A learning management method, system, intelligent terminal and storage medium

Abstract

The application relates to a learning management method and system, an intelligent terminal and a storage medium, and relates to the technical field of data processing.The application comprises the following steps: acquiring preset learned knowledge points of a target user; judging whether the learned knowledge points meet preset old user knowledge point learning requirements; if not, acquiring user evaluation data and target user basic information; analyzing the user evaluation data, the target user basic information and a preset user group forgetting parameter library to generate a user forgetting curve, and continuing to acquire the learned knowledge points for cyclic judgment; if yes, acquiring user review data, an adjacent review time interval and a historical user forgetting curve; correcting the historical user forgetting curve based on the user review data and the adjacent review time interval to generate a user forgetting curve; and analyzing the learned knowledge points and the user forgetting curve to generate a user review plan.The application has the effect of improving memory state detection accuracy.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a learning management method, system, smart terminal, and storage medium. Background Technology

[0002] An adaptive review system based on writing behavior and forgetting curves refers to an intelligent learning system that dynamically evaluates a user's memory status by acquiring behavioral data during the handwriting learning process and combining it with a forgetting curve model, and then generates a personalized review plan accordingly. This system is mainly applied in intelligent paper-and-pen learning scenarios to help users improve the efficiency of knowledge memorization and the focus of their reviews.

[0003] In related technologies, adaptive review systems mainly rely on the correctness or incorrectness of the user's test answers after completing a learning task as the sole feedback signal, and trigger subsequent reviews at fixed time intervals based on the quality of the test results, thereby determining the timing and content of the review; at the same time, users need to use electronic devices to record data such as the time of each review and the learning progress during the learning process.

[0004] Regarding the aforementioned technologies, when using adaptive review systems to analyze user forgetting patterns and generate review plans, they typically only obtain user answer results data. Furthermore, because the feedback method relying solely on answer accuracy cannot distinguish between the two distinct memory states of "fluent recall" and "answering correctly after hesitation," the system's accuracy in judging the user's true memory depth is insufficient, thus affecting the accuracy and personalization of the generated review plan. Additionally, since users need to use electronic devices for review, it is difficult to apply in purely paper-based learning scenarios without internet access or smart terminals, and prolonged use can negatively impact user eyesight, indicating room for improvement. Summary of the Invention

[0005] To improve the accuracy of memory state detection, this application provides a learning management method, system, smart terminal, and storage medium.

[0006] Firstly, this application provides a learning management method, which adopts the following technical solution: A learning management method, comprising: Obtain the learned knowledge points of the target users; Determine whether the learned knowledge points meet the preset learning requirements for existing users; If not, obtain user evaluation data and basic information of the target user; The system analyzes user evaluation data, target user basic information, and a pre-set user group forgetting parameter database to generate user forgetting curves and continues to acquire learned knowledge points for iterative judgment. If the conditions are met, then obtain the adjacent review time interval, the historical user forgetting curve, and the page identifier of the preset paper learning medium; Obtain user review data based on page identifiers; The historical user forgetting curves are corrected based on user review data and adjacent review time intervals to generate user forgetting curves; Analyze learned knowledge points and user forgetting curves to generate user review plans; Based on the user's forgetting curve and review plan, the preset adaptive review system performs multi-device synchronous operation.

[0007] Optionally, the steps of analyzing user evaluation data, target user basic information, and a pre-defined user group forgetting parameter library to generate user forgetting curves include: Find basic information about the user group in the user group forgetting parameter database; Substitute the target user's basic information and the user group's basic information into the preset similarity calculation formula to calculate the target user's forgotten matching similarity. The target user forgets the similarity of the match and sorts them to generate the minimum similarity of the match; Based on the minimum matching similarity, find the similar forgetting rate, similar memory stability coefficient and similar forgetting curve in the user group forgetting parameter database; Find the accuracy rate at each test time point in the user evaluation data; The similar forgetting rate, similar memory stability coefficient, test time point accuracy, and similar forgetting curve are analyzed to generate user forgetting curves.

[0008] Optionally, the steps of analyzing similar forgetting rates, similar memory stability coefficients, test time point accuracy, and similar forgetting curves to generate user forgetting curves include: The accuracy rate at the test time points is substituted into the preset least squares formula for calculation to generate the user-fitted forgetting rate. The similar forgetting rate and the user-fitted forgetting rate are substituted into a preset forgetting rate correction formula for calculation to generate the target user's initial forgetting rate. The similar memory stability coefficient is substituted into the preset memory stability coefficient correction formula for calculation to generate the initial memory stability coefficient of the target user; The similar forgetting curves are replaced based on the target user's initial forgetting rate and the target user's initial memory stability coefficient to generate the user's forgetting curve.

