Online course intelligent auxiliary platform based on data analysis
By acquiring user login data and content interaction data, analyzing users' learning status and completion rate, and pushing supplementary learning content, the problem of online learning systems being unable to comprehensively and accurately evaluate learning outcomes is solved, thereby improving learning effectiveness and content retention.
Patent Information
- Application Number
- CN202610048388.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-01-14
- Publication Date
- 2026-02-13
AI Technical Summary
Existing online learning systems cannot comprehensively and accurately evaluate users' learning outcomes, resulting in poor teaching effectiveness, especially when users complete in-class exercises in a short period of time but fail to memorize the content.
By acquiring users' login data and content interaction data, we can analyze users' learning status and completion rate, and use the data analysis platform to push supplementary learning content to improve learning effectiveness.
It enables accurate and comprehensive assessment of users' learning outcomes, improves learning effectiveness and content retention, and makes up for deficiencies in the learning process.
Smart Images
Figure CN121528067A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the technical field of online education, and in particular to a data analysis-based intelligent support platform for online courses. Background Technology
[0002] With the rapid development of internet and digital technology in education, online education has evolved from an emerging concept into an indispensable part of the global education system, experiencing explosive growth in recent years. In the process of online learning, besides the quality of the teaching content significantly impacting learning outcomes, the assessment of user learning results is also crucial. Accurate and timely assessment of user learning outcomes allows for the delivery of content that was not well-received, enabling users to review it again and thus improving overall learning effectiveness.
[0003] Existing online learning systems evaluate user learning outcomes through in-class exercises, using the scores to determine learning effectiveness. When a user's learning is deemed inadequate, subsequent learning content is adjusted to compensate for deficiencies in the previous lesson. However, relying solely on in-class exercises cannot provide a comprehensive and accurate assessment of learning outcomes. Furthermore, even if a user completes the exercises quickly, poor learning status can lead to a lack of retention of the material, resulting in suboptimal teaching effectiveness. Therefore, the fundamental problem this invention aims to solve is to accurately and comprehensively assess user learning outcomes based on their current learning status during online courses. Summary of the Invention
[0004] To accurately and comprehensively assess learning effectiveness based on users' learning status during online courses, this application provides a data-driven intelligent support platform for online courses, employing the following technical solution: A data-driven intelligent online course support platform includes: The user data collection unit is used to acquire user login data, content interaction data, and evaluation practice data. The user status analysis unit is used to determine the user's learning status based on login data and content interaction data. The learning status includes the user's learning regularity coefficient, whether the user's learning time on that day falls within the daily standard learning time range, and the learning status comparison results. The learning content analysis unit is used to determine the user's learning completion rate based on the user's learning status and content interaction data. The learning support module is used to push supplementary learning content based on the user's learning completion rate and assessment practice data.
[0005] Optionally, the user's learning status assessment process includes: Retrieve the daily login count and daily study time from the login data; Obtain the number of learning pauses and the percentage of replayed content from the content interaction data; The user's learning regularity coefficient is calculated based on the number of daily logins and the daily learning duration, and the user's daily standard learning time range is calculated based on the daily learning duration. The number of times a user pauses in learning and the percentage of replayed content are compared with the corresponding standards for the learning content to obtain the learning status comparison results; Determine whether the user's learning time falls within the daily standard learning time range. Based on the determination result, the user's learning regularity coefficient, and the learning status comparison result, obtain the user's learning status reference value. Based on the learning status reference value and content interaction data, determine the user's learning completion rate.
[0006] Optionally, the process of calculating the user learning regularity coefficient includes: Select user's historical login data for m days; Through equations Calculate and obtain the user learning regularity coefficient Sw; Where i∈[1,m], Let i be the number of logins on day i. Let be the average number of logins over m days. Let i be the study duration on day i. Let m be the average of the study time over m days. These are the weighting coefficients; The process of obtaining a user's daily standard learning time range includes: Determine whether : If so, then the standard daily study time range will be determined as follows: ; If not, then proceed according to Eliminate from largest to smallest in descending order and then recalculate. until The standard daily study time range is set as follows: ; in, To preset the comparison standard deviation, , These represent the shortest and longest daily study time over m days, respectively. , These represent the conditions satisfied after excluding the data. The shortest and longest study time per day over several days.
