Teaching platform online live broadcast management method and system

Abstract

The invention relates to the technical field of online education, in particular to a teaching platform online live broadcast management method and system, and the method comprises the following steps: based on the identity information provided by a student, checking the mobile phone number, user name and original record of the student, verifying the legality of the identity, retrieving the previous behavior data, judging the authority requirement, and classifying and distributing basic authority for the student. And generating a student identity authentication and basic permission allocation result. In the invention, through checking student identity information and behavior data, precision and flexibility of student classification and authority distribution are realized, student live broadcast behavior data are monitored, a participation effect is optimized, interaction between students and teaching contents is enhanced, student activeness is evaluated, and curriculum interaction forms such as question and answer links, group discussion and real-time feedback are adjusted according to grades. The learning interest is stimulated, the teaching experience is improved, behavior data are summarized to generate student archives, a basis is provided for teaching planning and evaluation, and intelligent management and personalized development of online education are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of online education, and particularly to an online live broadcast management method and system for a teaching platform. Background Art

[0002] The technical field of online education involves teaching activities and learning management based on the Internet, including distance education, the design, development, and management methods of online learning platforms. The key technologies in this field cover live broadcast technology, video conferencing technology, the distribution and management of teaching resources, real-time interaction, learning progress tracking, student assessment and feedback systems, etc. Online education technology aims to provide users with a convenient and efficient learning environment through a network platform, support remote teaching and diverse learning methods, break the limitations of time and space, and promote the popularization and personalization of educational resources.

[0003] Among them, the online live broadcast management method for a teaching platform is a video live broadcast control and management technology for an online education platform. Its main purpose is to optimize the organization and management process of online live broadcast teaching, and ensure the stability and efficiency of aspects such as video and audio quality, interaction links, and student participation during the live broadcast. Through a series of management measures and technical means, this method improves the fluency, interactivity, and real-time feedback effect of teaching activities, thereby improving the online education experience and promoting the wider application of distance education.

[0004] In the process of teaching resource distribution and live broadcast management in the prior art, although it can ensure the basic video, audio, and interaction quality, there are significant deficiencies in the personalized participation of students. Due to the lack of real-time monitoring and dynamic adjustment capabilities for student behavior data, the participation of some students is low, and the effective distinction between active and inactive students cannot be achieved, affecting the overall balance of teaching effects. The evaluation of students' learning status in the prior art only stays at the basic participation data statistics, and the performance of students cannot be refined through multi-dimensional behavior data, resulting in a single design for interaction links and the inability to dynamically adjust course content according to the activity levels of different students. In live broadcast teaching, the decline in the interaction quality of students cannot be detected and addressed in a timely manner, leading to a weakening of learning effects. The construction of student behavior profiles in the prior art also lacks comprehensiveness and systematicness, making it difficult to provide sufficient data support for subsequent teaching planning and student evaluation, thus limiting the development potential of personalized and refined management in online education. Summary of the Invention

[0005] In order to solve the technical problems that in the process of teaching resource distribution and live broadcast management in the prior art, although the basic video, audio and interaction quality can be guaranteed, there are significant deficiencies in the personalized participation of students. Due to the lack of real-time monitoring and dynamic adjustment capabilities for students' behavior data, the participation of some students is low, the effective distinction between active and inactive students cannot be achieved, affecting the overall balance of teaching effects. The evaluation of students' learning status in the prior art only stays at the basic participation data statistics, and the performance of students cannot be refined through multi-dimensional behavior data, resulting in a single design of the interaction session and the inability to dynamically adjust the course content according to the activity of different students. In live teaching, the decline in the interaction quality of students cannot be discovered and addressed in a timely manner, resulting in a weakening of learning effects. The construction of students' behavior archives in the prior art also lacks comprehensiveness and systematicness, making it difficult to provide sufficient data support for subsequent teaching planning and student evaluation, restricting the development potential of personalized and refined management in online education. The embodiments of the present invention provide a method and system for online live broadcast management of a teaching platform. The technical solutions are as follows: On the one hand, a method for online live broadcast management of a teaching platform is provided, and the method includes: S1: Based on the identity information provided by the student, check the student's mobile phone number, username and original records, verify the legality of the identity, retrieve the past behavior data to determine the permission requirements, classify and assign basic permissions to the student, and generate the student identity authentication and basic permission assignment result; S2: Based on the student identity authentication and basic permission assignment result, monitor the number of speeches, participation frequency, viewing duration, and discussion activity during the live broadcast, adjust the permissions in combination with the real-time data, and generate the behavior analysis and permission adjustment result; S3: Based on the behavior analysis and permission adjustment result, refine the student behavior data, count the number of speeches, question frequency, interaction content type and emotional expression, evaluate the learning participation situation, divide the activity level, and generate the student activity level; S4: According to the student activity level, adjust the course interaction session, increase Q&A and group discussions when the activity is low, evaluate the participation situation and dynamically match the content form, combine open discussion with real-time questions, and generate the interaction guidance and course adjustment result; S5: According to the interaction guidance and course adjustment result, summarize the student behavior data after the course ends, record the participation duration, question times and interaction quality, classify the student behavior, and generate the student behavior archive and course interaction summary.

[0006] As a further solution of the present invention, the student identity authentication and basic permission allocation results include mobile phone number verification status, username matching result, original record verification result, basic permission category, and initial permission configuration. The behavior analysis and permission adjustment results include real-time behavior data records, permission status updates, interaction permission status, and dynamic permission adjustment. The student activity level includes the number of speeches, question frequency, interaction content type, emotional expression index, and learning participation level. The interaction guidance and course adjustment results include Q&A session settings, group discussion configurations, real-time feedback forms, and interaction content adjustments. The student behavior file and course interaction summary include student participation duration, number of questions, interaction quality, participation classification results, and course data summary.

[0007] As a further solution of the present invention, based on the identity information provided by the student, the steps of verifying the legitimacy of the identity by checking the student's mobile phone number, username, and original record, retrieving past behavior data to determine the permission requirements, and classifying and allocating basic permissions to the student to generate the student identity authentication and basic permission allocation results are specifically as follows: S101: Based on the identity information provided by the student, check the mobile phone number, username, and original record provided by the student, segmentally extract the core fields of each piece of information, count the proportion and quantity of missing fields by comparing the matching degree of each field, check for field conflicts and record abnormal data, and obtain the student identity information verification result; S102: Based on the student identity information verification result, extract the past behavior records associated with the matching fields, identify behavior characteristics and perform classification label marking by screening the parameters of login behavior, access records, and operation logs, and determine whether the permission conditions are met to generate the student permission attribute classification result; S103: Based on the student permission attribute classification result, screen and allocate content according to the permission conditions marked in the classification label, check the levels of viewing, speaking, and interaction permissions in the permission allocation record, and combine the key characteristic values in the behavior data to check the allocation range to obtain the student identity authentication and basic permission allocation result.