[0009] Optionally, the step of refining historical user forgetting curves based on user review data and adjacent review time intervals to generate user forgetting curves includes: Obtain historical user memory depth scores, historical user forgetting rates, and historical user memory stability coefficients; Find the user's review accuracy rate, review time, and number of corrections in the user's review data; The system analyzes the time interval between adjacent review sessions, the user's review accuracy, the user's review duration, and the number of corrections made by the user to generate a user memory depth score. The user's memory depth score, historical user memory depth score, historical user forgetting rate, and adjacent review time interval are substituted into the preset prediction review effect error formula to calculate the user's prediction review effect error. Calculate the ratio between the user's predicted review effect error and the preset full-score predicted review effect score to generate the user's review effect error percentage; The historical user forgetting curve is corrected based on the percentage of error in user review effectiveness, the historical user forgetting rate, and the historical user memory stability coefficient to generate a user forgetting curve.

[0010] Optionally, the steps to generate a user memory depth score include analyzing adjacent review time intervals, user review accuracy, user review duration, and number of corrections made by the user: The user's review accuracy, review time, and number of corrections are substituted into a preset review difficulty index formula to generate a user review difficulty index. The user's review difficulty index is substituted into the preset memory depth scoring formula for calculation to generate an initial user memory depth score. Determine whether the time interval between adjacent review sessions is not less than the preset review interruption time interval; If it is less than, then the initial user memory depth score will be determined as the user memory depth score; If it is not less than, then the historical user forgetting rate and historical user memory stability coefficient will be substituted into the preset memory decay coefficient formula for calculation to generate the interruption review decay coefficient. Calculate the product of the interrupted review decay coefficient and the adjacent review time interval to generate the total interrupted review decay; The initial user memory depth score and the total decay of interrupted review are substituted into the preset interrupted memory depth score formula to calculate the user memory depth score.

[0011] Optionally, the steps to correct the historical user forgetting curve based on the user's review effect error ratio, historical user forgetting rate, and historical user memory stability coefficient to generate the user forgetting curve include: Calculate the product of the user's review effect error ratio, the preset forgetting rate decay coefficient, and the historical user forgetting rate to generate the user forgetting rate update. Calculate the difference between the historical user forgetting rate and the updated user forgetting rate to generate the updated user forgetting rate; Calculate the product of the user's review effect error ratio, the preset stable decay coefficient, and the historical user memory stability coefficient to generate the user memory stability coefficient update amount; Calculate the sum of the historical user memory stability coefficient and the user memory stability coefficient update amount to generate the user updated memory stability coefficient; The historical user forgetting curves are replaced with user-updated forgetting rates and user-updated memory stability coefficients to generate user forgetting curves.

[0012] Optionally, the steps to analyze learned knowledge points and user forgetting curves to generate user review plans include: Find the user's actual forgetting rate from the user's forgetting curve; Obtain the user's actual memory depth score and the number of times the knowledge point was reviewed; The user's actual forgetting rate, the user's actual memory depth score, and the number of times the knowledge point was reviewed are substituted into a preset formula for the degree of forgetting of the knowledge point to calculate the degree of memory of the knowledge point. Based on the knowledge points already learned, the degree of knowledge point memorization is compared with the preset knowledge point memorization threshold to generate knowledge points to be reviewed and the degree of memorization of the knowledge points to be reviewed; Calculate the quotient between the degree of memorization of the knowledge points to be reviewed and the preset total degree of memorization of knowledge points, in order to generate the percentage of memorization of the knowledge points to be reviewed; The candidate knowledge points to be reviewed are ranked according to the degree of memorization of each knowledge point. The system filters candidate knowledge points to be reviewed based on a preset maximum daily review volume to generate a review plan for the user.

[0013] Secondly, this application provides a learning management system, which adopts the following technical solution: A learning management system, comprising: The acquisition module is used to acquire learned knowledge points, user evaluation data, target user basic information, user review data, time intervals between adjacent reviews, and historical user forgetting curves. A memory for storing a program of a learning management method as described in any of the preceding items; The processor and the program in the memory can be loaded and executed by the processor to implement a learning management method as described in any of the above.

