An online education effect feedback statistical analysis system based on a big data model

Abstract

The application discloses an online education effect feedback statistical analysis system based on a big data model, relates to the technical field of online education effect analysis, and solves the technical problem that the existing online education platform generally lacks an effective learning effect evaluation mechanism and it is difficult to accurately analyze the learning behavior and performance of students, specifically, after the execution of an online education course, a learning behavior analysis unit analyzes the learning behavior of students participating in online education, infers whether the learning behavior of the students meets the requirements according to the execution stage of the online education course, and thus analyzes the online education effect; a learning performance analysis unit detects the learning performance of the students participating in online education, and infers the online education effect of the current stage through the learning performance detection; and a learning progress achievement analysis unit analyzes the progress completion of real-time courses of the students participating in online education.

Description

Technical Field

[0001] This invention relates to the field of online education effectiveness analysis technology, specifically to an online education effectiveness feedback statistical analysis system based on a big data model. Background Technology

[0002] With the rapid development of internet technology, online education has become an important part of the education field. However, existing online education platforms generally lack effective learning outcome evaluation mechanisms, making it difficult to accurately analyze students' learning behaviors and performance. Traditional methods mainly rely on manual observation and simple statistics, which cannot acquire and analyze comprehensive data on students during both online and offline learning phases in real time, resulting in lagging and one-sided evaluations of educational outcomes. Especially when processing large-scale online education data, existing technologies struggle to accurately identify students' learning behavior patterns, performance characteristics, and progress differences, failing to provide timely and effective decision-making basis for adjusting teaching.

[0003] Furthermore, existing systems lack in-depth analysis of the correlation between learning behavior and educational outcomes, and cannot dynamically adjust teaching strategies based on students' real-time feedback, thus affecting the quality and efficiency of online education.

[0004] To address the aforementioned technical shortcomings, a solution is proposed. Summary of the Invention

[0005] The purpose of this invention is to solve the problems mentioned above by proposing an online education effectiveness feedback statistical analysis system based on a big data model.

[0006] The objective of this invention can be achieved through the following technical solutions:

[0007] A statistical analysis system for online education effectiveness feedback based on a big data model includes a statistical analysis platform, which is connected to a learning behavior analysis unit, a learning performance analysis unit, and a learning progress and outcome analysis unit.

[0008] After the online education courses are implemented, the learning behavior analysis unit analyzes the learning behavior of students who participated in the online education. Based on the implementation stage of the online education courses, it infers whether the students' learning behavior meets the requirements, thereby analyzing the effectiveness of the online education.

[0009] The learning performance analysis unit assesses the learning performance of students participating in online education and infers the effectiveness of online education at the current stage based on the assessment results.

[0010] The learning progress and outcome analysis unit analyzes the progress of students participating in online education's real-time courses.

[0011] In a preferred embodiment of the present invention, the learning behavior analysis unit operates as follows:

[0012] The online education courses are divided into online learning phases and offline learning phases. The online learning phase yields active and passive feedback from students. Feedback is represented by the number of questions asked and the number of discussions participated in during the online education courses. Active and passive feedback are represented by the two methods: students actively requesting and online teachers assigning them.

[0013] During the offline learning phase, students actively intervened and accepted the learning content. Active intervention was reflected in the control of video viewing time, the number of pauses, and the number of fast-forwards. Active acceptance was reflected in the fact that the learning videos were played in a chapter-based format without intervention.

[0014] In a preferred embodiment of the present invention, online learning phase analysis is performed by collecting the error rate ratio between active and passive feedback during the online learning phase and marking it as an online analysis parameter. If the online analysis parameter exceeds a set error rate ratio threshold, it is inferred that the current student's learning enthusiasm is high but the online education effect is low. A course type adjustment signal is then generated and sent to the statistical analysis platform. After receiving the signal, the statistical analysis platform adjusts the type of the current student's online education course, such as adjusting the knowledge scope covered by the course chapters or adjusting the number of knowledge covered by the chapters.

[0015] If the online analysis parameters do not exceed the set error rate ratio threshold, it is inferred that the current student's learning motivation is low and the online education effect is low. A learning monitoring signal is generated and sent to the statistical analysis platform. The statistical analysis platform monitors the current student's learning, increases the frequency of passive feedback, and statistically analyzes the accuracy rate to determine whether the current student has completed the current learning course.

[0016] In a preferred embodiment of the present invention, the offline learning phase analysis collects the ratio of the cumulative duration of courses with active intervention in adjacent course chapters within the offline learning phase to the cumulative duration of courses with continuous active acceptance, and marks it as an offline analysis parameter; if the offline analysis parameter exceeds the preset duration ratio, it is inferred that the current student's learning initiative is declining, an offline content adjustment signal is generated and sent to the statistical analysis platform, and after receiving it, the statistical analysis platform conducts an offline course video assessment for the current student and uses the content that fails the assessment as the content for the next offline course;

[0017] If the offline analysis parameters do not exceed the preset duration ratio, it is inferred that the current student's learning initiative is on a normal trend, and an offline course efficiency signal is generated and sent to the statistical analysis platform.

[0018] As a preferred embodiment of the present invention, the process of learning the performance analysis unit is as follows:

[0019] In the current online education course, when the error rate corresponding to the online analysis parameters decreases synchronously, the frequency of subsequent errors of the corresponding knowledge points learned by students during the online learning phase is collected. At the same time, when the offline analysis parameters continue to decrease, the probability of errors corresponding to the same type of knowledge points in the homework completed by the corresponding students during the offline learning phase is collected, and the values ​​are compared.

