Network teaching interaction auxiliary management system based on Internet

Abstract

The invention belongs to the technical field of teaching management, and particularly discloses and provides an internet-based network teaching interaction auxiliary management system, which comprises a student portrait construction module, a student portrait updating module, a real-time interaction monitoring module, an interaction management evaluation module and a student teaching management terminal. According to the invention, by integrating the functions of multiple modules such as student portrait construction, teaching recommendation, portrait updating, real-time interaction monitoring and evaluation and the teaching management terminal, the learning condition and interaction performance of the student can be accurately grasped, and the personalized recommendation of the teaching course and the dynamic updating of the learning portrait based on the individual characteristics of the student are realized; meanwhile, a corresponding interaction management strategy is flexibly triggered and implemented according to the interaction monitoring data, so that the teaching pertinence and the interaction effectiveness are effectively improved, the teaching quality is improved, the learning experience of students is optimized, the interaction experience of the students in courses is guaranteed, and the overall effect and adaptability of network teaching are enhanced.

Description

Technical Field

[0001] The invention belongs to the technical field of teaching management, and in particular relates to an Internet-based network teaching interactive auxiliary management system. Background Art

[0002] The traditional teaching model is restricted by multiple factors such as time, location, and teaching materials, and it is difficult to meet the diverse learning needs of students. At the same time, the interaction between teachers and students is limited, making it difficult to form an effective learning community. The network-based teaching management platform can break these limitations, provide a more flexible and convenient learning method, promote communication and interaction between teachers and students, thereby stimulating students' interest in learning and improving their independent learning ability.

[0003] Prior art, such as the Chinese invention patent application with application number 202410239181.7, discloses an interactive teaching data management system and method based on cloud computing, which includes: an interactive teaching monitoring module, a teaching data collection module, a cloud platform management center, a copying progress analysis module and a teaching data management module. The calligraphy teaching process is monitored in real time through the interactive teaching monitoring module, and the student copying information and student location information in the calligraphy teaching process are collected through the teaching data collection module. All collected data are stored and managed through the cloud platform management center, and the student copying progress is analyzed through the copying progress analysis module. The degree of difference in the student copying progress is analyzed through the teaching data management module, and the data transmission of the student copying images is managed, and the data transmission method is flexibly adjusted to help teachers pay attention to students whose copying progress is relatively backward in time and provide assistance, thereby balancing the overall copying level of students in the calligraphy teaching process and improving the teaching quality.

[0004] Another example of the prior art is an AI intelligent interactive course system and method based on a large language model disclosed in the Chinese invention patent application with application number 202311010600.1, which includes a user end, a knowledge base, a course module, a learning management module, an intelligent education auxiliary module, an intelligent education review module, an intelligent creation auxiliary module and an AI digital human teaching module. The knowledge base, course module, learning management module, intelligent education auxiliary module, intelligent education review module and intelligent creation auxiliary module all provide teaching data support for the AI ​​digital human teaching module. It is an AI intelligent interactive course system based on a large language model that integrates intelligent education assistance, intelligent education review, and intelligent creation assistance.

[0005] With regard to the above technical solutions, it is obvious that the interactive auxiliary management of online teaching has optimized the interaction to a certain extent. However, the current interactive auxiliary management of online teaching still has the following deficiencies: 1. Although it is currently possible to provide personalized learning paths and content recommendations based on students' learning data, such recommendations are often not accurate and detailed enough, and the degree of personalization needs to be improved. In addition, no tracking strategy has been established, which makes it difficult to make subsequent personalized adjustments, and thus cannot further meet the personalized teaching needs of different students.

[0006] 2. Interactive messages have not been optimized. When multiple people participate in real-time interactive scenarios such as online discussions and collaborative learning at the same time, the smoothness of the interaction cannot be guaranteed. At the same time, messages in the interactive process are prone to being disordered, lost, or received repeatedly, making it impossible to smoothly promote interactive communication, which greatly undermines the interactive experience. Summary of the invention

[0007] In view of this, in order to solve the problems raised in the above background technology, an Internet-based network teaching interactive auxiliary management system is now proposed.

[0008] The purpose of the present invention can be achieved through the following technical solutions: The present invention provides an Internet-based online teaching interactive auxiliary management system, which includes: a student portrait construction module, which is used to collect the accumulated learning data of each student corresponding to the current teaching platform and construct a learning portrait of each student.

[0009] The student portrait update module is used to track the learning footprints of each student in each recommended teaching course within each recommendation cycle, and conduct learning portrait update analysis based on the accumulated learning data and recommended learning footprints to obtain the appropriate update frequency of the corresponding learning portrait of each student.

[0010] The real-time interaction monitoring module is used to monitor the number of interactive students participating in the current course, the number of online students and interaction record data.

[0011] The interactive management assessment module is used to determine whether interactive management is triggered. If triggered, the interactive management indicators are confirmed; otherwise, the pre-set interactive management indicators are maintained.