[0008] As a further solution of the present invention, based on the student identity authentication and basic permission allocation results, the steps of monitoring the number of speeches, participation frequency, viewing duration, and discussion activity during the live broadcast, and adjusting the permissions in combination with real-time data to generate the behavior analysis and permission adjustment results are specifically as follows: S201: Based on the student identity authentication and basic permission allocation results, extract the behavior data of the number of speeches, count the total amount of data and mark abnormal items to obtain the student real-time behavior data monitoring result; S202: Based on the student real-time behavior data monitoring result, analyze the fluctuation trend of the behavior data, screen abnormal data segments and mark potential adjustment items to obtain the student permission adjustment reference data; S203: Based on the reference data for adjusting the trainee's permissions, use a behavior prediction algorithm to check the triggering conditions of the permission adjustment items, update the permission status, record the adjustment time and scope, and obtain the results of behavior analysis and permission adjustment.

[0009] As a further solution of the present invention, the formula of the behavior prediction algorithm is as follows: ; Wherein, represents the probability value of permission adjustment, represents the activity frequency of the trainee in the recent week, represents the task completion degree of the trainee recently, represents the number of unfinished tasks of the trainee, represents the number of task delays of the trainee, represents the weight coefficient of the influence of activity frequency on permission adjustment, represents the weight coefficient of the influence of task completion degree on permission adjustment, represents the behavior duration of the trainee, represents the adjustment coefficient of the influence of the trainee's behavior duration on permission adjustment, represents the learning difficulty level of the trainee in the recent month, represents the adjustment coefficient of the influence of learning difficulty on permission adjustment.

[0010] As a further solution of the present invention, based on the results of the behavior analysis and permission adjustment, refine the trainee's behavior data, count the number of speeches, question frequencies, types of interaction content, and emotional expressions, evaluate the learning participation, divide the activity levels, and the steps for generating the trainee activity levels are specifically as follows: S301: Based on the results of the behavior analysis and permission adjustment, extract the indicators of the number of speeches, question frequencies, types of interaction content, and emotional expressions in the trainee's behavior data, classify and count them according to the data type and summarize them to generate the statistical results of the trainee's behavior data; S302: Based on the statistical results of the trainee's behavior data, analyze the correlation between the types of interaction content and the emotional expression indicators, evaluate the activity of the trainee's learning participation, group the trainee participation levels through classified behavior characteristics, and generate the trainee participation level classification results; S303: Based on the trainee participation level classification results, record the level and corresponding participation of each trainee, check the classification results of active and inactive trainees, mark the status of trainees with different levels, and obtain the trainee activity levels.

[0011] As a further solution of the present invention, according to the activity level of the students, the course interaction session is adjusted. When the activity level is low, Q&A and group discussions are increased. The steps of evaluating the participation situation and dynamically matching the content form, combining open discussion and real-time questions, and generating the interaction guidance and course adjustment results are specifically as follows: S401: Based on the activity level of the students, identify the group of students with low activity levels, set the interaction tasks of adding Q&A sessions and group discussions, allocate the corresponding interaction tasks according to the activity levels of the students, and generate the interaction task allocation results; S402: Based on the interaction task allocation results, track the performance of the students in participating in the Q&A and discussion sessions in real time, record the participation frequency and the number of speeches, analyze the participation data of the interaction session, evaluate the participation effect and the trend of behavior changes of the students, and generate the evaluation results of the students' participation effect; S403: Based on the evaluation results of the students' participation effect, select the matching content form in combination with the trend of behavior changes, calculate the adjustment coefficient of the trend of behavior changes of the students, adjust the content forms of open discussion, real-time questions and instructor feedback, record the adjustment process and the status of the course form, and obtain the interaction guidance and course adjustment results.

[0012] As a further solution of the present invention, the steps for obtaining the adjustment coefficient of the trend of behavior changes of the students are as follows: ; Wherein, represents the adjustment coefficient of the th type of students, represents the total number of students, represents the th student's eigenvalue in the behavior change, represents the mean value of the th type of students in terms of characteristics, represents the th type of students' standard deviation in terms of behavior change characteristics.

[0013] As a further solution of the present invention, according to the interaction guidance and course adjustment results, after the course ends, summarize the students' behavior data, record the participation duration, the number of questions and the interaction quality, classify the students' behaviors, and the steps for generating the students' behavior profiles and course interaction summaries are specifically as follows: S501: Based on the interaction guidance and course adjustment results, extract the data of the students' participation duration, the number of questions and the interaction quality, divide and classify the data according to time periods, screen out the outliers and adjust the data format, and generate the summary results of the students' course behavior data; S502: Analyze the interaction relationship between the number of questions asked by the trainee and the participation duration based on the summary result of the trainee course behavior data. Compare the interactive quality score with the behavior performance, classify the trainees into different levels according to characteristics and label the characteristics to generate the classification result of the trainee interaction level. S503: Based on the classification result of the trainee interaction level, record the classified trainee behavior data, match and integrate the number of questions asked and the interactive quality, summarize the overall course interaction data and file it to obtain the trainee behavior file and the course interaction summary.

[0014] On the other hand, an electric vehicle status monitoring system is provided. The electric vehicle status monitoring system is used to execute the above-mentioned electric vehicle status monitoring method, and the system includes: The trainee identity authentication module checks the mobile phone number and user name provided by the trainee based on the identity information provided by the trainee, matches the record fields of the mobile phone number and user name, checks the integrity and consistency of the identity data item by item, and obtains the trainee identity verification result. The basic permission allocation module filters the trainee behavior records based on the trainee identity verification result, counts the viewing duration and the number of speeches, judges the permission range that the behavior records meet the conditions, and obtains the basic permission allocation information. The behavior monitoring and analysis module records the trainee behavior data based on the basic permission allocation information, extracts the speech duration and viewing frequency parameters, compares the behavior data with the index standard value, and generates the trainee behavior statistical result. The permission dynamic adjustment module classifies and extracts the behavior frequency data based on the trainee behavior statistical result, counts the difference between the parameter and the adjustment threshold, updates the applicable range of the trainee interaction permission, and obtains the permission adjustment information. The trainee activity evaluation module classifies and summarizes the speech frequency and the number of questions asked based on the permission adjustment information, statistically calculates the activity level parameters layer by layer, matches the level interval table, and generates the trainee activity level. The course interaction management module statistically calculates the trainee interaction data based on the trainee activity level, classifies and extracts the content type and the number of participations, adjusts the frequency and content form of the course interaction session, and generates the trainee behavior file and the course interaction summary.