[0007] Optionally, the process of obtaining the learning state comparison result includes: Through equations Calculate the learning state comparison value Rc and use the learning state comparison value Rc as the learning state comparison result. Where p represents the percentage of the user's replayed learning content. Let Q be the average percentage of replayed content corresponding to the user's learning content, Q be the number of pauses during the user's learning session, and t be the duration of the user's learning session. This represents the average number of pauses per unit of time corresponding to the user's learning content. and These are the first and second defined functions, respectively. When within the corresponding preset range ,otherwise, ,when When within the corresponding preset range ,otherwise, , , These are the weighting coefficients; The process of obtaining the learning state reference value includes: Through equations The learning state reference value Z is calculated and obtained; in, For the third defined function, when t falls within the daily standard learning time range, ,otherwise, The value is obtained by referring to the preset correspondence according to the range of t.
[0008] Optionally, the process for determining the learning completion rate includes: Through equations Calculate and obtain the learning completion rate Cp; Where C represents the course completion rate. This is a correction factor.
[0009] Optionally, the calculation process for the course viewing completion rate includes: The viewing period is divided according to the interaction time point, and D time periods are obtained, j∈[1,D]; the interaction time point includes the pause time point, the magnification adjustment time point and the timestamp adjustment time point; Through equations Calculate and obtain the course viewing completion rate C; in, The playback rate for the j-th time segment is... For the j-th time segment, the number of times the video is played. express The corresponding number of time periods, k∈[1, ], Indicates the kth The duration of the corresponding time period.
[0010] Optionally, the learning assistance module operates by including: Through equations Calculate the final completion coefficient Xc, and compare the final completion coefficient Xc with the preset value: When Xc ≥ preset value, supplementary learning content will not be pushed. When Xc < preset value, push different amounts of supplementary learning content based on the difference between the preset value and the final completion coefficient Xc; Where Cy is the user evaluation value.
[0011] Optionally, the calculation process of the user evaluation value includes: Through equations Calculate and obtain the user evaluation value Cy; Where E is the number of assessment questions, x∈[1,E], For the score of the xth assessment question, Let x be the score for the xth assessment question. Let x be the correlation coefficient between the x-th assessment question and the learning content.
[0012] In summary, this application includes at least one of the following beneficial technical effects: This invention assesses a user's learning status and completion rate by acquiring their login data and content interaction data. Based on the completion rate and assessment practice data, it pushes supplementary learning content. This method can more accurately and comprehensively assess a user's learning outcomes. At the same time, by pushing supplementary learning content, users can better retain the learning content, make up for any deficiencies in the learning process, and improve the overall learning effect. Attached Figure Description
[0013] Figure 1 This is a logical block diagram of the data analysis-based intelligent online course support platform of this invention. Detailed Implementation
[0014] The embodiments of this application are described in detail below, and examples of the embodiments are shown in the accompanying drawings.
[0015] In the description of this specification, the references to "certain embodiments," "one embodiment," "some embodiments," "illustrative embodiment," "example," "specific example," or "some examples" refer to specific features, structures, materials, or characteristics described in connection with the described embodiment or example, which are included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.
[0016] This application discloses an online course intelligent assistance platform based on data analysis, referring to... Figure 1 The system includes a user data acquisition unit, a user status analysis unit, a learning content analysis unit, and a learning assistance module. The user data acquisition unit acquires user login data, content interaction data, and assessment practice data. The user status analysis unit determines the user's learning status based on the login data and content interaction data. The learning content analysis unit determines the user's learning completion rate based on the learning status and content interaction data. The learning assistance module pushes supplementary learning content based on the user's learning completion rate and assessment practice data. In this embodiment, by acquiring the user's login data and content interaction data to determine their learning status and completion rate, and then pushing supplementary learning content based on the learning completion rate and assessment practice data, this method can more accurately and comprehensively assess the user's learning outcomes. Furthermore, by pushing supplementary learning content, the user can better retain the learning content, compensate for any deficiencies in the learning process, and improve the overall learning effectiveness.