[0012] The student teaching management terminal is used to update the learning portrait based on the appropriate update frequency of each student's corresponding learning portrait, and to manage teaching interactions based on interaction management indicators.

[0013] Compared with the prior art, the beneficial effects of the present invention are as follows: (1) The present invention can accurately grasp the students' learning status and interactive performance by integrating multiple module functions such as student portrait construction, teaching recommendation, portrait update, real-time interactive monitoring and evaluation, and teaching management terminal, and realize personalized recommendation of teaching courses and dynamic update of learning portraits based on individual characteristics of students. At the same time, it can flexibly trigger and implement corresponding interactive management strategies based on interactive monitoring data, effectively improve the pertinence of teaching and the effectiveness of interaction, promote the improvement of teaching quality and optimization of students' learning experience, and enhance the overall effect and adaptability of online teaching.

[0014] (2) The present invention constructs a learning radar chart for each student by analyzing the learning participation of the recommended courses, the consistency of the recommended independent participation, and the interest level of each course type. This chart serves as a learning portrait for the student, providing a more reliable and accurate data reference for subsequent personalized learning path planning and learning resource allocation. This effectively solves the problem that current course recommendations are not precise and detailed enough, and further improves the personalization of course recommendations.

[0015] (3) By analyzing the degree of consistency between the recommendations and the autonomous participation of each student, the present invention can comprehensively and specifically calculate the consistency between the recommended courses and the self-selected courses of each student, thereby providing strong data support for understanding the learning preferences of students, optimizing learning recommendations, and adjusting teaching strategies.

[0016] (4) The present invention tracks the learning data of each student on the recommended teaching courses and conducts learning portrait update analysis, which effectively makes up for the current deficiency of not establishing a tracking strategy, provides a reliable basis for subsequent personalized adjustments, and facilitates further meeting the personalized teaching needs of different students.

[0017] (5) The present invention effectively solves the problem of the current lack of interactive message optimization by performing interactive management trigger judgment, confirming interactive management items, and confirming management indicators under interactive management items. It can ensure the smoothness of interaction when multiple people simultaneously participate in real-time interactive scenarios such as online discussions and collaborative learning, and avoid the occurrence of problems such as disordered, lost, or repeated reception of messages during the interactive process, thereby facilitating the smooth promotion of interactive communication and ensuring the interactive experience of students in the course. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for describing the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0019] Figure 1 It is a schematic diagram of the connection of each module of the system of the present invention.

[0020] Figure 2 It is a schematic flow chart of the overall implementation steps of the present invention.

[0021] Figure 3 It is a flow chart of the interactive management assessment module of the present invention.

[0022] Figure 4 This is a schematic diagram of the interactive management judgment process of triggering the present invention. DETAILED DESCRIPTION

[0023] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0024] See also Figure 1 and Figure 2 As shown, the present invention provides an Internet-based network teaching interactive auxiliary management system, which includes: a student portrait construction module, a student portrait update module, a real-time interactive monitoring module, an interactive management evaluation module and a student teaching management terminal.

[0025] In the above, the student portrait updating module is connected to the student portrait constructing module and the student teaching management terminal respectively, and the interactive management evaluation module is connected to the real-time interactive monitoring module and the student teaching management terminal respectively.

[0026] The student portrait construction module is used to collect the accumulated learning data of each student corresponding to the current teaching platform and construct a learning portrait of each student.

[0027] Specifically, the accumulated learning data includes but is not limited to the marked source tags of each course, the course type of each course, the setting information and the learning footprint, wherein the setting information includes the set number of class hours, the number of interactive nodes and the set course content, etc., and the learning footprint records the number of learning class hours, the last learning date and the number of interactive nodes participated in, etc.

[0028] It should be added that the data source of the current teaching platform's cumulative learning data for each student is mainly the teaching platform database, which contains all the students' learning behavior records, personal information, and relevant data on participation in interactions on the platform. It can also be provided by external systems with authorization, such as other education platform data related to the student's previous learning experience, or industry databases related to the student's occupation.

[0029] In yet another specific aspect, the constructing of the learning profile of each student includes: A1, extracting the marked source label of each course from the accumulated learning data of each student, wherein the marked source label includes recommended learning and self-selected learning.

[0030] A2. Record each course marked with a source tag as recommended learning as a recommended course, and count each student's participation in the recommended course.

[0031] A3. Extract the course type, setting information and learning footprint of each course from the accumulated learning data of each student, and calculate the recommended autonomous participation consistency of each student.

[0032] A4. Integrate all courses under the same course type to obtain courses under each course type corresponding to each student.

[0033] A5. Extract the set number of class hours from the setting information, and extract the number of learning class hours and the last learning date from the learning footprint.

[0034] A6. Compare the number of learning hours with the number of set hours, record the ratio as the completion ratio, and calculate the average course completion ratio k of each student for each course type by the mean. ij , i represents the student number, i=1,2,......n, j represents the course type number, j=1,2,......m.