[0015] The beneficial effects brought by the technical solution provided by the embodiment of the present invention at least include: By verifying the identity information, behavior data, and permission requirements of students, dynamically adjust the legitimacy of students' identities and basic permissions, achieve the accuracy and flexibility of student classification and permission allocation, significantly improve the refinement of permission management, monitor the behavior data of students during the live broadcast, combine real-time speech, viewing duration, and discussion activity, optimize the participation effect of students by dynamically adjusting permissions, enhance the interaction depth between students and teaching content, based on the further refined processing of behavior data, complete the multi-dimensional evaluation of student activity by counting indicators such as the number of speeches, question frequency, and emotional expression, achieve the scientific classification of students' participation, provide data support for the subsequent course link design, in the design of the course interaction link, flexibly adjust the interaction form according to the student activity level, including setting up Q&A sessions, group discussions, and real-time question feedback, effectively stimulate students' learning interest, improve the teaching experience and participation quality of online education, after the course ends, summarize the behavior data of students, combine information such as interaction quality and the number of questions, generate a student behavior profile, provide a comprehensive basis for subsequent teaching planning and student performance evaluation, through the real-time tracking and dynamic adjustment of students' behavior, not only improve the intelligent management level of online education, but also promote the efficient application and precise personalized development of the distance education model. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a schematic diagram of the working process of the present invention; Figure 2 It is a detailed flowchart of S1 of the present invention; Figure 3 It is a detailed flowchart of S2 of the present invention; Figure 4 It is a detailed flowchart of S3 of the present invention; Figure 5 It is a detailed flowchart of S4 of the present invention; Figure 6 It is a detailed flowchart of S5 of the present invention; Figure 7 It is a system flowchart of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0017] The following describes the technical solutions in the present invention with reference to the accompanying drawings.

[0018] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to represent examples, illustrations, or explanations. Any embodiment or design solution described as "example" in the present invention should not be construed as being more preferred or having more advantages than other embodiments or design solutions. Exactly speaking, the use of the word "example" is intended to present concepts in a specific way. In addition, in the embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one of the two.

[0019] In the embodiments of the present invention, the terms "image" and "picture" can sometimes be used interchangeably. It should be noted that when the difference between them is not emphasized, their intended meanings are the same. The terms "of", "corresponding", and "corresponding to" can sometimes be used interchangeably. It should be noted that when the difference between them is not emphasized, their intended meanings are the same.

[0020] In the embodiments of the present invention, sometimes a subscript such as W1 may be written in a non-subscript form such as W1. When the difference between them is not emphasized, their intended meanings are the same.

[0021] To make the technical problems, technical solutions, and advantages to be solved by the present invention clearer, the following will be described in detail with reference to the accompanying drawings and specific embodiments.

[0022] Please refer to Figure 1 , embodiments of the present invention provide an online live management method for a teaching platform. The processing flow of this method may include the following steps: S1: Based on the identity information provided by the students, by checking the students' mobile phone numbers, user names, and original records, determine the legitimacy of the students' identities, retrieve past behavior data to determine whether the basic permission requirements are met, classify the students, allocate viewing, speaking, and interaction permissions, and if the students' original behaviors meet the key criteria, increase the permissions to generate the student identity authentication and basic permission allocation results; S2: Based on the student identity authentication and basic permission allocation results, monitor the behavior data of the students during the live broadcast, including the number of speeches, participation frequency, video viewing duration, and discussion activity. Combine the real-time behavior data, and by dynamically adjusting the permissions, determine whether the students need interaction permissions to generate the behavior analysis and permission adjustment results; S3: Based on the behavior analysis and permission adjustment results, refine the behavior data of the students, count the number of speeches, question frequency, interaction content type, and emotional expression indicators, evaluate the learning participation, classify the activity levels of the students, record the participation of each student, identify active and inactive students, and generate the student activity levels; S4: According to the student activity levels, adjust the interactive sessions of the course in real time. If the student activity level is low, stimulate student participation by adding Q&A sessions and group discussions. After evaluating the student participation, dynamically select the matching content forms, including open discussions and allowing students to obtain lecturer feedback through real-time questions, to generate the interactive guidance and course adjustment results; S5: According to the interactive guidance and course adjustment results, summarize the behavior data of the trainees after the course ends, record the interaction situation of each trainee, including the participation duration, the number of questions asked, and the interaction quality, classify the trainees, record the course participation data, and generate trainee behavior profiles and course interaction summaries based on the interaction data and trainee performance.

[0023] The results of trainee identity authentication and basic permission allocation include mobile phone number verification status, username matching results, original record verification results, basic permission categories, and initial permission configurations. The results of behavior analysis and permission adjustment include real-time behavior data records, permission status updates, interactive permission status, and dynamic permission adjustments. The trainee activity levels include the number of speeches, question frequencies, types of interactive content, emotional performance indicators, and learning participation levels. The results of interactive guidance and course adjustment include Q&A session settings, group discussion configurations, real-time feedback forms, and interactive content adjustments. The trainee behavior profiles and course interaction summaries include trainee participation durations, the number of questions asked, interaction quality, participation classification results, and course data summaries.

[0024] Specifically, as Figure 2 shown, based on the identity information provided by the trainees, verify the legality of the identity by checking the trainees' mobile phone numbers, usernames, and original records, retrieve past behavior data to determine permission requirements, classify the trainees, and allocate basic permissions. The steps for generating the results of trainee identity authentication and basic permission allocation are as follows: S101: Based on the identity information provided by the trainees, check the mobile phone numbers, usernames, and original records provided by the trainees, segment and extract the core fields of each piece of information, count the proportion and quantity of missing fields by comparing the matching degree of each field, check for field conflicts and record abnormal data, and obtain the trainee identity information verification results; By dividing the trainee identity information into independent paragraphs according to the mobile phone number, username, and original record, respectively compare whether the mobile phone number conforms to the specified length, whether the username conforms to the character limit, and whether the original record contains the specified fields, count the proportion and quantity of missing fields, record the missing fields in a separate missing list for missing situations, and at the same time check whether the mobile phone number and username are registered repeatedly. For the mobile phone number matching rule, it is required that only one same value appears in each group of data, and the username must not be exactly the same as the existing usernames. Check for field conflicts and record abnormal data, mark the data with repeated mobile phone numbers and usernames as conflict data, further screen for blank items or illegal characters in the original record, and obtain the trainee identity information verification results through comparison and analysis.

