learning progress monitoring and analysis, monitoring the learning progress data of each student in each monitoring period, and obtaining the learning efficiency of students in each monitoring period according to the learning progress data analysis of each student in each monitoring period;
[0049] S6: Learning behavior evaluation, adopting a method combining learning analysis technology and data mining algorithms, establishing an evaluation model for the learning effect of theoretical online courses according to the attendance rate, student interest and learning progress generated by students' online learning on the theoretical course online learning platform, evaluating the learning effect of students, and outputting the evaluation results in the form of charts and numbers through visualization technology. Through the various learning behavior data provided by the theoretical course learning platform, integrating the physiological signals collected by the smart bracelet and the student learning ability information, establishing a multi-modal information database for the student learning process; using machine learning technology, establishing a multi-modal information fusion theoretical course evaluation system, providing theoretical and technical method support for the precise and personalized real-time evaluation of the online course learning process and the dynamic adjustment of teaching plans.
[0050] Particularly in the present invention, the multi-modal networked teaching data analysis method uses the maximum information coefficient MIC for feature screening to remove irrelevant factors. The specific steps are as follows: First, calculate the correlation coefficient P = (p1, p2,..., pn) between the feature space X and the score space S column by column, and select the feature corresponding to the maximum value Pmax of the correlation coefficient as the first feature, assumed to be f1 = Xk; then calculate the MIC values M = (m1,..., mk-1, mk+1,..., mn) between the feature f1 and other features, let L = 0.5 * P(i≠k) + 0.5 * (1 - M), and select the feature corresponding to Lmax as the second feature f2; remove the first feature and repeat the above steps until the sufficient number of features is obtained, or the maximum value of the current MIC is less than a certain threshold. The attendance data includes the classroom attendance rate and the assignment submission rate, the learning interest data includes the classroom participation degree and the course playback completion degree, and the learning progress data includes the assignment completion progress and the material reading progress.
[0051] The present invention also proposes an online student learning behavior evaluation system based on multimodal feature analysis, including a terminal platform. The terminal platform includes a feature extraction and screening module, an evaluation model acquisition module, an attendance rate detection and analysis module, a student interest detection and analysis module, a learning progress monitoring and analysis module, and an evaluation result output module. The feature extraction and screening module is connected to the evaluation model acquisition module to transmit the extracted and screened feature data to the evaluation model acquisition module to generate a model. The attendance rate detection and analysis module, the student interest detection and analysis module, and the learning progress monitoring and analysis module are all connected to the evaluation result output module, so that the data obtained by the attendance rate detection and analysis module, the student interest detection and analysis module, and the learning progress monitoring and analysis module are transmitted to the evaluation result output module to evaluate the learning behavior of students. The attendance rate of students in each monitoring period is obtained by analyzing the attendance data of each student in each monitoring period. The learning interest coefficient of students in each monitoring period is obtained by analyzing the learning interest data of each student in class in each monitoring period. The learning efficiency of students in each monitoring period is obtained by analyzing the learning progress data of each student in each monitoring period. Furthermore, the learning behavior evaluation index of students in each monitoring period is comprehensively analyzed, which can help schools and teachers better optimize teaching management, timely discover student problems, take effective measures, and improve teaching quality and student learning motivation.
[0052] It should be particularly noted that the attendance rate detection and analysis module is used to detect the attendance rate data of each student in each monitoring period and obtain the attendance rate of students in each monitoring period according to the attendance data of each student in each monitoring period. By counting the number of late arrivals and absences of each student, the learning status and problems of students can be timely discovered, which helps to timely adjust the teaching plan and intervene in the learning situation of students. The student interest detection and analysis module is used to detect the learning interest data of each student in class in each monitoring period and obtain the learning interest coefficient of students in each monitoring period according to the learning interest data of each student in class in each monitoring period. By evaluating the completeness of course playback of students, possible learning problems of students can be timely discovered and guidance can be given to improve the learning effect of students. The learning progress monitoring and analysis module is used to monitor the learning progress data of each student in each monitoring period and obtain the learning efficiency of each student in each monitoring period according to the learning progress data of each student in each monitoring period. It can real-time monitor the progress of students' homework completion and material reading, help teachers understand the learning progress of students, and timely adjust teaching content and methods. The evaluation result output module is used to integrate and analyze the learning ability data, physiological data, and learning behavior data generated by students in the online learning platform of theoretical courses by using a method combining learning analysis technology and data mining algorithms, establish an evaluation model for the learning effect of theoretical online courses, evaluate the learning effect of students, and output the evaluation result in the form of charts and numbers by using visualization technology.
[0053] In the description in this document, it should be noted that relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article or device.