[0035] A7. Set the interest correction factor μ for each student for each course type ij , count the interest of each student in each course type β ij , u represents the number of courses, e is a natural constant, and k′ is the completion ratio of the set reference.

[0036] A8. Use the recommended course participation, recommended autonomous participation consistency and interest in each course type as portrait elements to construct a learning radar chart for each student as their learning portrait.

[0037] Understandably, self-selected learning refers to courses selected by students themselves through search and other means, and recommended learning refers to learning recommended to students by the platform and in which students participate.

[0038] Understandably, in order to more intuitively display the learning portrait of students, visualization technology is used to present the portrait results. Charts such as radar charts can be used to display the characteristic distribution and proportional relationship of students in different dimensions such as learning behavior, preference, and interaction. For example, a radar chart is used to display the performance of students in multiple characteristics such as recommended course participation, recommended autonomous participation consistency, and interest in each course type. The shape and area of ​​the radar chart can be used to intuitively compare the differences in learning portraits between different students, helping teachers and teaching management personnel to better understand the learning characteristics and needs of students, thereby providing a basis for personalized teaching.

[0039] It should be added that the completion ratio of the reference can be set to a value of 0.8.

[0040] It should also be added that the construction of each student's learning radar chart starts from the center of the radar chart. The axes extending outward represent different portrait elements, and the scales on the axes correspond to the quantitative coding values. The radar chart is used to show the characteristics of students in terms of learning participation in recommended courses, consistency with recommended independent participation, and interest in each course type. The shape of the radar chart can be used to intuitively compare the differences in learning portraits between different students. Teachers can quickly identify students' learning characteristics and tendencies, thereby providing a basis for personalized teaching.

[0041] In another specific embodiment, the learning portrait can also be a labeled description, such as "active interactive learner", "in-depth scholar", "practice-oriented student", etc., which can be matched and compared with the numerical range corresponding to the pre-set label based on the above-mentioned statistics of recommended course learning participation, recommended autonomous participation consistency and interest in each course type, and then the label can be set.

[0042] The embodiment of the present invention analyzes the learning participation of recommended courses, the consistency of recommended independent participation and the interest of each course type, and then constructs a learning radar chart for each student, which serves as their learning portrait, providing more reliable and accurate data reference for subsequent personalized learning path planning and learning resource allocation, effectively solving the problem that current course recommendations are not accurate and detailed enough, and further improving the personalization of course recommendations.

[0043] Furthermore, in step A2, the participation of each student in the recommended course is counted, including: A21, extracting the number of set interactive nodes from the setting information, and extracting the number of participating interactive nodes from the learning footprint, and recording the ratio of the two as the interaction ratio.

[0044] A22. Count the number of recommended courses for each student whose corresponding completion ratio is greater than the set reference completion ratio, and divide it by the total number of recommended courses for the corresponding student to obtain the recommended learning achievement ratio of each student.

[0045] A23. Count the number of recommended courses for each student whose corresponding interaction ratio is greater than the set reference interaction ratio, and divide it by the total number of recommended courses for the corresponding student to obtain the recommended interaction achievement ratio of each student.

[0046] A24. Set the weights of the recommended learning achievement ratio and the recommended interaction achievement ratio, and calculate the recommended course participation of each student through weighted average.

[0047] In a specific embodiment, the course content is usually designed according to a certain logical order and knowledge architecture. For example, in a programming course, from basic data types and variable definitions to complex algorithms and data structures, each part is progressive. The students' learning of the learning content also directly reflects their attention to the course. Therefore, the weight of the recommended learning achievement ratio is set to be greater than the weight of the recommended interaction achievement ratio. For example, the weights of the recommended learning achievement ratio and the recommended interaction achievement ratio can be 0.55 and 0.45 respectively.

[0048] Furthermore, in step A3, the degree of consistency of each student's recommended autonomous participation is counted, including: A31, extracting the set course content from the setting information, and recording each course marked with a source label as autonomously selected learning as each main course.

[0049] A32. Based on the set course content, the similarity of the course content of each student's main course and each recommended course is calculated by a similarity measurement algorithm.

[0050] A33. If the similarity between a self-directed course and a recommended course exceeds the set value, the recommended course will be used as a similar recommended course for the self-directed course.

[0051] A34. Count the number of similar recommended courses for each student's main course, take the maximum number and divide it by the total number of corresponding recommended courses to obtain the recommended independent content matching ratio of each student.

[0052] A35. For each student, count the number of courses in which they participate in independent learning that are of the same type as the recommended courses, and divide it by the total number of recommended courses to obtain the matching ratio of the recommended independent course type for each student.

[0053] A36. According to the statistical method of recommended course participation, the autonomous course participation of each student is statistically obtained in the same way. The recommended course participation and autonomous course participation of each student are respectively denoted as δ i and δ i ′, calculate the matching ratio λ of each student’s recommended autonomous participation degree i ,

[0054] A37. Set weights for the recommended autonomous content match ratio, recommended autonomous course type match ratio, and recommended autonomous participation degree match ratio, and calculate each student's recommended autonomous participation match ratio through weighted average.