[0025] S102: Based on the verification result of the trainee's identity information, extract the past behavior records associated with the matching fields. By screening the parameters of login behavior, access records, and operation logs, identify the behavior characteristics and conduct classification label marking. Determine whether the permission conditions are met, and generate the classification result of the trainee's permission attributes; By constructing a set of behavior data sets including the time distribution of login behavior, the key path of access records, and the categories of operation logs, screen the part that matches the trainee's identity. In the login behavior records, count the consecutive login days and time period distribution to determine the active behavior characteristics. The access records are divided according to different modules, and the proportional distribution between the frequently accessed modules and the less accessed modules is analyzed. Extract the commonly used operation instructions and time distribution characteristics from the operation logs, identify the behavior characteristics and conduct classification label marking. Mark the active trainees as "highly active", and those with a long login interval or no operation records as "lowly active". Determine whether the permission conditions are met, compare the behavior classification results with the permission allocation rules, and further mark the trainees who meet the advanced permission conditions as "high-permission", generating the classification result of the trainee's permission attributes.

[0026] S103: Based on the classification result of the trainee's permission attributes, screen and allocate the content according to the permission conditions marked in the classification label. By checking the levels of viewing, speaking, and interaction permissions in the permission allocation record, and combining the key feature values in the behavior data to verify the allocation range, obtain the trainee's identity authentication and basic permission allocation result; Match the classification result of the trainee's permission attributes with the current permission content list one by one, screen the permission content corresponding to the trainee category. By checking the levels of viewing, speaking, and interaction permissions in the permission allocation record, check in turn whether the allocated content contains all the fields in the permission list, and combine the key feature values in the behavior data to verify the allocation range. For example, grant full permissions to highly active trainees, limit the basic viewing permissions to lowly active trainees, and screen the interaction permissions according to the active interaction data in the behavior records, marking whether the permission range is consistent with the allocation standard, and complete the final verification of the permission conditions to obtain the trainee's identity authentication and basic permission allocation result.

[0027] Specifically, as Figure 3 shown, based on the trainee's identity authentication and basic permission allocation result, monitor the number of speeches, participation frequency, viewing duration, and discussion activity during the live broadcast, and adjust the permissions in combination with the real-time data. The steps to generate the behavior analysis and permission adjustment result are specifically as follows: S201: Based on the trainee's identity authentication and basic permission allocation result, extract the behavior data of the number of speeches, count the total amount of data and mark the abnormal items, obtaining the monitoring result of the trainee's real-time behavior data; Group the speech records of trainees in each interactive scenario according to the time series, count the distribution of the number of speeches in each time period, focus on recording the speech content and time intervals of trainees with high-frequency speeches. For cases of continuous high-frequency speeches, mark them as abnormal items. Screen the speech data for duplicate or contentless records, classify the invalid records as abnormal data, calculate the total amount of speech data, obtain the total number of valid speeches after removing the abnormal items, analyze the average speech interval of each trainee, record the fluctuations in speech behavior, and highlight the data of trainees with large fluctuations. For example, if a trainee suddenly has a sharp increase in the number of speeches in a short period of time, it means abnormal permission usage. Through the above process, generate the monitoring results of trainees' real-time behavior data and use it as the basis for subsequent data analysis.

[0028] S202: Based on the monitoring results of trainees' real-time behavior data, analyze the fluctuation trend of the behavior data, screen abnormal data segments and mark potential adjustment items to obtain reference data for trainees' permission adjustment; Plot the statistical speech data on the time axis as a behavior curve to observe the amplitude and frequency of data fluctuations. Classify the fluctuation range of trainees' speech behavior, mark the time periods with significantly higher fluctuation amplitudes than the average value as abnormal data segments, and extract the behavior characteristics of the time periods for in-depth analysis. For example, whether there is repeated or single content when there is high-frequency speech. For trainees whose speech behavior contains abnormal data, count the proportion of their abnormal data segments in the total behavior data, calculate the impact degree of the abnormal data on the overall behavior, and further screen potential adjustment items that affect permission allocation. For example, mark the high-frequency abnormal speech data as a factor triggering permission downgrade, record the duration and frequency of the fluctuations in the behavior data, and combine the identity authentication and permission allocation results of the trainees to provide a reliable reference basis for permission adjustment, and finally obtain reference data for trainees' permission adjustment.