[0017] The online course intelligent assistance platform in this embodiment provides a function for simulation demonstration. Taking the junior high school Internet of Things learning process as an example, after completing the learning through video, the user can enter the virtual experiment interface through the simulation demonstration function. By selecting the main control board and various sensors, and then connecting through the endpoint, the virtual operation can be carried out. The results of the virtual operation can then be used as evaluation practice data to judge the user's learning outcomes. By setting multiple virtual simulation processes as evaluation questions according to the teaching content, the score of each evaluation question can be judged based on the user's completion level.
[0018] The user's learning status assessment process includes: first, obtaining the daily login count and daily learning duration from the login data; then, obtaining the number of learning pauses and the percentage of replayed content from the content interaction data; calculating the user's learning regularity coefficient based on the daily login count and daily learning duration, and calculating the user's daily standard learning time range based on the daily learning duration; the user's learning regularity coefficient can determine whether the user's learning process forms a regularity, while the user's daily standard learning time range can determine whether the user's learning process is carried out in a state of learning fatigue; then, comparing the user's learning pause count and the percentage of replayed content with the corresponding standards of the learning content to obtain the learning status comparison. As a result, the learning status comparison results can reflect the user's concentration during the learning process. Finally, by judging whether the user's learning time on that day falls within the daily standard learning time range, a reference value for the user's learning status is obtained based on the judgment result, the user's learning regularity coefficient, and the learning status comparison results. Therefore, the learning status reference value can comprehensively reflect the user's learning status by considering factors such as the regularity of the user's learning and the concentration during the learning process. The user's learning completion rate is judged based on the learning status reference value and content interaction data. Compared with judging the learning completion rate solely based on whether the user has watched all the teaching videos, the learning completion rate obtained in this embodiment can more objectively judge the user's learning success.
[0019] In one embodiment, the process of calculating the user learning regularity coefficient includes: firstly, selecting m days of user historical login data; the value of m is selected and set according to the size of the user's historical data; when the user's historical data is large, the number of m is increased, thereby improving the accuracy of the user learning regularity coefficient, through the equation... Calculate the user learning regularity coefficient Sw; where i∈[1,m], Let i be the number of logins on day i. Let be the average number of logins over m days. Let i be the study duration on day i. Let m be the average of the study time over m days. These are the adjustment coefficients; the magnitude of the adjustment coefficients is based on... and The numerical range difference was set based on test data. Therefore, the more regular a user's login frequency and the more uniform the duration of each study session, the stronger the regularity of the user's learning, i.e., the larger the value of the user's learning regularity coefficient Sw. Thus, the magnitude of the user's learning regularity coefficient Sw is used to judge the regularity of the user's learning. Meanwhile, the process of obtaining the user's daily standard study time range includes: determining whether... ,in, The preset standard deviation is set based on empirical data fitting, therefore, when it meets the following conditions... When the data shows relatively small fluctuations, the standard daily learning time range is determined to be [time range missing]. , , These represent the shortest and longest daily study time over m days, respectively; if not satisfied, then satisfied. When this occurs, it indicates that all data fluctuates significantly; in this case, further adjustments should be made according to... Using this as the standard time range for daily learning can lead to significant errors in the daily standard time range. Therefore, according to... Eliminate from largest to smallest in descending order and then recalculate. until The standard daily study time range is set as follows: , , These represent the conditions satisfied after excluding the data. The shortest and longest daily study time over several days can be used to determine the average daily study time range obtained through the above method, thus reflecting the user's average status.