[0055] The similarity measurement algorithms described in step A32 include but are not limited to the cosine similarity algorithm, the edit distance algorithm, and the Jaccard similarity algorithm, etc. The selected similarity measurement algorithm is used to perform pairwise comparison calculations on the contents of each student's independent course and recommended course. For example, student A has a self-taught basic painting course and three recommended painting-related courses. It is necessary to calculate the content similarity between the independent course and the three recommended courses respectively. Among them, the reference course content similarity can be set to a value of 0.8. For example, if the similarity value calculated by the cosine similarity algorithm to a recommended course is 0.8, it means that the two courses have a high degree of similarity in content, and there may be many overlapping knowledge points and teaching contents.

[0056] In a specific embodiment, if the learning scenario is mainly based on the construction of a knowledge system, that is, the focus of learning is on the coherence and relevance of knowledge content, the weight of the recommended autonomous content matching ratio can be relatively high. For example, in a highly professional academic course system, such as computer science course learning, the systematic nature of knowledge is critical. If the course content of the student's autonomous learning, such as machine learning algorithms, is highly related to the recommended course content such as deep learning frameworks, then the content matching ratio weight can be set to the highest, and illustratively, it can be taken as 0.4, because the close matching of content helps students build a complete knowledge system and avoid knowledge gaps. The weights of the recommended autonomous course type matching ratio and the recommended autonomous participation degree matching ratio can be taken as 0.3 respectively. If the learning scenario is mainly based on vocational skills training, in vocational skills training, the actual participation and learning effect of the students are directly related to whether they can master vocational skills, so the participation degree matching ratio weight is higher. The matching of content and course type is also important, but relatively speaking, it is slightly less important in this scenario. Exemplarily, the weights of the recommended autonomous content matching ratio, the recommended autonomous course type matching ratio, and the recommended autonomous participation degree matching ratio may be set to 0.3, 0.3, and 0.4, respectively.

[0057] Furthermore, step A7 sets an interest correction factor for each student corresponding to each course type, including: A71, comparing the last study date with the current date to obtain the last study interval duration, and comparing the study interval duration of each course under each course type for each student with the set reference time window.

[0058] A72. If the learning interval of a course exceeds the set reference time window and the completion ratio of the course is lower than the set reference completion ratio, the course will be recorded as an interrupted course. The number of interrupted courses of each student in each course type will be counted and divided by the total number of courses. The ratio will be used as the interest correction factor for each student in each course type.

[0059] It should be added that the purpose of setting the interest correction factor for each student for each course type is to more accurately measure the student's real interest in each course type. The simple completion ratio may overestimate the student's interest, because some courses may be started by students for some reason, but then not continued. For example, a student may start learning a programming language course because of course promotion, but lose interest and give up halfway. By considering the learning interval and the completion ratio to determine the interrupted course, it can more accurately reflect the student's current interest in the course type. At the same time, learning interest changes dynamically, and the interest correction factor can capture this change. Over time, the student's interest may shift or disappear. Setting this factor can timely discover which course types the student's interest has weakened, providing a basis for subsequent learning recommendations or teaching strategy adjustments. For example, a student was very interested in history courses before, but has not studied history-related courses for a long time recently, and the completion rate of previous courses is not high, which indicates that his interest in history courses may have decreased, and this change can be reflected through the interest correction factor.

[0060] Understandably, if we know that students’ interest in certain types of courses has declined, we can reduce recommendations or resource investment in these areas, and direct more resources to the types of courses that students are really interested in. For example, on an online learning platform, based on the results of the interest correction factor, we can reduce the number of math course advertisements pushed to students whose interest in math courses has declined, and instead recommend science experiment courses that they may be interested in.

[0061] In a specific embodiment, the learning interval is obtained by comparing the last learning date with the current date, and the interrupted course is judged in combination with the completion ratio. This method comprehensively considers two important factors: time and learning progress. It does not rely solely on a single indicator, avoiding one-sidedness. For example, only looking at the completion ratio may ignore the situation that the student has completed part of the course but has not continued to study for a long time. And only looking at the learning interval may misjudge that the courses that are temporarily interrupted but have a high completion rate are not of interest to the students. The interest correction factor is obtained by comparing the number of interrupted courses with the number of courses, and the degree of interest change is quantified. This makes it possible to compare the interest changes between different students or the same student in different course types. For example, 30% of the courses of student A in the literature course type are interrupted courses, and 60% of the courses of student B in the literature course type are interrupted courses. It can be intuitively seen that the degree of interest decline of student B in the literature course is more serious than that of student A. And this setting method can flexibly adapt to the characteristics of various course types. Different courses have different learning cycles, difficulty and other factors. By comparing with the set reference time window and completion ratio, the interest changes of students can be judged according to the characteristics of the course itself. For example, for short-term skill training courses, the reference time window can be set shorter. For long-term academic research courses, the time window can be appropriately extended to more reasonably measure students' interest in different course types. For example, for short-term skill training courses, the reference time window can be set to one week, and for long-term academic research courses, the time window can be set to one month.