[0029] S203: Based on the reference data for trainees' permission adjustment, use a behavior prediction algorithm to check the triggering conditions of the permission adjustment items, update the permission status and record the adjustment time and scope to obtain the results of behavior analysis and permission adjustment; The formula of the behavior prediction algorithm is as follows: ; Among them, represents the probability value of permission adjustment, represents the activity frequency of the trainee in the recent week, represents the task completion rate of the trainee recently, represents the number of unfinished tasks of the trainee, represents the number of task postponements of the trainee, represents the weight coefficient of the impact of activity frequency on permission adjustment, represents the weight coefficient of the impact of task completion rate on permission adjustment, Represents the behavior duration of the trainee, Indicates the adjustment coefficient of the impact of the trainee's behavior duration on permission adjustment, Represents the learning difficulty level of the trainee in the most recent month, Indicates the adjustment coefficient of the impact of learning difficulty on permission adjustment; Detailed explanation of the formula and the derivation process of formula calculation: This formula is used to calculate the probability value of trainee permission adjustment and generate a permission adjustment strategy. By weighted analysis of multiple factors such as the trainee's activity frequency, task completion rate, number of unfinished tasks and delay times, trainee behavior duration, and learning difficulty, the most appropriate timing and amplitude of permission adjustment are obtained; : The activity frequency of the trainee in the most recent week (unit: times). According to the trainee activity data automatically recorded by the system, such as the number of times the trainee logged in to the system and interacted in the past week, it is set to 25 times, which means the trainee had 25 interactions in the past week. This data reflects the trainee's activity level. Frequent activities indicate that the trainee has a strong need to increase permissions to provide more functions; : The task completion rate of the trainee in the near future (unit: percentage). By monitoring and recording the trainee's task completion situation through the system, it is set to 80%, indicating that the trainee has completed 80% of the tasks. This data is calculated by the ratio of the total number of tasks to the number of completed tasks, reflecting the trainee's learning efficiency and task completion situation. A higher completion rate indicates that the trainee has stronger capabilities and needs to be provided with more permissions or higher-difficulty tasks; : The number of unfinished tasks of the trainee (unit: pieces). The number of unfinished tasks is obtained through the trainee's task management system and is set to 5 pieces, indicating that the trainee has 5 unfinished tasks, measuring the backlog of the trainee's current tasks. A larger number of unfinished tasks indicates that the trainee has an excessive burden, so the permissions need to be adjusted to optimize the trainee's learning path; : The number of task delay times of the trainee (unit: times). By recording the completion time of each task of the trainee, the number of task delay times is obtained and is set to 3 times, indicating that the trainee has 3 task delay times in the near future, reflecting the frequency of the trainee not completing tasks on time within the specified time. A larger number of task delay times requires an adjustment of the trainee's permissions to reduce the task backlog; : The weight coefficient of the impact of activity frequency on permission adjustment. According to historical data analysis and the trainee behavior analysis model, it is set is 0.6. This value is set based on the degree of influence of the trainee's activity level on permission adjustment. Generally speaking, trainees with higher activity levels have a greater weight in permission adjustment. The weight coefficient of the activity frequency affects the contribution of the trainee's frequent activity to permission adjustment; : The weight coefficient of the influence of task completion on permission adjustment, obtained through data analysis, is set to be 0.4, indicating that the influence of task completion on permission adjustment is relatively small, but still of a certain importance. The weight coefficient of task completion affects the weight of the trainee's task completion situation in permission adjustment; : The behavior duration of the trainee (unit: hours). The active duration of the trainee in the system is calculated through the trainee's login records and behavior monitoring system, and is set to be 30 hours, indicating that the trainee has used the system for a total of 30 hours in the recent week. The behavior duration measures the trainee's degree of investment. The longer the active time, the greater the trainee's need for the platform and the more permission support is required; : The adjustment coefficient of the influence of the trainee's behavior duration on permission adjustment, is set to be 0.7, meaning that the influence coefficient of the trainee's active duration on permission adjustment is relatively large. The longer the active time, the greater the weight coefficient, adjusting the influence of the behavior duration on permission adjustment. A relatively high value indicates that the influence of the trainee's behavior duration on permission adjustment is relatively strong; : The learning difficulty level of the trainee (unit: level), set according to the difficulty settings of the trainee's tasks and courses, is set to be level 5, indicating that in the tasks the trainee has learned recently, the average difficulty is level 5. The learning difficulty encountered by the trainee determines the need for functions and permissions. High-difficulty tasks require more permission support to ensure the trainee's learning efficiency; : The adjustment coefficient of the influence of learning difficulty on permission adjustment, is set to be 0.5, indicating that the influence of learning difficulty on permission adjustment is relatively weak, but still needs to be considered, adjusting the influence of learning difficulty on permission adjustment. A relatively low value indicates that the influence of learning difficulty on permission adjustment is relatively small; Substitute the known parameter values, the formula is as follows: ; This result shows that the probability value of permission adjustment is 10.12. This value indicates that based on the current trainee behavior data, the predictability of permission adjustment is very high. The larger the value, the stronger the trainee's need for permission improvement under the current conditions, and it is necessary to optimize and adjust the permissions to improve the trainee's learning experience and efficiency.

[0030] Specifically, as Figure 4 shown, based on the results of behavior analysis and permission adjustment, the steps of refining the processing of trainee behavior data, counting the number of speeches, question frequencies, types of interaction content, and emotional expressions, evaluating learning participation, dividing the activity levels, and generating the trainee activity level are as follows: S301: Based on the results of behavior analysis and permission adjustment, extract the indicators of the number of speeches, question frequencies, types of interaction content, and emotional expressions from the trainee behavior data, classify and count them according to the data types and summarize them to generate the statistical results of trainee behavior data; Group the number of speeches by time period and count the total number of daily speeches and the average speech interval. Summarize the question frequency data according to the time period of each interaction. Screen the trainees with high question frequencies and count the repetition rate of their question content. Classify the types of interaction content by theme and keywords, divide the interaction content into question type, discussion type, and feedback type, and count the proportion of each type of content. At the same time, record the number of active trainees and the main contributors of each type of content. For the emotional expression indicators, through the analysis of keywords in the trainee interaction content, combined with the frequency of modal particles and sentence pattern features, divide the emotions into three categories: positive, neutral, and negative, and count the distribution ratios of the three types of emotions in different interaction scenarios. Classify and organize all the extracted data according to the behavior types, and comprehensively summarize the results to generate the statistical results of trainee behavior data, providing clear basic data support for subsequent analysis.

[0031] S302: Based on the statistical results of trainee behavior data, analyze the correlation between the types of interaction content and the emotional expression indicators, evaluate the activity of trainee learning participation, group the trainee participation levels through classified behavior characteristics, and generate the trainee participation level classification results; By comparing the distribution characteristics of emotional expressions in different types of interaction content, calculate the correlation ratio of positive emotions and question-type content, the correlation ratio of neutral emotions and discussion-type content, and the correlation ratio of negative emotions and feedback-type content, and count the quantity distribution of each correlation relationship to further evaluate the activity of trainee learning participation. Use the number of speeches and question frequencies as the main measurement indicators of trainee participation, and classify and analyze the behavior characteristics of trainees in combination with the types of interaction content and emotional expression indicators. For example, classify trainees with high numbers of speeches, high question frequencies, and a large proportion of positive emotions as "highly active trainees", while classify trainees with few speeches, low question frequencies, and a large proportion of negative emotions as "low-active trainees". According to the above classification results, divide the trainees into multiple levels and divide them into groups according to the specific characteristics of their participation behaviors to generate the trainee participation level classification results, providing data support for subsequent level adjustment and optimization.

[0032] S303: Based on the classification results of the participation levels of the trainees, record the levels of each trainee and their corresponding participation status, check the classification results of active and inactive trainees, mark the status of trainees with different levels, and obtain the activity levels of the trainees. Compare the classification results with the original behavior data to confirm the matching degree between the trainee levels and their participation behaviors. For example, verify whether "highly active trainees" have records of frequent speaking and asking questions. For the classification results of active and inactive trainees, conduct statistics according to the level distribution, and focus on analyzing the differences in the number of trainees and behavior characteristics between different levels. For example, statistically analyze the significant differences in the question-asking frequency and types of interaction content between "moderately active trainees" and "low-active trainees", mark the status of trainees with different levels, record the trainees with cross-level characteristics as objects to be adjusted, and dynamically track the status of the trainees. Through the above comparison and marking, comprehensively obtain the activity levels of the trainees, ensure that the classification results are accurate and can reflect the actual participation of the trainees, and provide basic support for the subsequent optimization of the learning experience.