[0020] In a preferred embodiment of the present invention, if the frequency of subsequent errors in the knowledge points learned by the corresponding student during the online learning phase exceeds the continuous frequency threshold, or if the probability of errors in the same type of knowledge points in the homework completed by the corresponding student during the offline learning phase exceeds the error probability threshold, it is inferred that the online education effect of the corresponding student is abnormal, an online education abnormality signal is generated and sent to the statistical analysis platform, the statistical analysis platform modifies the current student's online education course, and compiles the error knowledge points and prints paper review materials.

[0021] If the frequency of subsequent errors in the knowledge points learned by the corresponding student during the online learning phase does not exceed the continuous frequency threshold, and the probability of errors in the same type of knowledge points in the homework completed by the corresponding student during the offline learning phase does not exceed the error probability threshold, then it is inferred that the online education effect of the corresponding student is normal, and a normal online education signal is generated and sent to the statistical analysis platform.

[0022] As a preferred embodiment of the present invention, the process of the learning progress result analysis unit is as follows:

[0023] When online education analysis is normal, the speed deviation between the planned completion speed of the total number of online education courses and the actual completion speed of the course volume is obtained. At the same time, the speed deviation of the online education course progress deviation and the increase rate of the course volume after the speed deviation occurs are obtained, and data analysis is performed on them.

[0024] As a preferred embodiment of the present invention, if the speed deviation between the planned completion speed of the total number of chapters in the online education course and the actual completion speed of the course exceeds the speed deviation threshold, or if the increase speed of the progress deviation of the online education course after the speed deviation occurs exceeds the increase speed threshold, it is inferred that the online education course is not suitable for the current student, and an online course total cycle planning signal is generated and sent to the statistical analysis platform. After receiving the signal, the statistical analysis platform re-plans the total online course cycle of the current student and the course completion progress of the corresponding cycle nodes to reduce the course learning speed.

[0025] If the speed difference between the planned completion speed of the total number of online education courses and the actual completion speed does not exceed the speed difference threshold, and the increase rate of the progress deviation of the online education courses after the speed difference occurs does not exceed the increase rate threshold, then it is inferred that the online education courses are suitable for the current students, and a signal of matching the total online course cycle is generated and sent to the statistical analysis platform.

[0026] Compared with the prior art, the beneficial effects of the present invention are:

[0027] 1. In this invention, through the above technical solution, this application can effectively distinguish the data characteristics of students' autonomous behavior and teaching intervention behavior, accurately identify the real learning participation in online interaction and the self-control ability in offline learning, thereby providing a more accurate behavioral analysis dimension for evaluating the effectiveness of online education, and helping teachers to optimize the course interaction design and learning content arrangement strategy in a targeted manner.

[0028] In this invention, the above-described technical solution solves the technical problem that existing systems cannot dynamically identify the correlation between students' active learning behavior and knowledge acquisition effectiveness. By comparing the error rates of active and passive feedback, the true state of students' enthusiasm and knowledge absorption efficiency can be distinguished, thereby enabling targeted selection of course optimization or monitoring and reinforcement strategies.

[0029] In this invention, through the above technical solution, the application can dynamically adjust the course content according to the behavioral characteristics of students in the offline learning stage, ensure that students master weak knowledge points, avoid repetitive errors caused by decreased learning initiative, and thus improve the personalized adaptation capability and learning effect of online education. Attached Figure Description

[0030] To facilitate understanding by those skilled in the art, the present invention will be further described below with reference to the accompanying drawings;

[0031] Figure 1 This is a system principle block diagram of the present invention;

[0032] Figure 2 This is a flowchart of the method of the present invention. Detailed Implementation

[0033] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0034] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0035] In existing technologies, online education platforms typically employ basic statistical methods for effectiveness evaluation, such as recording course completion rates or test scores. These methods suffer from limitations such as limited analytical dimensions and delayed feedback, making it difficult to track the correlation between behavioral characteristics and knowledge acquisition during the learning process in real time. One online programming teaching platform found that homework accuracy alone could not explain the phenomenon of some students exhibiting high course participation but poor application of knowledge, reflecting the difficulty of traditional assessment systems in identifying the deeper contradiction between learning behavior patterns and educational outcomes.

[0036] To address these issues, the R&D team realized the need to build a multi-dimensional data analysis model. Analysis revealed that examining learning time or test scores in isolation easily overlooks the impact of behavioral patterns on knowledge absorption efficiency. After multiple experimental verifications, the team proposed dividing the learning process into two stages: online interaction and offline self-directed learning, establishing behavioral feature collection mechanisms for each. Further research revealed the need for a dynamic analysis mechanism to correlate behavioral data with knowledge mastery, thereby forming a closed-loop feedback system.

[0037] Therefore, please refer to Figure 1-2 As shown, this application proposes an online education effectiveness feedback system that includes a statistical analysis platform. This platform integrates a learning behavior analysis unit, a learning performance analysis unit, and a learning progress and outcome analysis unit. After the online education course is completed, the learning behavior analysis unit assesses the compliance of learning behavior by analyzing students' online question-and-answer interactions and offline video operation behaviors. The learning performance analysis unit simultaneously monitors the mastery of knowledge points, and the learning progress and outcome analysis unit tracks course progress deviations in real time.

[0038] The learning behavior analysis unit is a system component that acquires online interactive behavior characteristics and offline learning trajectories through a data acquisition module. Specifically, it can connect to the teaching platform's log system via an API interface to capture behavioral data such as the number of questions asked and the frequency of video pauses in real time. The learning performance analysis unit is a processing module that correlates behavioral data with the degree of knowledge mastery. Specifically, it can establish an error-knowledge point mapping model to identify the correspondence between high-frequency errors and specific learning behaviors. The learning progress and outcome analysis unit is an evaluation module that dynamically monitors the execution of course progress. Specifically, it can use a progress deviation algorithm to compare the preset learning path with the actual progress curve.