[0062] The student portrait update module is used to track the learning footprints of each student in each recommended teaching course within each recommendation cycle, and perform learning portrait update analysis based on the accumulated learning data and recommended learning footprints to obtain the appropriate update frequency of the learning portrait corresponding to each student.

[0063] Specifically, the learning portrait update analysis is performed, including: B1. Based on the learning footprints of each student in each recommended teaching course in each recommendation cycle, the recommended course participation of each student in each recommendation cycle is obtained by statistically analyzing the statistical method of the recommended course participation of each student.

[0064] B2. With the recommendation period as the horizontal axis and the recommended course participation as the vertical axis, a recommended course participation change curve for each student is constructed, and the slope of the curve is extracted as the learning participation change rate, recorded as (k c ) i .

[0065] B3. Count the number of recommended cycles in which the participation rate of each student in the recommended course is greater than the set reference participation rate, and divide it by the total number of recommended cycles to obtain the matching ratio of each student's recommended course learning participation cycle, recorded as (k b )i .

[0066] B4. Count the appropriate update frequency p of each student's learning profile i , k c ′ and k b ′ represents the learning participation change rate and participation cycle matching ratio of the set reference respectively, and p0 is the set automatic update frequency of the initial learning profile.

[0067] It should be added that, in a specific embodiment, k c ′ can be set to -0.1, k b ′ can be set to 0.3, and the specific value of p0 is the default value of the initial setting of the teaching platform, which is provided by the teaching platform. Depend on, The integration is obtained, among which, It represents the update compensation factor. When the slope of the curve is smaller, it indicates that with the change of the recommendation cycle, the participation in the recommended courses gradually decreases, that is, the attention to the recommended courses is also getting lower and lower. When the attention is getting lower and lower, the update frequency needs to be appropriately increased in order to better control the courses that students are interested in.

[0068] The embodiment of the present invention tracks the learning data of each student for the recommended teaching courses and performs learning portrait update analysis, which effectively makes up for the current deficiency of not establishing a tracking strategy, provides a reliable basis for subsequent personalized adjustments, and facilitates further meeting the personalized teaching needs of different students.

[0069] The real-time interaction monitoring module is used to monitor the number of interactive students participating in the current course, the number of online students and interaction record data.

[0070] Specifically, the interaction record data records the cumulative number of interaction IDs, the number of interaction conversations, the initiating interaction ID of each interaction conversation, the conversation content and initiation time, and the identity tag of each interaction ID.

[0071] See also Figure 3 As shown, the interaction management evaluation module is used to determine whether to trigger interaction management. If triggered, the interaction management index is confirmed, otherwise the preset interaction management index is maintained.

[0072] Specifically, see Figure 4 As shown, judging whether to trigger interaction management includes: E1. comparing the number of interactive students participating in the current online course with the number of online students to obtain an online interaction ratio.

[0073] E2. Extract the cumulative number of interaction IDs, the number of interaction conversations, and the initiation interaction ID, conversation content, and initiation time of each interaction conversation from the interaction record data, and calculate the online interaction frequency index.

[0074] E3. The online interaction ratio greater than the set reference interaction ratio is defined as judgment condition 1, and the interaction square degree greater than the set reference interaction frequency index is defined as judgment condition 2.

[0075] E4. Determine whether the above judgment conditions are met. If so, it will be triggered as the judgment result, otherwise it will not be triggered as the judgment result.

[0076] Furthermore, in step E2, the online interaction frequency index is counted, including: E21, based on the initiation time of each interactive dialogue, the starting initiation time and the final initiation time are extracted, and the interval between the two is used as the interaction time. The number of interactive dialogues is divided by the interaction time to obtain the interaction frequency, which is recorded as f.

[0077] E22. Compare all interactive dialogues in pairs and calculate the content similarity between two dialogues using a similarity measurement algorithm. When the content similarity between two interactive dialogues is greater than or equal to a set threshold, the two interactive dialogues are recorded as repeated interactive dialogue pairs, and the number of repeated dialogue pairs is counted, which is recorded as M.

[0078] E23. Divide the number of interactive conversations by the number of interactive IDs recorded cumulatively to obtain the number of interactive conversations for a single interactive ID, recorded as D, and the number of interactive conversations as T;

[0079] E24. Calculate the online interaction frequency index γ, r1, r2 and r3 represent the weights corresponding to the interaction frequency, the number of repeated conversations and the number of interactive conversations for a single interactive ID, respectively. f′ and D′ represent the interaction frequency and the number of single-person interactive conversations for the set reference, respectively.