[0033] Specifically, as Figure 5 shown, according to the activity levels of the trainees, adjust the course interaction session. When the activity level is low, increase Q&A and group discussions, evaluate the participation status and dynamically match the content form, combine open discussions with real-time questions. The specific steps for generating the interactive guidance and course adjustment results are as follows: S401: Based on the activity levels of the trainees, identify the groups of trainees with low activity levels, set interactive tasks for increasing Q&A sessions and group discussions, allocate corresponding interactive tasks according to the activity levels of the trainees, and generate the interactive task allocation results. By screening the behavior data of low-active trainees, determine their specific performances in terms of the number of speeches, question-asking frequency, and interactive emotional expressions, and classify the low-active trainees into three categories: completely non-participating, intermittently participating, and low-frequency participating. For different categories of low-active trainees, set appropriate interactive tasks respectively. For completely non-participating trainees, allocate basic Q&A sessions, and the task content includes answering simple knowledge points and sharing learning experiences. For intermittently participating trainees, allocate group discussion tasks, and the task requires the trainees to complete the topic discussion with group members within a specified time period and submit a summary. For low-frequency participating trainees, allocate interesting Q&A tasks, such as a quiz game format, to enhance the trainees' interest. Allocate corresponding interactive tasks according to the activity levels of the trainees, moderately mix and group active trainees and low-active trainees to enhance the interactive effect, ensure that the task allocation is reasonable and can stimulate the trainees' willingness to participate, generate the interactive task allocation results, and provide a basis for subsequent tracking and evaluation.

[0034] S402: Based on the interactive task assignment results, track in real time the performance of trainees in the Q&A and discussion sessions, record the participation frequency and the number of speeches, analyze the participation data in the interactive session, evaluate the participation effect of trainees and the trend of behavior changes, and generate the evaluation results of trainees' participation effect; Record the participation frequency, the number of speeches and the task completion status of each trainee. For the Q&A session, record in detail the answering time and accuracy rate for each answer, and match the content of the trainee's answer with the question difficulty to identify whether there are participation barriers caused by excessive task difficulty for the trainee. For the group discussion session, record the speaking order of the trainee, the depth and frequency of the contributed content, and the overall cooperation performance of the group. By comparing the task completion degrees of the group members, analyze whether the performance of low-active trainees is driven by high-active trainees. Combine the real-time monitoring data to analyze the change trend of the participation data in the interactive session, such as whether there is a significant increase in the number of speeches of trainees, whether the task completion status tends to be consistent, etc., evaluate the participation effect of trainees and the trend of behavior changes, and generate the evaluation results of trainees' participation effect to provide data support for adjusting the interactive strategy.

[0035] S403: Based on the evaluation results of trainees' participation effect, select a matching content form in combination with the trend of behavior changes, calculate the adjustment coefficient of the trend of trainees' behavior changes, adjust the content forms of open discussion and real-time question and instructor feedback, record the adjustment process and the status of the course form, and obtain the results of interactive guidance and course adjustment; The steps to obtain the adjustment coefficient of the trend of trainees' behavior changes are as follows: ; where represents the adjustment coefficient of the th type of trainees, represents the total number of trainees, represents the eigenvalue of the th trainee in the behavior change, represents the mean value of the th type of trainees in terms of characteristics, represents the standard deviation of the th type of trainees in the behavior change characteristics; Detailed explanation of the formula and the derivation process of the formula calculation: This formula is used to calculate the adjustment coefficient after grouping trainees , and this coefficient determines the intensity of the interactive design and course adjustment for a specific group of trainees. During the calculation process, it is necessary to quantify the specific direction and degree of adjustment based on the behavior characteristics of trainees, the mean value and standard deviation of this group of trainees; : the The adjustment coefficient for a certain type of students is used to determine the adjustment intensity of this group in the course interaction design. The calculation result reflects the relationship between the participation degree of this type of students and the course content adjustment. It is set that ; : The total number of all students, which reflects the size of the sample size and directly affects the representativeness of the clustering calculation. It is set that , which is the data obtained through the monitoring and collection of the participation of all students by the course platform; : The value of a certain feature of the th student in the behavior change. For example, the number of questions asked by the student. This feature is obtained from the behavior data recorded in real time by the course interaction system. It is set that the number of questions asked by a certain student is ; : The mean value of the behavior characteristics of the th type of students, which reflects the overall behavior trend of this group. This mean value is obtained by calculating the average value of the behavior characteristic values of all students within this group. It is set that the mean value is ; : The standard deviation of the behavior characteristics of the th type of students, which reflects the magnitude of the behavior fluctuations within this group. The larger the standard deviation, the greater the behavioral differences of this type of students. This standard deviation is obtained by calculating the behavior characteristic fluctuation values of all students within this group. It is set that ; Substitute the set parameters into the formula to obtain the adjustment coefficient : ; Specifically for a certain student (set as student 1): ; In the calculation of all 100 students, it is set that the average result of this value is ; This result shows that in this student group, the adjustment coefficient of the behavior characteristics is 2.4, indicating that the behavioral differences of this group are relatively large. Therefore, in the course design, greater interaction and content adjustments need to be made to better meet the needs of this group. By analyzing the behaviors of all students, the course designer can optimize the course content, interaction methods, and instructor feedback for different student groups according to this adjustment coefficient.

[0036] Specifically, as Figure 6 shown, according to the interaction guidance and course adjustment results, after the course ends, the steps of summarizing the student behavior data, recording the participation duration, the number of questions asked, and the interaction quality, classifying the student behaviors, and generating the student behavior archives and course interaction summaries are specifically as follows: S501: Based on the interactive guidance and course adjustment results, extract the participation duration, number of questions, and interactive quality data of the trainees. Divide and classify the data by time period, filter out outliers, and adjust the data format to generate the summary result of the trainees' course behavior data. Group the participation data of each trainee by course period, record the specific participation duration within each period, and count the distribution of the total duration and average duration. The number of questions data is statistically grouped by trainee and refined to different course modules, recording the proportion of high-frequency and low-frequency question modules to analyze the question behavior characteristics of the trainees. The interactive quality data is scored based on the depth, breadth, and contribution of the trainees' participation content, and each score data is associated with the participation record of the specific period. For the extracted data, filter out outliers according to statistical principles, such as excluding extreme participation duration data (such as too short or too long period participation records) and repeated or blank questions. When adjusting the data format, convert various data into a unified time period and trainee identification form to ensure the consistency of all participation data in different analysis steps, generating the summary result of the trainees' course behavior data to provide a complete data basis for subsequent behavior analysis.