[0039] Specifically, the system obtains real-time behavioral log data from the teaching platform via API. The learning behavior analysis unit inputs active questioning and passive questioning data from the online phase into a classification model, while analyzing the temporal distribution characteristics of video viewing behavior during the offline phase. When a correlation is detected between high-frequency fast-paced behavior and knowledge point errors, the learning performance analysis unit triggers an early warning mechanism. The learning progress and outcome analysis unit simultaneously calculates the deviation between the chapter completion speed and the preset schedule, and initiates a course plan adjustment procedure when a continuous deviation is detected.

[0040] Compared to existing technologies, traditional systems only provide static completion statistics, while this solution constructs a dynamic correlation model between behavioral characteristics and knowledge mastery. The independent progress tracking function in existing technologies forms a data loop with behavioral analysis in this solution, enabling the identification of anomalies where progress appears to be met but actual results are poor. By establishing a multi-dimensional data analysis system, the problem of misjudgment caused by single-dimensional evaluation is solved.

[0041] Through the above technical solutions, this application achieves comprehensive monitoring of the online learning process, enabling the identification of learning patterns characterized by high participation but low efficiency. By correlating behavioral characteristics with the distribution of knowledge errors, the system accurately pinpoints teaching segments requiring improvement. The dynamic progress monitoring mechanism can promptly identify situations where progress appears normal but knowledge acquisition is weak, providing data support for adjusting the curriculum plan.

[0042] This application further proposes an online education effectiveness feedback statistical analysis system based on a big data model. The learning behavior analysis unit process is as follows: Based on the online education course, online learning stage and offline learning stage are obtained. Through the online learning stage, active and passive feedback from students are obtained. Feedback is represented by the number of questions asked and the number of discussions participated in within the online education course. Active and passive feedback are represented by the two methods: students actively requesting and online teachers assigning. In the offline learning stage, active intervention and active acceptance from students are obtained. Active intervention is represented by the control of video viewing time, the number of pauses, and the number of fast-forwards. Active acceptance is represented by the playback of learning videos according to the set chapter divisions without intervention.

[0043] Among these, "active feedback" refers to interactive behaviors initiated by students independently without teacher intervention. This can be achieved by monitoring the number of times the question button is triggered on the course interface and the frequency of students speaking voluntarily in the discussion area, reflecting students' spontaneous learning participation. "Passive feedback" refers to interactive behaviors assigned by the teacher, specifically measured by the frequency of teachers calling on students to answer questions and the completion rate of mandatory discussion tasks, assessing students' learning responsiveness under external guidance. "Active intervention" refers to students' manual manipulation of the course videos during the learning process, specifically measured by recording the distribution of pause times, fast-forward frequency, and single viewing duration, identifying fluctuations in students' attention during self-directed learning. "Active acceptance" refers to students watching videos in a pre-set chapter order, specifically measured by detecting the continuous playback status of the video progress bar and the interval between chapter transitions, measuring students' structured acceptance of the course content.

[0044] Specifically, in the online learning phase, by distinguishing between two interaction methods—active application and teacher-assigned instruction—it is possible to separate the impact of student self-motivation and teacher guidance on learning behavior. In the offline learning phase, by quantifying video manipulation behavior and adherence to chapter sequence, a behavioral model reflecting the patterns of student self-directed learning can be constructed. For example, in teacher-assigned discussion tasks, if the frequency of passive feedback is significantly higher than that of active feedback, it indicates that students rely on external incentives; if the active intervention indicators formed by the number of video pauses and fast-forwards exceed a threshold, it reflects a problem of inattention during self-directed learning.

[0045] Compared to existing technologies, traditional online education systems typically only count the total number of questions asked and the total video viewing time, failing to distinguish the differences in the impact of proactive behavior versus passive response on learning outcomes. This solution, by establishing a multi-dimensional behavior classification system, can accurately identify the interaction between students' intrinsic learning motivation and external teaching strategies, providing more granular data support for subsequent personalized teaching adjustments.

[0046] Through the above technical solution, this application can effectively distinguish the data characteristics of students' autonomous behavior and teaching intervention behavior, accurately identify the real learning participation in online interaction and the self-control ability in offline learning, thereby providing a more accurate behavioral analysis dimension for evaluating the effectiveness of online education and helping teachers to optimize the course interaction design and learning content arrangement strategies in a targeted manner.

[0047] This application further proposes a specific implementation method for online learning phase analysis, including collecting the error rate ratio of active feedback and passive feedback during the online learning phase and marking it as an online analysis parameter, generating a course type adjustment signal or learning monitoring signal based on the comparison result of the online analysis parameter and the set error rate ratio threshold, and adjusting the course type or learning monitoring strategy through a statistical analysis platform.

[0048] Among these, active feedback refers to students' proactive interaction with teachers, which can be achieved through a statistical module recording the number of questions asked or the frequency of participation in discussion forums, reflecting the degree of students' independent participation. Passive feedback refers to teachers assigning students to answer questions or participate in discussions, which can be achieved through a classroom attendance system or task assignment module, used to assess students' passive response ability. The error rate ratio is the ratio of errors in active feedback to errors in passive feedback, specifically obtained through a data calculation unit by performing difference analysis on the accuracy rates of the two types of feedback and then standardizing the results, used to quantify the difference in knowledge mastery between students in active and passive states. The course type adjustment signal is an instruction that triggers adjustments to the course structure, specifically by dynamically adjusting the coverage of chapter knowledge or the density of knowledge points based on a preset algorithm, used to match students' current learning status. The learning monitoring signal is an instruction to enhance the supervision of the learning process, specifically by increasing the frequency of teacher-assigned tasks or shortening the feedback cycle, used to increase the intensity of passive feedback data collection.