[0014] Thirdly, this application provides a smart terminal, which adopts the following technical solution: A smart terminal includes a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and executed as described in any of the preceding claims.

[0015] Fourthly, this application provides a computer storage medium capable of storing corresponding programs, which facilitates improving the accuracy of memory state detection, and adopts the following technical solution: A computer-readable storage medium storing a computer program that can be loaded by a processor and executed by any of the above-described learning management methods.

[0016] In summary, this application includes at least one of the following beneficial technical effects: 1. By determining whether the learned knowledge points meet the learning requirements of existing users, if not, the system analyzes user evaluation data, target user basic information, and user group forgetting parameter database to obtain the user forgetting curve, and continues to obtain learned knowledge points for iterative judgment; if they meet the requirements, the system obtains user review data based on page identifiers, and corrects the historical user forgetting curve based on the obtained user review data and adjacent review time intervals to obtain the user forgetting curve. The system analyzes the learned knowledge points and the user forgetting curve to obtain the user review plan, and then controls the preset adaptive review system to perform multi-terminal synchronous operation based on the user forgetting curve and user review plan. Furthermore, by integrating writing process behavior data and multi-dimensional memory assessment of answer correctness, the system improves the accuracy of memory status detection. 2. By retrieving basic user information from the user group forgetting parameter database, and substituting the target user's basic information and the user group's basic information into the similarity calculation formula, the target user's forgetting matching similarity is obtained. The target user's forgetting matching similarity is sorted to obtain the minimum matching similarity. Based on the minimum matching similarity, the similar forgetting rate, similar memory stability coefficient, and similar forgetting curve are retrieved from the user group forgetting parameter database. The test time point accuracy is retrieved from the user evaluation data. Thus, the user forgetting curve is obtained by analyzing the similar forgetting rate, similar memory stability coefficient, test time point accuracy, and similar forgetting curve. Furthermore, the accuracy of memory state detection is improved by using a hybrid initialization of group statistical prior and individual test data. 3. By identifying user review accuracy, review duration, and number of corrections in user review data, and analyzing adjacent review time intervals, user review accuracy, review duration, and number of corrections, a user memory depth score is obtained. The user memory depth score, historical user memory depth scores, historical user forgetting rates, and adjacent review time intervals are substituted into the formula for predicting review effectiveness error to calculate the user's predicted review effectiveness error. The quotient of the user's predicted review effectiveness error and the full-score predicted review effectiveness score is used to obtain the user's review effectiveness error percentage. Based on the user's review effectiveness error percentage, historical user forgetting rates, and historical user memory stability coefficients, the historical user forgetting curve is corrected to obtain the user forgetting curve. Furthermore, by incorporating writing process behavior data into memory state assessment, the accuracy of memory state detection is improved. Attached Figure Description

[0017] Figure 1 This is a flowchart of a learning management method in an embodiment of this application.

[0018] Figure 2 This is a flowchart of the steps in this application embodiment to analyze user evaluation data, target user basic information, and a preset user group forgetting parameter library to generate a user forgetting curve.

[0019] Figure 3 This is a flowchart of the steps in this application embodiment to analyze similar forgetting rates, similar memory stability coefficients, test time point accuracy, and similar forgetting curves to generate user forgetting curves.

[0020] Figure 4 This is a flowchart of the steps in this application embodiment to modify the historical user forgetting curve based on user review data and adjacent review time intervals to generate a user forgetting curve.

[0021] Figure 5 This is a flowchart of the steps in this application embodiment to analyze adjacent review time intervals, user review accuracy, user review duration, and number of user corrections to generate a user memory depth score.

[0022] Figure 6 This is a flowchart of the steps in this application embodiment to correct the historical user forgetting curve based on the user's review effect error ratio, historical user forgetting rate, and historical user memory stability coefficient to generate the user forgetting curve.

[0023] Figure 7 This is a flowchart illustrating the steps in this application embodiment to analyze learned knowledge points and the user's forgetting curve to generate a user review plan. Detailed Implementation

[0024] To make the purpose, technical solution, and advantages of this application clearer, the following description is provided in conjunction with the appendix. Figures 1 to 7 The present application will be further described in detail below with reference to embodiments. It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of the application.