[0037] S502: Based on the summary result of the trainees' course behavior data, analyze the interaction relationship between the number of questions and participation duration of the trainees, compare the interactive quality score with the behavior performance, classify the trainees into different levels according to their characteristics and label their characteristics to generate the classification result of the trainees' interaction levels. Standardize the number of questions according to the proportion of the course duration to reflect the distribution characteristics of the question behavior in different participation durations. For trainees with significant differences in the number of questions and participation duration data, such as trainees with a high number of questions but a low participation duration or vice versa, label them as special cases and analyze their behavior characteristics. Compare the interactive quality score with the behavior performance of the trainees, and identify trainees who perform outstandingly in high-quality interactions through the correlation between the score and the participation depth, and label these trainees as the high-level interaction group. For trainees with a low score and scattered participation behavior, label them as the low-level interaction group. Based on the above analysis results, classify the trainees into different levels according to their behavior characteristics and record the typical characteristics of each level in detail. For example, high-participation trainees tend to ask questions in multiple modules, while low-participation trainees tend to focus on asking questions in a few modules, generating the classification result of the trainees' interaction levels.

[0038] S503: Based on the classification result of the trainees' interaction levels, record the classified trainees' behavior data, match and integrate the number of questions and interactive quality, summarize the overall course interaction data and file it to obtain the trainees' behavior archives and the summary of the course interaction. Integrate and match the data of the number of questions and interaction quality of each trainee to generate an individual behavior profile. When summarizing the overall interaction data of the course, classify and file the participation records of all trainees according to course modules, time periods, and level categories, and analyze the distribution of the overall interaction behavior in different course modules. For example, count the module with the most questions and the time period with the highest interaction quality. By integrating trainee behavior data and course interaction records, extract key data points, such as the interaction change trend of active trainees after course adjustment and the participation improvement of inactive trainees. Store the archived data in layers to facilitate quick retrieval and reuse for subsequent analysis, ensure the integrity and logical correlation of trainee behavior data and course interaction summaries, obtain trainee behavior profiles and course interaction summaries, and provide data support and direction guidance for course optimization.

[0039] As Figure 7 shown, an online live management system for a teaching platform, the system includes: The trainee identity authentication module, based on the identity information provided by the trainee, checks the mobile phone number and user name provided by the trainee, matches the record fields of the mobile phone number and user name, and checks the integrity and consistency of the identity data item by item to obtain the trainee identity verification result; The basic permission allocation module, based on the trainee identity verification result, filters the trainee behavior records, counts the viewing duration and the number of speeches, determines the permission range that meets the conditions of the behavior records, and obtains the basic permission allocation information; The behavior monitoring and analysis module, based on the basic permission allocation information, records the trainee behavior data, extracts the speech duration and viewing frequency parameters, compares the behavior data with the index standard values, and generates the trainee behavior statistical result; The permission dynamic adjustment module, based on the trainee behavior statistical result, classifies and extracts the behavior frequency data, counts the difference between the parameter and the adjustment threshold, updates the applicable range of the trainee interaction permission, and obtains the permission adjustment information; The trainee activity evaluation module, based on the permission adjustment information, classifies the speech frequency and question number data, statistically calculates the activity level parameters layer by layer, matches the level interval table, and generates the trainee activity level; The course interaction management module, based on the trainee activity level, counts the trainee interaction data, classifies and extracts the content type and participation times, adjusts the frequency and content form of the course interaction session, and generates the trainee behavior profile and the course interaction summary.

[0040] The above is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of changes or substitutions, which should be covered by the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

Claims

1. An online live broadcast management method for a teaching platform, characterized in that, It includes the following steps: Based on the identity information provided by the trainee, check the trainee's mobile phone number, username, and original record, verify the legitimacy of the identity, retrieve past behavior data to determine the permission requirements, classify and allocate basic permissions to the trainee, and generate the trainee identity authentication and basic permission allocation results; Based on the trainee identity authentication and basic permission allocation results, monitor the number of speeches, participation frequency, viewing duration, and discussion activity during the live broadcast, adjust the permissions in combination with real-time data, and generate the behavior analysis and permission adjustment results; Based on the behavior analysis and permission adjustment results, refine the trainee behavior data, count the number of speeches, question frequency, interaction content type, and emotional expression, evaluate the learning participation situation, divide the activity level, and generate the trainee activity level; According to the trainee activity level, adjust the course interaction session, increase Q&A and group discussions when the activity level is low, evaluate the participation situation and dynamically match the content form, combine open discussions and real-time questions, and generate the interaction guidance and course adjustment results; According to the interaction guidance and course adjustment results, summarize the trainee behavior data after the course ends, record the participation duration, number of questions, and interaction quality, classify the trainee behavior, and generate the trainee behavior profile and course interaction summary.

2. The online live broadcast management method of the teaching platform according to claim 1, wherein The trainee identity authentication and basic permission allocation results include the mobile phone number verification status, username matching result, original record verification result, basic permission category, and initial permission configuration. The behavior analysis and permission adjustment results include real-time behavior data records, permission status updates, interaction permission status, and permission dynamic adjustment. The trainee activity level includes the number of speeches, question frequency, interaction content type, emotional expression indicators, and learning participation level. The interaction guidance and course adjustment results include Q&A session settings, group discussion configurations, real-time feedback forms, and interaction content adjustments. The trainee behavior profile and course interaction summary include the trainee participation duration, number of questions, interaction quality, participation classification results, and course data summary.

3. The online live broadcast management method of the teaching platform according to claim 1, characterized in that, The step of, based on the identity information provided by the trainee, checking the trainee's mobile phone number, username, and original record, verifying the legitimacy of the identity, retrieving past behavior data to determine the permission requirements, classifying and allocating basic permissions to the trainee, and generating the trainee identity authentication and basic permission allocation results is specifically as follows: Based on the identity information provided by the trainee, check the mobile phone number, username, and original record provided by the trainee, segmentally extract the core fields of each piece of information, count the proportion and quantity of missing fields by comparing the matching degree of each field, check for field conflicts and record abnormal data, and obtain the trainee identity information check result; Based on the trainee identity information check result, extract the past behavior records associated with the matching fields, identify the behavior characteristics and perform classification label marking by screening the parameters of login behaviors, access records, and operation logs, judge whether the permission conditions are met, and generate the trainee permission attribute classification result; Based on the classification results of the student permission attributes, filter and allocate content according to the permission conditions marked in the classification tags. By checking the levels of viewing, speaking, and interaction permissions in the permission allocation records and combining the key feature values in the behavior data, verify the allocation scope to obtain the student identity authentication and basic permission allocation results.