[0049] Specifically, when the online analysis parameters exceed the set error rate threshold, it indicates that students have a high error rate during active participation, but the accuracy rate of passive feedback has not improved accordingly. At this point, the system automatically identifies this as a mismatch between learning enthusiasm and knowledge absorption efficiency, triggering a course structure adjustment mechanism. For example, the distribution of knowledge points in course chapters may be rearranged, complex knowledge points may be broken down into multiple subdivided learning units, or the number of knowledge points per lesson may be reduced. When the online analysis parameters do not exceed the threshold, it indicates insufficient active participation and low knowledge mastery among students. The system will increase the frequency of passive feedback collection, for example, by requiring students to submit a summary report after completing each knowledge point, while simultaneously analyzing accuracy data to determine whether to allow students to proceed to the next stage of learning.

[0050] Compared to existing technologies, traditional online education systems typically judge learning effectiveness based on a single feedback type (such as simply counting the number of questions asked), failing to distinguish the differences in the impact of active and passive behaviors on knowledge acquisition. This solution, by quantifying the correlation between error rates of the two types of feedback, can accurately identify the contradictory relationship between students' willingness to learn actively and their efficiency in knowledge absorption. For example, when students frequently ask questions but their error rate remains high, existing technologies may misjudge this as high motivation and maintain the original course pace. This solution, however, can adjust the course structure in a timely manner to avoid overloading students with too much knowledge.

[0051] Through the above technical solution, this application solves the technical problem that existing systems cannot dynamically identify the correlation between students' active learning behavior and knowledge acquisition effectiveness. By comparing the error rate differences between active and passive feedback, the true state of students' enthusiasm and knowledge absorption efficiency can be distinguished, thereby enabling targeted selection of course optimization or monitoring and reinforcement strategies. For example, in cases of abnormal error rate ratios, the system can avoid overlooking the problem of insufficient knowledge digestion capacity due to one-sided reliance on the number of questions asked, effectively preventing a disconnect between learning progress and knowledge acquisition.

[0052] This application further proposes an online education effectiveness feedback statistical analysis system based on a big data model, which includes a statistical analysis platform connected to a learning behavior analysis unit, a learning performance analysis unit, and a learning progress and achievement analysis unit. The learning behavior analysis unit includes an offline learning stage analysis module, which collects the ratio of the cumulative duration of courses with active intervention in adjacent course chapters within the offline learning stage to the cumulative duration of courses with continuous active learning, and marks it as an offline analysis parameter. If the offline analysis parameter exceeds a preset duration ratio, it is inferred that the current student's learning initiative is declining, an offline content adjustment signal is generated and sent to the statistical analysis platform, and after receiving it, the statistical analysis platform conducts an offline course video assessment for the current student and uses the content that fails the assessment as the content for the next offline course. If the offline analysis parameter does not exceed the preset duration ratio, it is inferred that the current student's learning initiative is on a normal trend, an offline course efficiency signal is generated and sent to the statistical analysis platform.

[0053] Active intervention refers to students' actions in controlling video viewing time, number of pauses, or number of fast-forwards during offline learning. This can be achieved through the video player's operation log module, such as by collecting pause event timestamps and fast-forward frequency data. Active acceptance refers to students not intervening in video playback during offline learning. This can be achieved through the video player's continuous playback duration monitoring module, such as by recording the uninterrupted playback completion status of video chapters.

[0054] Specifically, during the offline learning phase, the system assesses student initiative by comparing the ratio of proactive intervention time to continuous proactive learning time in adjacent course chapters. For example, when the proportion of proactive intervention time exceeds a preset threshold, the system automatically triggers a course content adjustment mechanism, prioritizing knowledge points that were not assessed in subsequent learning. If the proportion does not exceed the threshold, the original course arrangement is maintained to ensure that learning progress matches effectiveness.

[0055] Compared to existing technologies, current online education systems typically focus only on behavioral data during the online learning phase, lacking dynamic monitoring of initiative during the offline learning phase. This solution quantifies intervention and acceptance behaviors during the offline learning phase, enabling real-time assessment of learning initiative. Furthermore, it optimizes course content through an adaptive adjustment mechanism, addressing the problem that traditional methods cannot dynamically adjust teaching strategies based on offline learning status.

[0056] Through the above technical solution, this application can dynamically adjust the course content according to the behavioral characteristics of students during the offline learning stage, ensuring that students master the weak knowledge points, avoid repetitive errors caused by decreased learning initiative, and thus improve the personalized adaptation capability and learning effect of online education.

[0057] This application further proposes an online education effectiveness feedback statistical analysis system based on a big data model. The learning performance analysis unit process includes: when the error rate corresponding to the online analysis parameters decreases synchronously within the current online education course, collecting the subsequent continuous frequency of errors for the knowledge points learned by the corresponding students during the online learning phase; and when the offline analysis parameters continue to decrease, collecting the error probability corresponding to the same type of knowledge points in the homework completed by the corresponding students during the offline learning phase, and comparing the values ​​of these parameters.

[0058] The frequency of subsequent errors refers to the distribution pattern of the number of consecutive errors made by students when learning the same knowledge point. This can be achieved through an error count module combined with time series analysis, used to detect whether students' mastery of the knowledge point still has persistent deficiencies after multiple learning sessions. The probability of errors corresponding to the same knowledge point in homework refers to the error rate of students on the same knowledge point in their homework. This can be achieved by associating knowledge point tags with the homework grading system and calculating the error rate percentage, used to verify whether the learning effect in the offline learning stage is correlated with the online learning stage.

[0059] Specifically, when the error rate corresponding to the online analysis parameters shows a synchronous downward trend, the system will, for the knowledge point the student is currently learning, count the frequency of consecutive errors in subsequent exercises or tests, such as making mistakes on the same knowledge point in three exercises. Simultaneously, as the offline analysis parameters continue to decrease, the system will collect the probability of errors occurring in the student's homework involving similar knowledge points, for example, by associating homework questions with knowledge point tags to calculate the proportion of incorrect questions out of the total number of questions. After extracting the error data from both online and offline phases, the system will compare the two values, for example, comparing the consecutive error frequency with a preset threshold, or analyzing whether the error probability exceeds the historical average level.