[0020] In addition, the process of obtaining the learning state comparison results includes: through equations The learning state comparison value Rc is calculated, where p is the proportion of the user's current learning replay content. Let Q be the average percentage of replayed content corresponding to the user's learning content, Q be the number of pauses during the user's learning session, and t be the duration of the user's learning session. The average number of pauses per unit time for each user's learning content, and the average percentage of replay content for each user's learning content. and the average number of pauses per unit time corresponding to the user's learning content All settings are based on historical data of the user's learning content. and These are the first and second defined functions, respectively. When within the corresponding preset range ,otherwise, ,when When within the corresponding preset range ,otherwise, , , These are weighting coefficients, which are adaptively set based on the influence of different factors in historical data. Therefore, the learning state comparison value Rc is used as the learning state comparison result. By measuring the magnitude of the learning state comparison value Rc, the user's state during the learning process can be determined. Furthermore, the process of obtaining the learning state reference value includes: through equations... The learning state reference value Z is calculated; where, For the third defined function, when t falls within the daily standard learning time range, ,otherwise, The value of Z is obtained by comparing the range of t with a preset correspondence, which is set based on empirical data. Therefore, by calculating the learning state reference value Z, the learning state can be accurately judged by comprehensively considering the regularity of the user's learning, the learning duration, and the parameters of the learning process.
[0021] The process of judging learning completion includes: using equations... Calculate the learning completion rate Cp; where C is the course viewing completion rate. The correction coefficient is obtained by fitting test data, and the calculation process of course completion includes: dividing the viewing period according to the interaction time point to obtain D divided time periods, j∈[1,D]; the interaction time point includes the pause time point, the adjustment magnification time point, and the adjustment timestamp time point; through the equation The course completion rate C is calculated; where, The playback rate for the j-th time segment is... For the j-th time segment, the number of times the video is played. express The corresponding number of time periods, k∈[1, ], Indicates the kth By calculating the learning completion rate Cp within the corresponding time period, the user's completion status of the learning content can be accurately and objectively determined.
[0022] In addition, the learning assistance module's operation process includes: through equations The final completion coefficient Xc is calculated, where Cy is the user evaluation value. The calculation process includes: using the equation... The calculation yields the result, where E is the number of evaluation questions, and x∈[1,E]. For the score of the xth assessment question, Let x be the score for the xth assessment question. Let X be the correlation coefficient between the x-th assessment question and the learning content. This coefficient is selected by the question setter when the assessment question is created. The correlation coefficient is 1 when the assessment question and learning content are completely correlated, 0 when they are completely uncorrelated, and 0.5 when they are partially correlated. Therefore, the user's assessment value reflects their learning effectiveness during the assessment practice. By combining the final completion coefficient Xc, a more accurate judgment of the user's learning effectiveness can be made. The final completion coefficient Xc is compared with a preset value, which is set based on empirical data. When Xc ≥ the preset value, it indicates that the user's learning effectiveness is good, and no supplementary learning content is pushed. When Xc < the preset value, it indicates that the user's learning effectiveness is poor, and different amounts of supplementary learning content are pushed based on the difference between the preset value and the final completion coefficient Xc. The specific quantity and comparison relationship between the difference can be set based on empirical data. Through this process, supplementary learning content can be adaptively provided according to the user's learning effectiveness, thereby improving the user's overall learning effectiveness.
[0023] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of this application.
Claims
1. A data-driven intelligent online course support platform, characterized in that: include: The user data collection unit is used to acquire user login data, content interaction data, and evaluation practice data. The user status analysis unit is used to determine the user's learning status based on login data and content interaction data. The learning status includes the user's learning regularity coefficient, whether the user's learning time on that day falls within the daily standard learning time range, and the learning status comparison results. The learning content analysis unit is used to determine the user's learning completion rate based on the user's learning status and content interaction data. The learning support module is used to push supplementary learning content based on the user's learning completion rate and assessment practice data.