[0080] It should be added that the threshold value of the content similarity of two interactive conversations can be specifically set to 0.7, and this value can be determined based on actual testing and adjustment. When the content similarity of two interactive conversations is greater than or equal to this threshold, they can be identified as repeated interactive conversations.

[0081] It should be added that when there are a large number of repeated interactions, this to a certain extent indicates that the students have in-depth thinking and continuous attention to a specific topic or knowledge point. For example, in an online programming course forum, students repeatedly discussed the topic of "How to optimize the performance of Python code". They may continue to share their own code optimization experience, propose new optimization ideas, or repeatedly ask and answer other people's opinions. This large number of repeated interactions shows the students' deep involvement in this specific programming problem, and also shows the urgency and urgency of the interactive processing.

[0082] It should be added that by comprehensively considering factors such as interaction frequency, repeated conversations, and the number of interactive conversations for a single interactive ID, the frequency of online interactions can be more comprehensively reflected. The interaction frequency can reflect the students' real-time participation enthusiasm in the course learning process and has the highest weight. Repeated conversations may mean that students are confused or interested in specific knowledge points and have the second highest weight. The number of interactive conversations for a single interactive ID can reflect the individual participation of students. However, in comparison, the course pays more attention to the creation of an overall interactive atmosphere and achieving teaching goals through various forms of interaction, so its weight is relatively lower. For example, r1, r2, and r3 can be taken as 0.45, 0.35, and 0.2, respectively.

[0083] In a specific embodiment, the specific values ​​of f′ and D′ need to be set according to the interaction duration. Assuming that the interaction duration is half an hour, f′ can be set to 5 minutes / item, and D′ can be set to 6 items.

[0084] In another specific embodiment, the interactive management indicator is confirmed, including: J1. If only judgment condition 1 is established, the interactive partition is used as a management item; if judgment condition 2 is established, the message queue management is used as a management item.

[0085] J2. If the management item is interactive partitioning, each interactive conversation is allocated based on a preset load balancing algorithm, and the backend server corresponding to each interaction is obtained and used as an interactive management indicator.

[0086] J3. If the management item is message queue management, extract the identity tag of each interaction ID from the interaction record data, set the processing priority coefficient of each interaction ID, and sort the interaction IDs from large to small according to their processing priority coefficients as the message processing order of each interaction ID and as the interaction management indicator.

[0087] In a specific embodiment, the pre-set load balancing algorithm includes but is not limited to a polling algorithm, a weighted polling algorithm, a least connection algorithm, and an IP hash-based algorithm. For example, the allocation of each interactive conversation based on the pre-set load balancing algorithm is specifically referred to as follows: 1) Polling algorithm: The interactive requests are allocated to each server in the back-end server cluster in sequence.

[0088] 2) Weighted round-robin algorithm: Different weights are assigned to each server based on the performance differences of the servers, such as the number of CPU cores, memory size, etc. Servers with stronger performance will be allocated more requests.

[0089] 3) Least Connection Algorithm: F5 monitors the current number of connections of each server in real time, and always distributes new interactive requests to the server with the least number of connections to ensure that the load of each server is relatively balanced.

[0090] 4) You can also use an IP hash-based algorithm to calculate a hash value based on the client's IP address, and then assign the request to a certain server based on this hash value. This algorithm is suitable for scenarios where you need to maintain session consistency between the client and the server. For example, in some interactive applications that require a login status, it can ensure that multiple requests from the same client are routed to the same server for processing.

[0091] Furthermore, the processing priority coefficient of each interactive ID is set, including: the processing priority coefficient of the interactive ID with the identity tag of teacher is recorded as 1.

[0092] The interaction ID whose identity tag is not that of the teacher is extracted and recorded as the target ID. Based on the initiating interaction ID of each interactive dialogue, the interactions under the target ID are filtered out.

[0093] Based on the content and initiation time of each interactive conversation under the target ID, the interaction frequency and number of repeated conversation pairs of the target ID are counted, denoted as f″ and M′ respectively. As the processing priority coefficient of the target ID, denoted as ε, T represents the number of interactive conversations, and the processing priority coefficient μ of each interactive ID is obtained by setting d , μ d The value is 1 or ε, d represents the interaction ID number, d = 1, 2, ... u.

[0094] The embodiment of the present invention effectively solves the problem of no current interactive message optimization by performing interactive management trigger judgment, confirming interactive management items, and confirming management indicators under interactive management items. It can ensure the smoothness of interaction when multiple people simultaneously participate in real-time interactive scenarios such as online discussions and collaborative learning, and avoid the occurrence of problems such as disordered, lost, or repeated reception of messages during the interactive process, thereby facilitating the smooth promotion of interactive communication and ensuring the interactive experience of students in the course.

[0095] The student teaching management terminal is used to update the learning portrait based on the appropriate update frequency of the learning portrait corresponding to each student, and to manage the teaching interaction based on the interaction management index.