[0060] Compared to existing technologies, traditional online education systems typically focus only on error rate analysis in a single learning phase, such as simply counting the number of errors in online tests, without verifying the correlation between online and offline learning effectiveness data. This solution, by combining the continuous frequency of errors in the online phase with the error probability of assignments in the offline phase, can more comprehensively identify students' knowledge gaps in different learning scenarios, avoiding misjudgments due to the limitations of a single data source.

[0061] Through the above technical solution, this application can effectively identify persistent deficiencies in students' learning of knowledge points. For example, when the error rate decreases during the online learning phase but errors still occur frequently in subsequent exercises, the system can promptly determine that the student may have a misunderstanding of the knowledge points, thereby triggering targeted intervention measures and improving the accuracy of online education effectiveness analysis and the efficiency of dynamic adjustment.

[0062] This application further proposes an online education effectiveness feedback statistical analysis system based on a big data model. When the continuous frequency of subsequent errors in the knowledge points learned by the corresponding student during the online learning phase exceeds a continuous frequency threshold, or when the probability of errors in the same type of knowledge points in the homework completed by the corresponding student during the offline learning phase exceeds an error probability threshold, it is inferred that the online education effect of the corresponding student is abnormal. An online education abnormality signal is generated and sent to the statistical analysis platform. The statistical analysis platform modifies the current student's online education course, counts the erroneous knowledge points, and prints out paper review materials. If the continuous frequency of subsequent errors in the knowledge points learned by the corresponding student during the online learning phase does not exceed the continuous frequency threshold, and the probability of errors in the same type of knowledge points in the homework completed by the corresponding student during the offline learning phase does not exceed an error probability threshold, it is inferred that the online education effect of the corresponding student is normal. An online education normality signal is generated and sent to the statistical analysis platform.

[0063] The consecutive error frequency refers to the number of times the same knowledge point is answered incorrectly consecutively during the online learning phase. This can be achieved by recording students' answers during interactive Q&A sessions, and is used to determine whether there are systematic deviations in students' understanding of the knowledge point. The error probability threshold refers to the critical value of the error rate for similar knowledge points in homework assignments during the offline learning phase. This can be determined by statistically analyzing the average error rate of a similar group of students on that knowledge point, and is used to identify students' weaknesses during self-study. Online education anomaly signals refer to electronic instructions generated based on anomaly detection. These signals can be transmitted using message queue technology and are used to trigger a dynamic adjustment mechanism for the teaching content. Paper-based review materials refer to specialized learning documents compiled for incorrect knowledge points. These can be automatically generated by linking to a knowledge point database to provide targeted reinforcement training.

[0064] Specifically, when the system detects that a student answers a specific knowledge point incorrectly more than three times consecutively during the online learning phase, it automatically marks that knowledge point as abnormal. Simultaneously, it calculates the student's error rate for similar knowledge points in offline assignments. If either indicator exceeds a preset threshold, the system immediately generates an anomaly signal and adjusts subsequent course arrangements through the analysis platform. For example, it might replace the planned course with a dedicated explanation of the incorrect knowledge point and automatically generate printed review materials containing typical examples, which are then sent to the student. If both indicators are within the normal range, the original teaching plan is maintained, and a normal signal is issued.

[0065] Compared to existing technologies, current online education systems typically only assess error rates in isolation at a single learning stage. This solution, however, establishes a two-dimensional correlation analysis mechanism between online interaction and offline practice, enabling more accurate identification of learning difficulties caused by misunderstandings of knowledge points. For example, traditional systems might adjust courses solely based on online test results, while this solution, by combining homework data, can uncover the correlations between students' knowledge gaps across different learning scenarios.

[0066] Through the above technical solution, this application effectively solves the problem of declining learning efficiency caused by the accumulation of students' cognitive biases in online education. By monitoring the correlation anomalies of cross-stage learning data in real time, it can quickly locate weak links in teaching and generate targeted review materials, avoiding misjudgments caused by traditional single-dimensional analysis and ensuring the timeliness and accuracy of teaching adjustments.

[0067] This application further proposes an online education effect feedback statistical analysis system based on a big data model. In this system, when the online education analysis is normal, the learning progress and achievement analysis unit obtains the speed deviation between the planned completion speed of the total number of online education courses and the actual completion speed of the course volume. At the same time, it obtains the real-time increase rate of the online education course progress deviation after the speed deviation occurs and performs data analysis on it.

[0068] The total course volume and planned completion rate refers to the progress standard for course chapters as preset in the syllabus. This can be achieved by standardizing the chapter completion time nodes using a time series model. The actual course volume and completion rate refers to the rate at which students actually complete course chapters. This can be calculated by collecting the time difference between course access timestamp data and chapter completion marker data. The rate of increase in course volume due to progress deviation refers to the growth rate of the number of uncompleted course chapters when the actual learning progress deviates from the planned progress. This can be achieved by dynamically monitoring the changing trend of the number of lagging chapters using a sliding window algorithm.

[0069] Specifically, when the system detects that the online education effect is normal, it calculates the speed deviation value by comparing the preset chapter progress plan with the actual learning rate. If this deviation value continues to accumulate, the system will simultaneously monitor the growth rate of the number of unfinished courses. By setting a speed deviation threshold and increasing the speed threshold, when either indicator exceeds the threshold, the system automatically determines that the current course arrangement does not match the student's actual learning ability, triggering a course total cycle replanning mechanism. This mechanism dynamically optimizes the course progress arrangement by adjusting the chapter learning cycle length or adding phased review nodes.