Claims
1. An online student learning behavior evaluation method based on multimodal feature analysis, characterized in that: The following steps are involved: S1: Feature extraction, extract the effective area of each student's face for feature collection, use the maximum information coefficient MIC for feature screening, remove irrelevant factors, calculate the correlation coefficient between the feature space X and the score space S column by column, select the feature corresponding to the maximum value of the correlation coefficient as the first feature, calculate the MIC value between feature f1 and other features, and select the feature corresponding to Lmax as the second feature f2; remove the first feature, and repeat the above steps until the maximum number of features is obtained; S2: Generate a model. After using MIC analysis to screen features, the screened features are reorganized into feature space, and random forest regression is used to obtain the final evaluation model. S3: Attendance rate detection and analysis, detecting the attendance rate data of each student in each monitoring period, and obtaining the attendance rate of each student in each monitoring period according to the attendance data analysis of each student in each monitoring period; S4: Student interest detection and analysis, detecting the learning interest data of each student in class in each monitoring period, and obtaining the learning interest coefficient of each student in each monitoring period according to the learning interest data of each student in class in each monitoring period; S5: Learning progress monitoring and analysis, monitoring the learning progress data of each student in each monitoring period, and obtaining the learning efficiency of each student in each monitoring period based on the learning progress data of each student in each monitoring period; S6: Learning behavior evaluation uses a method that combines learning analysis technology with data mining algorithms. According to students' attendance rate, student interests and learning progress generated by learning on the theoretical course online learning platform, a theoretical online course learning effect evaluation model is established to evaluate students' learning effects, and the evaluation results are output in the form of charts and numbers using visualization technology.
2. The online student learning behavior evaluation method based on multimodal feature analysis according to claim 1 is characterized in that: The multimodal network teaching data analysis method adopts the maximum information coefficient MIC for feature screening and removes irrelevant factors. The specific steps are as follows: first, the correlation coefficient P=(p1, p2, ..., pn) between the feature space X and the score space S is calculated column by column, and the feature corresponding to the maximum value Pmax of the correlation coefficient is selected as the first feature, assuming that f1=Xk; then the MIC value M=(m1, ..., mk-1, mk+1, ..., mn) between the feature f1 and other features is calculated, and L=0.5*P(i≠k)+0.5*(1-M) is set, and the feature corresponding to Lmax is selected as the second feature f2; remove the first feature, and repeat the above steps until the maximum number of features is obtained, or the current maximum value of MIC is less than a certain threshold.
3. The online student learning behavior evaluation method based on multimodal feature analysis according to claim 1 is characterized in that: The attendance data includes class attendance rate and homework submission rate.
4. The online student learning behavior evaluation method based on multimodal feature analysis according to claim 3 is characterized in that: The learning interest data includes class participation level and course playback completeness.
5. The online student learning behavior evaluation method based on multimodal feature analysis according to claim 4 is characterized in that: The learning progress data includes homework completion progress and material reading progress.
6. An online student learning behavior evaluation system based on multimodal feature analysis, including a terminal platform, characterized in that: The terminal platform includes a feature extraction and screening module, an evaluation model acquisition module, an attendance rate detection and analysis module, a student interest detection and analysis module, a learning progress monitoring and analysis module and an evaluation result output module. The feature extraction and screening module is connected to the evaluation model acquisition module so that the extracted and screened feature data is transmitted to the evaluation model acquisition module so that it generates a model. The attendance rate detection and analysis module, the student interest detection and analysis module, and the learning progress monitoring and analysis module are all connected to the evaluation result output module so that the data obtained by the attendance rate detection and analysis module, the student interest detection and analysis module, and the learning progress monitoring and analysis module are transmitted to the evaluation result output module so that it evaluates the student's learning behavior.
7. The online student learning behavior evaluation system based on multimodal feature analysis according to claim 6 is characterized in that: The attendance rate detection and analysis module is used to detect the attendance rate data of each student in each monitoring period, and obtain the attendance rate of the students in each monitoring period based on the attendance data of each student in each monitoring period; the student interest detection and analysis module is used to detect the learning interest data of each student in class in each monitoring period, and obtain the learning interest coefficient of the students in each monitoring period based on the learning interest data of each student in class in each monitoring period; the learning progress monitoring and analysis module is used to monitor the learning progress data of each student in each monitoring period, and obtain the learning efficiency of each student in each monitoring period based on the learning progress data of each student in each monitoring period.
8. The online student learning behavior evaluation system based on multimodal feature analysis according to claim 6 is characterized in that: The evaluation result output module is used to integrate and analyze the learning ability data, physiological data and learning behavior data generated by students learning on the theoretical course online learning platform by combining learning analysis technology with data mining algorithm, establish a theoretical online course learning effect evaluation model, evaluate the students' learning effect, and use visualization technology to output the evaluation results in the form of charts and numbers.