[0025] This application discloses a learning management method, which mainly addresses the problem of traffic light malfunctions. Specifically, it discloses a target user, an adaptive review system, and a processing terminal. The processing terminal communicates with the adaptive review system to achieve data interaction and control. The processing terminal determines whether the target user is a new user based on the learned knowledge points of the target user, determines the user's forgetting curve based on the user type and related review data, and then determines the user's review strategy based on the user's forgetting curve, thereby improving the accuracy of memory state detection.

[0026] Reference Figure 1 This application discloses a learning management method, including the following steps: Step S100: Obtain the pre-defined knowledge points already learned by the target user.

[0027] The learned knowledge points refer to the specific knowledge content that the target user has learned and stored in the processing terminal. The processing terminal searches in the user's learning database and extracts all the learned knowledge points stored in the user's learning database to obtain the learned knowledge points. In one embodiment, when the target user is a new user, the learned knowledge points are an empty set.

[0028] The target user refers to the user who is using the adaptive review system to learn and review knowledge points. This target user may be a new user who has just registered but has not yet started learning, or an old user who has already started learning and reviewing.

[0029] The user learning database refers to a database used by each target user to store information such as the knowledge points they have learned, the number of times each knowledge point has been reviewed, the review date of each review, the time of each review, the knowledge points reviewed each time, and parameters that quantify the status of each review. After each review by the target user, the processing terminal stores the review date and the knowledge points reviewed, and then counts the duplicates in all the stored knowledge points to obtain the number of times each knowledge point has been reviewed.

[0030] An adaptive review system is an intelligent learning system that acquires user behavior data during handwriting learning, combines it with a forgetting curve model to dynamically assess the user's memory status, and generates a personalized review plan accordingly. It can also be synchronized across multiple devices. This system is mainly used in smart paper-and-pen learning scenarios to help users improve the efficiency of knowledge memorization and the focus of their reviews.

[0031] Step S101: Determine whether the learned knowledge points meet the preset learning requirements for existing users.

[0032] The requirement for learning knowledge points for existing users means that the target user is an existing user when the set of learned knowledge points is not empty and the number of learned knowledge points is not less than the threshold number of learned knowledge points. In one embodiment, the threshold number of learned knowledge points is 10.

[0033] By processing the terminal, it can determine whether the learned knowledge points meet the learning requirements of existing users, thereby identifying whether the target user is an existing user or a new user, and then determining whether a learning ability assessment of the target user is necessary.

[0034] Step S1011: If not satisfied, obtain user evaluation data and basic information of the target user.

[0035] If the processing terminal determines that the learned knowledge points do not meet the learning requirements of the old user, it means that the target user is a new user and the learning ability of the target user needs to be assessed. Therefore, the processing terminal needs to determine the user assessment data and the basic information of the target user.

[0036] User evaluation data refers to a dataset used to assess the individual forgetting characteristics of users. It consists of the test accuracy rate at each test time point when the target user conducts a memory ability assessment. In one embodiment, when the target user selects English as the subject of study, the adaptive review system randomly sends multiple English words, such as 30, to the target user for memorization of word meanings and spellings. After memorization, at least three test time points are set, such as 5 minutes after the end of memorization, 30 minutes after the end of memorization, and 24 hours after the end of memorization, allowing the target user to conduct a memory test at the test time points, such as writing Chinese words from English or spelling English words from Chinese. After each test, the number of words correctly answered by the target user is recorded, and the number of correctly answered words is divided by the total number of English words to obtain the test accuracy rate at that test time point. After all the tests at all time points are completed, all test accuracy rates are summarized according to the test time points to obtain the user evaluation data.

[0037] The target user's basic information refers to the feature vector used to store the target user's age, grade, target subject, and self-assessment of the target learning subject. After a user registers an account with the adaptive review system, the system sends a basic information form to the target user. All items in the basic information form are optional. After the target user completes the form, the system reads the options selected by the target user from the basic information form, converts and summarizes the options according to the items in the form, and thus obtains the target user's basic information. The expression for the feature vector is: In the formula, This refers to the target user's age, such as when the target user is 8 years old. =0.08, This refers to the target user's grade level, with a value ranging from 0.1 to 1.2. For example, if the target user is in first grade, =0.1, the target users are first-year junior high school students. =0.7, the target users are high school seniors. =1.2, This refers to the target subject of the target user. If the target subject is English, then... =1, if the target subject is mathematics, then =2, if the target subject is Chinese, then =3, This refers to the target users' self-rating of the target subject. The value range is 0-10.