[0070] Compared to existing technologies, traditional online education systems only monitor learning behavior data or knowledge acquisition levels, lacking dynamic adaptation assessments of course progress and learning abilities. Existing methods cannot identify adaptation issues in the early stages of course progress lag, leading to delayed adjustments. This solution, through a dual threshold monitoring mechanism, can identify the matching status between course planning and learning abilities in the early stages of progress deviation accumulation, enabling preventative adjustments.

[0071] Through the above technical solution, this application effectively solves the problem of low knowledge absorption efficiency caused by the mismatch between course schedule planning and students' actual learning pace in online education. By dynamically monitoring the deviation and growth trend of course completion rate, it is possible to identify adaptation problems early in the teaching cycle and optimize course arrangements in a timely manner, avoiding systemic learning obstacles caused by the accumulation of progress delays, thereby improving the implementation effect of personalized education.

[0072] This application further proposes a system including a learning progress outcome analysis unit, wherein the process of the learning progress outcome analysis unit is as follows: When online education analysis is normal, the system obtains the speed deviation between the planned completion speed of the total number of online education course chapters and the actual completion speed of the course volume; simultaneously, it obtains the real-time increase rate of the online education course progress deviation course volume after the speed deviation occurs, and performs data analysis; if the speed deviation between the planned completion speed of the total number of online education course chapters and the actual completion speed of the course volume exceeds a speed deviation threshold, or if the real-time increase rate of the online education course progress deviation course volume after the speed deviation occurs exceeds an increase rate threshold... If the online education course is not suitable for the current student, an online course total cycle planning signal is generated and sent to the statistical analysis platform. After receiving the signal, the statistical analysis platform re-plans the current student's online course total cycle and the course completion progress of the corresponding cycle nodes, reducing the course learning speed. If the speed deviation between the planned completion speed of the total course volume chapters and the actual course volume completion speed does not exceed the speed deviation threshold, and the increase rate of the real-time online education course progress deviation course volume after the speed deviation occurs does not exceed the increase rate threshold, then the online education course is suitable for the current student, an online course total cycle alignment signal is generated and sent to the statistical analysis platform.

[0073] Speed ​​deviation refers to the difference between the planned completion speed and the actual completion speed of course chapters. This can be achieved by comparing the ratio of the planned completion time to the actual completion time, reflecting the degree of deviation between the student's current progress and the preset goal. The rate of increase in unfinished course content after the speed deviation occurs refers to the cumulative rate of unfinished course content. This can be quantified by the change in the number of unfinished chapters per unit time, used to assess the expanding trend of progress lag. The speed deviation threshold is a preset allowable critical value for speed difference, determined by the average deviation of students with normal progress in historical learning data, serving as a condition for triggering course planning adjustments. The rate of increase threshold is a preset allowable critical value for the cumulative rate of unfinished course content, determined by statistically analyzing the maximum acceptable rate of progress lag for a student group under reasonable learning intensity, used to identify the boundary conditions of course suitability issues.

[0074] Specifically, a big data model is used to collect learning progress data in real time. For example, the preset completion time and actual completion time of each course chapter are collected, and the ratio of the time difference between the two to the total planned duration is calculated as the speed deviation. When this ratio exceeds a speed deviation threshold set based on historical data, it indicates a significant mismatch between the student's current learning pace and the course plan. At this point, the rate at which the number of incomplete course chapters increases over time is further analyzed. If this rate exceeds the speed threshold, it is determined that the current course cycle arrangement exceeds the student's actual learning capacity. The total cycle duration needs to be adjusted and the chapter nodes redistributed, for example, extending the original 30-day course cycle to 40 days, while proportionally extending the deadlines for each chapter. If neither indicator exceeds the threshold, the current course plan is maintained to preserve the stability of the learning progress.

[0075] Compared to existing technologies, traditional online education systems typically employ fixed course cycles, failing to dynamically adjust based on students' real-time learning progress. This can easily lead to learning delays or excessive academic pressure. This solution, by establishing a dual threshold judgment mechanism, can identify the compatibility between the course cycle and students' abilities based on objective data and automatically trigger the cycle adjustment process, avoiding the lag of manual intervention.

[0076] Through the above technical solution, this application can monitor the expansion trend of course progress deviation in real time, identify mismatches between course planning and students' abilities in the early stages, and optimize the learning pace in a timely manner by dynamically adjusting the total cycle, preventing a decline in knowledge absorption efficiency or a decrease in learning enthusiasm due to improper course scheduling. At the same time, the quantitative threshold judgment mechanism can avoid the waste of system resources caused by frequent adjustments, ensuring the rationality and necessity of course planning adjustments.

[0077] This application further proposes the following process for the learning progress outcome analysis unit: When online education analysis is normal, the speed deviation between the planned completion speed of the total number of online course chapters and the actual completion speed of the course is obtained. Simultaneously, the real-time increase rate of the online education course progress deviation after the speed deviation occurs is obtained, and data analysis is performed. If the speed deviation between the planned completion speed of the total number of online course chapters and the actual completion speed exceeds a speed deviation threshold, or if the real-time increase rate of the online education course progress deviation after the speed deviation occurs exceeds an increase rate threshold, it is inferred that the online education course is not suitable for the current student. An online course total cycle planning signal is generated and sent to the statistical analysis platform. After receiving the signal, the statistical analysis platform re-plans the current student's online course total cycle and the course completion progress of the corresponding cycle nodes, reducing the course learning speed. If the speed deviation between the planned completion speed of the total number of online course chapters and the actual completion speed does not exceed a speed deviation threshold, and the real-time increase rate of the online education course progress deviation after the speed deviation occurs does not exceed an increase rate threshold, it is inferred that the online education course is suitable for the current student. An online course total cycle alignment signal is generated and sent to the statistical analysis platform.