2. The intelligent online course support platform based on data analysis according to claim 1, characterized in that, The process of judging a user's learning status includes: Retrieve the daily login count and daily study time from the login data; Obtain the number of learning pauses and the percentage of replayed content from the content interaction data; The user's learning regularity coefficient is calculated based on the number of daily logins and the daily learning duration, and the user's daily standard learning time range is calculated based on the daily learning duration. The number of times a user pauses in learning and the percentage of replayed content are compared with the corresponding standards for the learning content to obtain the learning status comparison results; Determine whether the user's learning time falls within the daily standard learning time range. Based on the determination result, the user's learning regularity coefficient, and the learning status comparison result, obtain the user's learning status reference value. Based on the learning status reference value and content interaction data, determine the user's learning completion rate.
3. The intelligent online course support platform based on data analysis according to claim 2, characterized in that, The process of calculating the user learning regularity coefficient includes: Select user's historical login data for m days; Through equations Calculate and obtain the user learning regularity coefficient Sw; Where i∈[1,m], Let i be the number of logins on day i. Let be the average number of logins over m days. Let i be the study duration on day i. Let m be the average of the study time over m days. These are the weighting coefficients; The process of obtaining a user's daily standard learning time range includes: Determine whether : If so, then the standard daily study time range will be determined as follows: ; If not, then proceed according to Eliminate from largest to smallest in descending order and then recalculate. until The standard daily study time range is set as follows: ; in, To preset the comparison standard deviation, , These represent the shortest and longest daily study time over m days, respectively. , These represent the conditions satisfied after excluding the data. The shortest and longest study time per day over several days.
4. The intelligent online course support platform based on data analysis according to claim 3, characterized in that, The process of obtaining the learning state comparison results includes: Through equations Calculate the learning state comparison value Rc and use the learning state comparison value Rc as the learning state comparison result. Where p represents the percentage of the user's replayed learning content. Let Q be the average percentage of replayed content corresponding to the user's learning content, Q be the number of pauses during the user's learning session, and t be the duration of the user's learning session. The average number of pauses per unit of time corresponding to the user's learning content. and These are the first and second defined functions, respectively. When within the corresponding preset range ,otherwise, ,when When within the corresponding preset range ,otherwise, , , These are the weighting coefficients; The process of obtaining the learning state reference value includes: Through equations The learning state reference value Z is calculated and obtained; in, For the third defined function, when t falls within the daily standard learning time range, ,otherwise, The value is obtained by referring to the preset correspondence according to the range of t.
5. The intelligent online course support platform based on data analysis according to claim 4, characterized in that, The process for determining the learning completion rate includes: Through equations Calculate and obtain the learning completion rate Cp; Where C represents the course completion rate. This is a correction factor.
6. The intelligent online course support platform based on data analysis according to claim 5, characterized in that, The calculation process for the course completion rate includes: The viewing period is divided according to the interaction time point, and D time periods are obtained, j∈[1,D]; the interaction time point includes the pause time point, the magnification adjustment time point and the timestamp adjustment time point; Through equations Calculate and obtain the course viewing completion rate C; in, The playback rate for the j-th time segment is... For the j-th time segment, the number of times the video is played. express The corresponding number of time periods, k∈[1, ], Indicates the kth The duration of the corresponding time period.
7. The intelligent online course support platform based on data analysis according to claim 6, characterized in that, The learning assistance module operates as follows: Through equations Calculate the final completion coefficient Xc and compare it with the preset value: When Xc ≥ preset value, supplementary learning content will not be pushed. When Xc < preset value, push different amounts of supplementary learning content based on the difference between the preset value and the final completion coefficient Xc; Where Cy is the user evaluation value.
8. The intelligent online course support platform based on data analysis according to claim 7, characterized in that, The calculation process for the user evaluation value includes: Through equations Calculate and obtain the user evaluation value Cy; Where E is the number of assessment questions, x∈[1,E], For the score of the xth assessment question, Let x be the score for the xth assessment question. Let x be the correlation coefficient between the x-th assessment question and the learning content.