[0038] Step S10111: Analyze user evaluation data, target user basic information, and a preset user group forgetting parameter library to generate a user forgetting curve, and continue to obtain learned knowledge points for iterative judgment.

[0039] The user forgetting curve is an exponential mathematical model used to describe the decay of a target user's retention rate of learned knowledge over time. The processing terminal analyzes user evaluation data, target user basic information, and a user group forgetting parameter database to obtain the user forgetting curve. Specific methods are described in [reference needed]. Figure 2 The system continues to iterate and judge the learned knowledge points. When the learned knowledge points meet the learning requirements of the old user, it means that the target user has started to use the adaptive review system to learn. Then, it can proceed to step S1012 to determine the user's forgetting curve, and then determine the review plan for each review of the target user. After determining the user's forgetting curve, the user's forgetting curve is stored in the user learning database, which serves as a reference for determining the user's forgetting curve again. When teachers or parents need to check the target user's learning status, it is retrieved from the target user's user learning database, thereby realizing real-time tracking of the target user's learning status.

[0040] The user group forgetting parameter database refers to a database that stores basic user information and forgetting curves of multiple typical users. It is used to provide statistical priors for new users without personal learning data. The database is obtained by processing the basic information and forgetting curves of users who use the adaptive review system, classifying and summarizing the collected data according to the subject and grade.

[0041] Step S1012: If satisfied, obtain the adjacent review time interval, the historical user forgetting curve, and the preset page identifier of the paper learning medium.

[0042] If the processing terminal determines that the learned knowledge points meet the learning requirements of the old user, it means that the target user is an old user and there is no need to assess the learning ability of the target user. Therefore, the processing terminal needs to determine the time interval between adjacent review sessions, the forgetting curve of historical users, and the page identifier of the paper learning medium.

[0043] Paper-based learning media refers to physical tools that use paper as the primary medium to carry and disseminate knowledge and information. Each page in a paper-based learning media is equipped with a unique page identifier, allowing users to determine the content they need to review and the corresponding page number based on the page identifier.

[0044] Page identifiers refer to the set of identifiers on the pages of the paper learning medium where the knowledge points that the target user needs to review are located. In one embodiment, page identifiers include page numbers, identifiable identifiers such as QR codes and barcodes set on the page, or feature image features extracted from the page image. When the page identifier is a QR code set on the page, multiple methods for obtaining page identifiers are supported to adapt to the usage needs of different target users. Specifically, these include: when the target user only has a wristband or watch, they first record their writing behavior during the review process using the wristband or watch, and then use their mobile phone to scan the QR code on the page corresponding to the knowledge point to be reviewed in the paper learning medium to obtain the page identifier; when the user only has a watch, but the watch has an identifier scanning function, they directly use their smartwatch to scan the QR code on the page corresponding to the knowledge point to be reviewed in the paper learning medium to obtain the page identifier; when the target user can use a mobile phone with internet access, they can directly scan the QR code on the page corresponding to the knowledge point to be reviewed in the paper learning medium to obtain the page identifier.

[0045] The adjacent review interval refers to the number of days between two consecutive reviews when a target user reviews the knowledge points of the same subject. When the target user is reviewing for the first time, the adjacent review interval is the number of days between the first learning time and the first review time. The processing terminal extracts the dates of the two most recent reviews from the user's learning database according to their time proximity, and subtracts the two dates to obtain the adjacent review interval.

[0046] The historical user forgetting curve refers to the most recently updated, stored data that describes the decay of the target user's memory retention rate of the learned knowledge over time. The processing terminal can obtain the historical user forgetting curve by searching the user's learning database.

[0047] Step S10121: Obtain user review data based on page identifier.

[0048] Herein, the user review data refers to behavior data generated when a target user reviews learned knowledge points in the most recent review, which consists of the user's review duration, the user's review accuracy rate, and the user's erasure count. In one embodiment, when the target user is in a scenario where electronic devices cannot be used, such as a school, the target user needs to fill the knowledge points to be reviewed in advance into a dedicated paper learning carrier with a page identification mark on each page by means of handwriting, pasting or printing, etc. The page identification mark is exemplified by a two-dimensional code, and the user scans the two-dimensional code to enter the adaptive review system, and stores the knowledge points written on each page under the corresponding page number respectively. The storage manner includes manual input, photographing, etc., so as to adapt to different types of knowledge points. During the review process, the target user manually records the review duration, review accuracy rate and user erasure count of each knowledge point to be reviewed, and when the condition for uploading data is met, the user scans the two-dimensional code on the current page to enter the adaptive review system, and fills in the review duration, review accuracy rate and user erasure count corresponding to each knowledge point on the current page into the corresponding positions. The user erasure count may be determined by means of post-review self-assessment, for example, the erasure count is divided into three grades: 0-1 time, 2-5 times and more than 5 times, which are self-selected by the target user; or during the review process, a stroke of the Chinese character "zheng" (counting character) is drawn for each erasure, and when the review of the current knowledge point is finished, the total number of strokes is counted to obtain the user erasure count.