[0096] The embodiment of the present invention integrates multiple module functions such as student portrait construction, teaching recommendation, portrait update, real-time interaction monitoring and evaluation, and teaching management terminal, so as to accurately grasp the student's learning status and interaction performance, realize personalized recommendation of teaching courses and dynamic update of learning portraits based on individual characteristics of students, and flexibly trigger and implement corresponding interaction management strategies according to interaction monitoring data, effectively improve the teaching pertinence and interaction effectiveness, promote the improvement of teaching quality and optimization of student learning experience, and enhance the overall effect and adaptability of online teaching.

[0097] The above contents are merely examples and explanations of the concept of the present invention. Those skilled in the art may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined by the present invention, they shall all fall within the protection scope of the present invention.

Claims

1. An Internet-based network teaching interactive auxiliary management system, characterized in that: The system includes: The student portrait building module is used to collect the accumulated learning data of each student on the current teaching platform and build a learning portrait for each student; The student portrait update module is used to track the learning footprints of each student in each recommended teaching course in each recommendation cycle, and conduct learning portrait update analysis based on the accumulated learning data and recommended learning footprints to obtain the appropriate update frequency of each student's corresponding learning portrait; Real-time interaction monitoring module, used to monitor the number of interactive students participating in the current course, the number of online students and interaction record data; The interactive management evaluation module is used to determine whether interactive management is triggered. If triggered, the interactive management indicators are confirmed; otherwise, the preset interactive management indicators are maintained; The student teaching management terminal is used to update the learning portrait based on the appropriate update frequency of each student's corresponding learning portrait, and to manage teaching interactions based on interaction management indicators.

2. The Internet-based network teaching interactive auxiliary management system according to claim 1, characterized in that: The construction of the learning profile of each student includes: Extracting the marked source labels of each course from the accumulated learning data of each student, where the marked source labels include recommended learning and self-selected learning; Each course marked with the source label "recommended learning" is recorded as a recommended course, and the participation rate of each student in the recommended course is counted; Extract the course type, setting information and learning footprint of each course from the accumulated learning data of each student, and calculate the recommended autonomous participation consistency of each student; Integrate all courses under the same course type to obtain all courses under each course type corresponding to each student; Extracting the set class hours from the setting information, and extracting the number of learning class hours and the last learning date from the learning footprint; The number of learning hours is compared with the number of set hours, and the ratio is recorded as the completion ratio. The average course completion ratio k of each student corresponding to each course type is obtained by average calculation. ij , i represents the student number, i=1,2,......n, j represents the course type number, j=1,2,......m; Set the interest correction factor μ for each student corresponding to each course type ij , count the interest of each student in each course type β ij , u represents the number of courses, e is a natural constant, and k′ is the completion ratio of the set reference; The recommended course participation, recommended autonomous participation consistency and interest in each course type are used as portrait elements to construct a learning radar chart for each student as his or her learning portrait.

3. The Internet-based network teaching interactive auxiliary management system according to claim 2, characterized in that: The statistics of each student's participation in the recommended courses include: Extracting the number of interactive nodes set from the setting information, and extracting the number of participating interactive nodes from the learning footprint, and recording the ratio of the two as the interaction ratio; Count the number of recommended courses for each student whose corresponding completion ratio is greater than the set reference completion ratio, and divide it by the total number of recommended courses for the corresponding student to obtain the recommended learning achievement ratio of each student; Count the number of recommended courses for each student whose corresponding interaction ratio is greater than the set reference interaction ratio, and divide it by the total number of recommended courses for the corresponding student to obtain the recommended interaction achievement ratio of each student; Set the weights of the recommended learning achievement ratio and the recommended interaction achievement ratio, and calculate the recommended course participation of each student through weighted average.

4. The Internet-based network teaching interactive auxiliary management system according to claim 2, characterized in that: The statistical analysis of the degree of consistency of each student's recommended autonomous participation includes: Extracting the set course content from the setting information, and recording each course marked with a source tag of self-selected learning as a main course; Based on the set course content, the similarity of the course content of each student's main course and each recommended course is calculated by a similarity measurement algorithm; If the similarity between a self-directed course and a recommended course exceeds the set value, the recommended course will be used as a similar recommended course for the self-directed course. Count the number of similar recommended courses for each student's main course, take the maximum number and divide it by the total number of corresponding recommended courses to get the recommended independent content matching ratio of each student; For each student, count the number of courses in which they participate in self-study that are of the same type as the recommended courses, and divide this number by the total number of recommended courses to obtain the matching ratio of the recommended self-study courses for each student. According to the statistical method of recommended course participation, the autonomous course participation of each student is statistically obtained in the same way. The recommended course participation and autonomous course participation of each student are respectively denoted as δ i and δ i ′, calculate the matching ratio λ of each student’s recommended autonomous participation degree i , Set weights for the recommended autonomous content match ratio, recommended autonomous course type match ratio, and recommended autonomous participation degree match ratio, and calculate each student's recommended autonomous participation match ratio through weighted average.