[0078] Speed ​​deviation refers to the difference between the theoretical completion speed of online education course chapters and the actual completion speed achieved by students during the learning process. Specifically, it can be calculated by the difference between the planned time and the actual time spent on each chapter, reflecting the degree of deviation between the learning progress and the preset goals. The rate of increase in the number of incomplete course chapters due to progress deviation refers to the rate at which the number of incomplete course chapters increases over time when speed deviation occurs. Specifically, it can be calculated by counting the number of newly added lagging chapters per unit time, used to dynamically assess the spread trend of abnormal learning progress. The speed deviation threshold can be a pre-set upper limit for acceptable progress deviation, such as the maximum allowed percentage of chapter completion delay time. The rate of increase threshold can be a pre-set critical value for the rate of course lag expansion, such as a limit on the number of newly added incomplete chapters per week.

[0079] Specifically, by continuously monitoring the difference between the planned completion speed of chapters and the actual learning speed, deviations from the preset learning path can be detected in a timely manner. When the speed deviation exceeds a threshold, it indicates that the current course pace is not compatible with the student's learning capacity. At this point, a signal is triggered to adjust the overall course cycle planning, prompting the system to readjust the learning stage divisions and node requirements. Monitoring the rate of increase in the amount of course content deviating from the schedule can effectively identify the worsening trend of the lag problem. When the rate of increase exceeds a threshold, the system reduces the learning intensity by extending the total learning cycle and breaking down dense knowledge points to prevent the progress from spiraling out of control. During the analysis, a dual-condition judgment mechanism is used, focusing on both the static deviation amount and tracking the dynamic trend, thereby accurately judging the course suitability.

[0080] In some specific implementations, the speed deviation can be calculated using a sliding time window algorithm, such as calculating the difference between the planned number of chapters completed and the actual number completed on a weekly basis. Monitoring the rate of increase in the amount of course content in progress deviation can be done using time series analysis models, such as fitting the slope of the change in the number of lagging chapters through linear regression. A specific operation for replanning the total course cycle could be to extend the original 12-week course to 15 weeks, and add review sessions in weeks 4 and 8.

[0081] Compared to existing technologies, current online education systems typically only focus on static progress completion rates, failing to identify the dynamic risks of progress deviations and easily leading to a loss of control over the learning pace. This solution introduces a two-dimensional analysis of speed deviation and acceleration, enabling early warning of potential progress risks and dynamically optimizing course planning based on real-time data. This avoids the systemic learning obstacles caused by the accumulation of lag issues in traditional methods.

[0082] Through the above technical solution, this application can effectively solve the problem of mismatch between the pace of online education courses and students' individual learning abilities. Through dynamic monitoring and intelligent adjustment mechanisms, it can prevent the problem of lag caused by unreasonable course schedule settings from continuing to worsen, ensure that the learning plan is synchronized with the students' actual learning ability, and improve the personalized adaptation capability of online education.

[0083] This application further proposes the following process for the learning progress outcome analysis unit: When online education analysis is normal, the speed deviation between the planned completion speed of the total number of online course chapters and the actual completion speed of the course is obtained. Simultaneously, the real-time increase rate of the online education course progress deviation after the speed deviation occurs is obtained, and data analysis is performed. If the speed deviation between the planned completion speed of the total number of online course chapters and the actual completion speed exceeds a speed deviation threshold, or if the real-time increase rate of the online education course progress deviation after the speed deviation occurs exceeds an increase rate threshold, it is inferred that the online education course is not suitable for the current student. An online course total cycle planning signal is generated and sent to the statistical analysis platform. After receiving the signal, the statistical analysis platform re-plans the current student's online course total cycle and the course completion progress of the corresponding cycle nodes, reducing the course learning speed. If the speed deviation between the planned completion speed of the total number of online course chapters and the actual completion speed does not exceed a speed deviation threshold, and the real-time increase rate of the online education course progress deviation after the speed deviation occurs does not exceed an increase rate threshold, it is inferred that the online education course is suitable for the current student. An online course total cycle alignment signal is generated and sent to the statistical analysis platform.

[0084] The total course volume and chapter completion speed refer to the theoretical completion progress of each chapter as preset in the teaching plan. This can be achieved by extracting time node data from the course syllabus and is used to measure the rationality of the preset teaching pace. The actual course volume completion speed refers to the time interval data of students actually completing each chapter's learning. This can be calculated by comparing learning log timestamps and is used to reflect the actual speed at which students absorb knowledge. Speed ​​deviation refers to the time difference between theoretical and actual progress, which can be calculated using a time series interpolation algorithm and is used to quantify the matching degree between the teaching plan and learning ability. The rate of increase in course volume due to progress deviation refers to the cumulative rate of increase in the number of uncompleted course chapters after a speed deviation occurs. This can be calculated by the ratio of the number of uncompleted chapters to the number of lagging days and is used to dynamically assess the expanding trend of learning lag risk.

[0085] Specifically, when the system detects that the online education effect is normal, it calculates the deviation between the theoretical and actual progress by comparing the preset chapter completion time points with the students' actual learning records. Furthermore, when the deviation exceeds a preset threshold, the system will track the growth rate of the number of incomplete chapters in real time. For example, if a student fails to complete a scheduled chapter for three consecutive weeks, with the number of incomplete chapters increasing by two each week, the system determines that their learning pace does not match the course plan. At this point, a course cycle restructuring mechanism is automatically triggered, reallocating the learning time points for each chapter and extending the total cycle length to create a flexible teaching plan that adapts to the student's learning capacity. Conversely, if both the deviation and the growth rate are within a reasonable range, a course adaptation confirmation signal is generated, maintaining the original teaching plan unchanged.