[0049] The user review duration refers to a data set of the total duration required by the target user in this review and the respective duration for reviewing each knowledge point; the user review accuracy rate refers to a data set of the accuracy rate of the review questions corresponding to each knowledge point done by the target user after the end of this review; the user erasure count refers to the number of modifications for each review question when the user does the review questions corresponding to each knowledge point.

[0050] Step S10122: Correct the historical user forgetting curve based on the user review data and the adjacent review time interval, so as to generate a user forgetting curve.

[0051] Herein, the user forgetting curve in this step is consistent with the user forgetting curve in the above step S10111. The processing terminal corrects the historical user forgetting curve based on the user review data and the adjacent review time interval to obtain the user forgetting curve, and the specific method refers to Figure 4 the foregoing step. After the user forgetting curve is determined, the user forgetting curve is stored in the user learning database, so as to be used as the historical user forgetting curve when the user forgetting curve is determined next time, and is retrieved from the user learning database of the target user when a teacher or parent needs to inquire about the learning situation of the target user, thereby realizing real-time tracking of the learning situation of the target user.

[0052] Step S10123: Analyze the learned knowledge points and the user forgetting curve to generate a user review plan.

[0053] The user review plan refers to the set of knowledge points that the target user needs to review next time, along with a list of the order in which these points should be reviewed. The processing terminal analyzes the learned knowledge points and the user's forgetting curve to obtain the user review plan. For specific methods, please refer to [link / reference needed]. Figure 7 The process involves determining the user's review plan and storing it in the user's learning database. When the target user reviews next time, the system will display the knowledge points to be reviewed based on the user's review plan.

[0054] Step S10124: Control the preset adaptive review system to perform multi-terminal synchronous operation based on the user's forgetting curve and user review plan.

[0055] Each time a target user completes a review session, the processing terminal uploads the updated user forgetting curve and the determined review plan for the next session as a change record to the cloud database. After the cloud database establishes a communication connection with the parent and teacher terminals, the processing terminal uses a message push mechanism to synchronously display the change instructions containing the latest user forgetting curve and review plan on the parent, student, and teacher terminals for parsing and display. This allows parents and teachers to track the target user's learning and review effectiveness in real time and provides the target user with a clear understanding after synchronization.

[0056] The adaptive review system in this step is the same as the adaptive review system in step S100 above, and will not be described again here.

[0057] Reference Figure 2 The steps for generating user forgetting curves include analyzing user evaluation data, basic information of target users, and a pre-set user group forgetting parameter database: Step S200: Find the basic information of the user group in the user group forget parameter database.

[0058] The basic information of the user group refers to the feature vector used to store the target user's age, grade, target subject, and self-rating of the target learning subject for multiple typical users. The processing terminal searches in the user group forgetting parameter database to obtain the basic information of the user group. The feature vector expression of each typical user is consistent with the values ​​of each component in the expression of the target user's basic information in step S1011 above. The feature vector expression of each typical user is as follows: In the formula, It refers to the first The age of a typical user It refers to the first The grade level of a typical user It refers to the first The target discipline of a typical user It refers to the first A typical user's self-rating of the target subject. This refers to the first item in the basic information of the target user or the basic information of the user group. Each component.

[0059] Step S201: Substitute the target user's basic information and the user group's basic information into the preset similarity calculation formula to calculate the target user's forgotten matching similarity.

[0060] Among them, the target user forgetting match similarity refers to the Euclidean distance between the target user's basic information and the user group's basic information. It is used to measure the closeness between the target user and typical users in the database in terms of the target user's age, grade, target subject, and self-rating of the target learning subject. The smaller the similarity value, the higher the similarity. The processing terminal substitutes the target user's basic information and the user group's basic information into the similarity calculation formula to calculate the target user forgetting match similarity.