5. The Internet-based network teaching interactive auxiliary management system according to claim 3, characterized in that: The setting of the interest correction factor for each student corresponding to each course type includes: Compare the last study date with the current date to obtain the last study break duration, and compare the study break duration of each course of each student under each course type with the set reference time window; If the learning interval of a course exceeds the set reference time window and the completion ratio of the course is lower than the set reference completion ratio, the course will be recorded as an interrupted course. The number of interrupted courses of each student under each course type will be counted and divided by the total number of courses. The ratio will be used as the interest correction factor for each student under each course type.

6. The Internet-based network teaching interactive auxiliary management system according to claim 2, characterized in that: The learning profile update analysis includes: Based on the learning footprints of each student in each recommended teaching course in each recommendation period, the participation rate of each student in the recommended courses in each recommendation period is obtained by using the same statistical method as the statistical method of the participation rate of each student in the recommended courses; With the recommendation period as the horizontal axis and the recommended course participation as the vertical axis, a recommended course participation change curve for each student is constructed, and the slope of the curve is extracted as the learning participation change rate, recorded as (k c ) i ; Count the number of recommended cycles in which the participation rate of each student in the recommended course is greater than the set reference participation rate, and divide it by the total number of recommended cycles to obtain the matching ratio of each student's recommended course learning participation cycle, recorded as (k b ) i ; Count the appropriate update frequency p of each student's learning profile i , k c ′ and k b ′ represents the learning participation change rate and participation cycle matching ratio of the set reference respectively, and p0 is the set automatic update frequency of the initial learning profile.

7. The Internet-based network teaching interactive auxiliary management system according to claim 1, characterized in that: The determining whether to trigger interactive management includes: Compare the number of interactive students participating in the current online course with the number of online students to obtain the online interaction ratio; Extract the cumulative number of interaction IDs, the number of interaction conversations, and the initiation interaction ID, conversation content, and initiation time of each interaction conversation from the interaction record data, and calculate the online interaction frequency index; The online interaction ratio greater than the set reference interaction ratio is defined as judgment condition 1, and the interaction square degree greater than the set reference interaction frequency index is defined as judgment condition 2; Determine whether the above judgment conditions are met. If so, it will be triggered as the judgment result, otherwise it will not be triggered as the judgment result.

8. The Internet-based network teaching interactive auxiliary management system according to claim 7, characterized in that: The statistical online interaction frequency index includes: Based on the initiation time of each interactive dialogue, the starting initiation time and the last initiation time are extracted, and the interval between the two is taken as the interaction duration. The number of interactive dialogues is divided by the interaction duration to obtain the interaction frequency, which is recorded as f; All interactive dialogues are compared pairwise, and the content similarity between two dialogues is calculated using a similarity measurement algorithm. When the content similarity between two interactive dialogues is greater than or equal to a set threshold, the two interactive dialogues are recorded as repeated interactive dialogue pairs, and the number of repeated dialogue pairs is counted, recorded as M; Divide the number of interactive conversations by the number of interactive IDs recorded cumulatively to obtain the number of interactive conversations for a single interactive ID, recorded as D, and the number of interactive conversations as T; Calculate the online interaction frequency index γ, r1, r2 and r3 represent the weights corresponding to the interaction frequency, the number of repeated conversations and the number of interactive conversations for a single interactive ID, respectively. f′ and D′ represent the interaction frequency and the number of single-person interactive conversations for the set reference, respectively.

9. The Internet-based network teaching interactive auxiliary management system according to claim 8, characterized in that: The confirmed interaction management indicators include: If only condition 1 is met, interactive partitioning is used as a management item. If condition 2 is met, message queue management is used as a management item. If the management item is interactive partitioning, each interactive conversation is allocated based on a preset load balancing algorithm, and the backend server corresponding to each interaction is obtained and used as an interactive management indicator; If the management item is message queue management, extract the identity tag of each interaction ID from the interaction record data, set the processing priority coefficient of each interaction ID, and sort the interaction IDs from large to small according to their processing priority coefficients as the message processing order of each interaction ID and as the interaction management indicator.

10. The Internet-based network teaching interactive auxiliary management system according to claim 9, characterized in that: The step of setting the processing priority coefficient of each interactive ID includes: The processing priority coefficient of the interaction ID with the identity label of teacher is recorded as 1; The interaction IDs whose identity tags are not teachers are extracted and recorded as target IDs. Based on the initiating interaction IDs of each interactive dialogue, each interaction under the target ID is filtered out; Based on the content and initiation time of each interactive conversation under the target ID, the interaction frequency and number of repeated conversation pairs of the target ID are counted, denoted as f″ and M′ respectively. As the processing priority coefficient of the target ID, denoted as ε, T represents the number of interactive conversations, and the processing priority coefficient μ of each interactive ID is obtained by setting d , μ d The value is 1 or ε, d represents the interaction ID number, d = 1, 2, ... u.

Images