4. The online live broadcast management method of the teaching platform according to claim 1, wherein, Based on the student identity authentication and basic permission allocation results, the steps for monitoring the number of speeches, participation frequency, viewing duration, and discussion activity during the live broadcast and adjusting permissions in combination with real-time data to generate the behavior analysis and permission adjustment results are as follows: Based on the student identity authentication and basic permission allocation results, extract the behavior data of the number of speeches, count the total amount of data and mark the abnormal items to obtain the monitoring results of the students' real-time behavior data; Based on the monitoring results of the students' real-time behavior data, analyze the fluctuation trend of the behavior data, screen the abnormal data segments and mark the potential adjustment items to obtain the reference data for students' permission adjustment; Based on the reference data for students' permission adjustment, use the behavior prediction algorithm to check the trigger conditions of the permission adjustment items, update the permission status and record the adjustment time and scope to obtain the behavior analysis and permission adjustment results.

5. The online live broadcast management method of the teaching platform according to claim 4, characterized in that, The formula of the behavior prediction algorithm is as follows: ； Among them, represents the probability value of permission adjustment, represents the activity frequency of the trainee in the recent week, represents the task completion rate of the trainee recently, represents the number of unfinished tasks of the trainee, represents the number of times the trainee's tasks are postponed, represents the weight coefficient of the influence of activity frequency on permission adjustment, represents the weight coefficient of the influence of task completion rate on permission adjustment, represents the behavior duration of the trainee, represents the adjustment coefficient of the influence of the trainee's behavior duration on permission adjustment, represents the learning difficulty level of the trainee in the recent month, represents the adjustment coefficient of the influence of learning difficulty on permission adjustment.

6. The online live broadcast management method of the teaching platform according to claim 1, characterized in that Based on the behavior analysis and permission adjustment results, refine the processing of the students' behavior data, count the number of speeches, question frequency, interaction content type, and emotional expression, evaluate the learning participation, divide the activity levels, and the steps for generating the students' activity levels are as follows: Based on the behavior analysis and permission adjustment results, extract the indicators of the number of speeches, question frequency, interaction content type, and emotional expression in the students' behavior data, classify and count according to the data type and summarize to generate the statistical results of the students' behavior data; Based on the statistical results of the students' behavior data, analyze the correlation between the interaction content type and the emotional expression indicators, evaluate the activity of the students' learning participation, and group the students' participation levels through classified behavior characteristics to generate the results of the students' participation level classification; Based on the results of the students' participation level classification, record the level and corresponding participation of each student, check the classification results of active and inactive students, and mark the status of students with different levels to obtain the students' activity levels.

7. The online live broadcast management method of the teaching platform according to claim 1, wherein According to the students' activity levels, adjust the course interaction session, increase Q&A and group discussions when the activity level is low, evaluate the participation and dynamically match the content form, combine open discussions and real-time questions, and the steps for generating the interaction guidance and course adjustment results are as follows: Based on the students' activity levels, identify the groups of students with low activity levels, set the interaction tasks of increasing Q&A sessions and group discussions, and allocate the corresponding interaction tasks according to the students' activity levels to generate the interaction task allocation results; Based on the interaction task allocation results, track the performance of the students' participation in the Q&A and discussion sessions in real time, record the participation frequency and the number of speeches, analyze the participation data in the interaction session, evaluate the participation effect of the students and the trend of behavior changes to generate the evaluation results of the students' participation effect; Based on the evaluation results of the participation effects of the trainees, select the matching content form in combination with the trend of behavior changes, calculate the adjustment coefficient of the trend of trainees' behavior changes, adjust the content forms of open discussions and real-time questions and instructor feedback, record the adjustment process and the status of the course form, and obtain the results of interaction guidance and course adjustment.

8. The online live broadcast management method of the teaching platform according to claim 7, characterized in that, The steps for obtaining the adjustment coefficient of the trend of the trainees' behavior changes are as follows: ； Among them, represents the adjustment coefficient of the th type of trainee, represents the total number of trainees, represents the th trainee's eigenvalue in behavioral changes, represents the th type of trainee's mean value in characteristics, represents the th type of trainee's standard deviation in behavioral change characteristics.

9. The online live broadcast management system of the teaching platform according to claim 1, characterized in that, According to the results of the interaction guidance and course adjustment, after the course ends, summarize the trainees' behavior data, record the participation duration, the number of questions, and the interaction quality, and the steps for classifying the trainees' behavior and generating the trainees' behavior archives and the summary of course interactions are specifically as follows: Based on the results of the interaction guidance and course adjustment, extract the data of the participation duration, the number of questions, and the interaction quality of the trainees, divide and classify the data by time period, screen out the outliers and adjust the data format, and generate the summary result of the trainees' course behavior data; Based on the summary result of the trainees' course behavior data, analyze the interaction relationship between the number of questions and the participation duration of the trainees, compare the interaction quality score with the behavior performance, classify the trainees into different levels according to their characteristics and label the characteristics, and generate the result of the classification of the trainees' interaction levels; Based on the result of the classification of the trainees' interaction levels, record the classified trainees' behavior data, match and integrate the number of questions and the interaction quality, summarize the overall interaction data of the course and file it, and obtain the trainees' behavior archives and the summary of course interactions.

10. An online live broadcast management system for a teaching platform, characterized in that, According to the online live broadcast management method of the teaching platform according to any one of claims 1-9, the system includes: The trainee identity authentication module, based on the identity information provided by the trainee, checks the mobile phone number and user name provided by the trainee, matches the record fields of the mobile phone number and the user name, and checks the integrity and consistency of the identity data item by item to obtain the trainee identity verification result; The basic permission allocation module, based on the trainee identity verification result, screens the trainee behavior records, counts the viewing duration and the number of speeches, judges the scope of permissions that the behavior records meet the conditions, and obtains the basic permission allocation information; The behavior monitoring and analysis module, based on the basic permission allocation information, records the trainee behavior data, extracts the speech duration and viewing frequency parameters, compares the behavior data with the standard values of the indicators, and generates the trainee behavior statistical result; The permission dynamic adjustment module, based on the trainee behavior statistical result, classifies and extracts the behavior frequency data, counts the difference between the parameters and the adjustment threshold, updates the applicable range of the trainee interaction permissions, and obtains the permission adjustment information; The trainee activity evaluation module, based on the permission adjustment information, classifies the speech frequency and number of questions data, statistically analyzes the activity level parameters layer by layer, matches the level interval table, and generates the trainee activity level; The course interaction management module, based on the trainee activity level, statistically analyzes the trainee interaction data, classifies and extracts the content type and the number of participations, adjusts the frequency and content form of the course interaction session, and generates the trainee behavior archives and the summary of course interactions.

Images