[0086] Compared to existing technologies, traditional online education systems only record the percentage of learning progress completed, without establishing a dynamic deviation tracking and rate analysis mechanism. For example, existing technologies typically only show that a student is three chapters behind, but cannot determine whether the number of chapters behind is continuously increasing or stabilizing. This solution, by introducing a dual analysis model of speed deviation and growth rate, can accurately identify the risk of persistent lag, proactively adjust the teaching plan in the early stages, and avoid a decline in learning effectiveness due to schedule pressure.

[0087] Thresholds, preset values, preset ranges, etc. are set for result comparison and analysis to determine whether they are good or bad. The value of these thresholds is determined by a combination of large-scale model analysis of sample data and human experience. They can also be adjusted appropriately based on seasonal or common-sense influences.

[0088] Furthermore, the settings for weighting ratios, influence factors, etc., are based on the magnitude of each parameter's influence on the results. The specific values ​​are allocated to ultimately reflect the impact on the results. The settings for input and storage are also determined by a combination of large-scale model analysis of sample data and human experience. Appropriate adjustments can also be made based on seasonal or rational influence conditions.

[0089] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to any specific implementation. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A statistical analysis system for online education effectiveness feedback based on a big data model, characterized in that, It includes a statistical analysis platform, which comprises a learning behavior analysis unit, a learning performance analysis unit, and a learning progress and outcome analysis unit. Following the implementation of online education courses, a learning behavior analysis unit is conducted to analyze the learning behaviors of students participating in online education. Based on the stage of online education course implementation, it is inferred whether students' learning behaviors meet the requirements, thereby analyzing the effectiveness of online education. The process of the learning behavior analysis unit is as follows: The online education courses are divided into online learning phases and offline learning phases. The online learning phase yields active and passive feedback from students. Feedback is represented by the number of questions asked and the number of discussions participated in during the online education courses. Active and passive feedback are represented by the two methods: students actively requesting and online teachers assigning them. During the offline learning phase, students actively intervened and accepted the learning content. Active intervention was reflected in the control of video viewing time, the number of pauses, and the number of fast-forwards. Active acceptance was reflected in the fact that the learning videos were played in a pre-defined chapter division without intervention. Online learning phase analysis: Collect the error rate ratio of active feedback and passive feedback during the online learning phase and mark it as an online analysis parameter. If the online analysis parameter exceeds the set error rate ratio threshold, a course type adjustment signal is generated and sent to the statistical analysis platform; if the online analysis parameter does not exceed the set error rate ratio threshold, a learning monitoring signal is generated and sent to the statistical analysis platform. Offline learning phase analysis involves collecting the ratio of the cumulative duration of courses with active intervention in adjacent course chapters within the offline learning phase to the cumulative duration of courses with continuous active learning, and marking this as an offline analysis parameter. If the offline analysis parameter exceeds the preset duration ratio, it is inferred that the current student's learning initiative is declining, an offline content adjustment signal is generated and sent to the statistical analysis platform, and after receiving it, the statistical analysis platform conducts an offline course video assessment for the current student and uses the content that fails the assessment as the content for the next offline course. If the offline analysis parameters do not exceed the preset duration value ratio, it is inferred that the current student's learning initiative is in a normal trend, and an offline course efficiency signal is generated and sent to the statistical analysis platform. The learning performance analysis unit assesses the learning performance of students participating in online education and infers the effectiveness of online education at the current stage based on the assessment results. The learning progress and outcome analysis unit analyzes the real-time course progress of students participating in online education.

2. The online education effectiveness feedback statistical analysis system based on a big data model according to claim 1, characterized in that, The process of analyzing learning performance in the unit is as follows: In the current online education course, when the error rate corresponding to the online analysis parameters decreases synchronously, the frequency of subsequent errors of the corresponding knowledge points learned by students during the online learning phase is collected. At the same time, when the offline analysis parameters continue to decrease, the probability of errors corresponding to the same type of knowledge points in the homework completed by the corresponding students during the offline learning phase is collected, and the values ​​are compared.

3. The online education effectiveness feedback statistical analysis system based on a big data model according to claim 2, characterized in that, If the frequency of subsequent errors for a student's knowledge points during the online learning phase exceeds the continuous frequency threshold, or if the probability of errors for the same type of knowledge points in the student's homework during the offline learning phase exceeds the error probability threshold, an online education anomaly signal will be generated and sent to the statistical analysis platform. If the frequency of subsequent errors for the knowledge points learned by the corresponding student during the online learning phase does not exceed the continuous frequency threshold, and the probability of errors for the same type of knowledge points in the homework completed by the corresponding student during the offline learning phase does not exceed the error probability threshold, then a normal online education signal is generated and sent to the statistical analysis platform.

4. The online education effectiveness feedback statistical analysis system based on a big data model according to claim 3, characterized in that, The process for analyzing learning progress and outcomes in this unit is as follows: When online education analysis is normal, the speed deviation between the planned completion speed of the total number of online education courses and the actual completion speed of the course volume is obtained. At the same time, the speed deviation of the online education course progress deviation and the increase rate of the course volume after the speed deviation occurs are obtained, and data analysis is performed on them.

5. The online education effectiveness feedback statistical analysis system based on a big data model according to claim 4, characterized in that, If the speed of completion of the planned total number of online education courses exceeds the speed of completion of the actual number of courses, or if the speed of increase of the progress deviation of the online education courses after the speed deviation occurs exceeds the speed of increase, then an online course total cycle planning signal will be generated and sent to the statistical analysis platform. If the speed difference between the planned completion speed of the total number of online education courses and the actual completion speed does not exceed the speed difference threshold, and the increase rate of the progress deviation of the online education courses after the speed difference occurs does not exceed the increase rate threshold, then a total online course cycle alignment signal is generated and sent to the statistical analysis platform.

Images