[0061] The formula for calculating similarity is: .

[0062] In the formula, This refers to the target user's forgotten matching similarity. This refers to the total number of items in the target user's basic information; for example, 4. This refers to the first item in the target user's basic information. One portion, This refers to the first item in the basic information of the user group. The first typical user Each component.

[0063] Step S202: Sort the target user forgotten match similarity to generate the minimum match similarity.

[0064] The minimum matching similarity refers to the basic information of the user group that has the highest similarity to the basic information of the target user. The processing terminal sorts the target user's forgetting matching similarity in ascending order to form a similarity sequence, and the value at the first position in the sequence is the minimum matching similarity.

[0065] Step S203: Based on the minimum matching similarity, find the similar forgetting rate, similar memory stability coefficient, and similar forgetting curve in the user group forgetting parameter database.

[0066] Among them, the similarity forgetting rate refers to the decay coefficient in the similarity forgetting curve. The similarity forgetting rate determines the rate at which a user's memory retention rate decreases over time. The higher the value of the similarity forgetting rate, the faster the user forgets the knowledge points. The similarity memory stability coefficient is a data used to quantify the strength of a user's memory of knowledge points. The higher the value of the similarity memory stability coefficient, the more stable the user's memory of knowledge points and the stronger the ability to resist forgetting. The similarity forgetting curve is an exponential mathematical model used to describe the decay law of the memory retention rate of the learned knowledge of the typical user with the lowest similarity over time. The processing terminal searches for the similarity forgetting curve in the user group forgetting parameter database based on the minimum matching similarity, and then extracts data from the corresponding position of the expression of the similarity forgetting curve to obtain the similarity forgetting rate and the similarity memory stability coefficient. The expression of all forgetting curves is: .

[0067] In the formula, This refers to the forgetting curve. This refers to the rate of forgetting. This refers to review time. This refers to the memory stability coefficient. The differences in the expressions of the forgetting curves of different users lie in the different forgetting rates and memory stability coefficients.

[0068] Step S204: Find the accuracy rate of the test time points in the user evaluation data.

[0069] Among them, the test time point accuracy refers to the set of test accuracy data of the target user at each test time point when evaluating the individual forgetting characteristics of the target user. The processing terminal can obtain the test time point accuracy by searching in the user evaluation data.

[0070] Step S205: Analyze the similar forgetting rate, similar memory stability coefficient, test time point accuracy, and similar forgetting curve to generate the user forgetting curve.

[0071] The user forgetting curve in this step is consistent with the user forgetting curve in step S10111 above. The processing terminal analyzes the similar forgetting rate, similar memory stability coefficient, test time point accuracy, and similar forgetting curve to obtain the user forgetting curve. For specific methods, refer to [link to relevant documentation]. Figure 3 The steps.

[0072] Reference Figure 3 The steps for generating user forgetting curves by analyzing similar forgetting rates, similar memory stability coefficients, test time point accuracy, and similar forgetting curves include: Step S300: Substitute the accuracy of the test time points into the preset least squares formula for calculation to generate the user-fitted forgetting rate.

[0073] The user-fitted forgetting rate refers to the data obtained by fitting the accuracy of all test time points after evaluating the memory ability of the target user. It is used to quantify how fast the target user forgets. The processing terminal substitutes the accuracy of the test time points into the least squares formula to calculate the user-fitted forgetting rate.

[0074] The least squares formula is: .

[0075] In the formula, This refers to the user's fitted forgetting rate. This refers to the total number of test time points; for example, 3. It refers to the first The number of days between each test time point and the end of the test. For example, if there are three test time points, 5 minutes, 30 minutes, and 24 hours after the end of the memory period, respectively, then... , , , It refers to the first The accuracy of each test time point.

[0076] Step S301: Substitute the similar forgetting rate and the user-fitted forgetting rate into the preset forgetting rate correction formula to calculate the initial forgetting rate for the target user.

[0077] The initial forgetting rate of the target user refers to the rate at which the user's memory retention rate decreases over time, predicted based on similar typical users and the results of memory assessment. The processing terminal substitutes the similar forgetting rate and the user's fitted forgetting rate into the forgetting rate correction formula to calculate the initial forgetting rate of the target user.

[0078] The formula for correcting the rate of forgetting is: .

[0079] In the formula, This refers to the initial forgetting rate of the target user. This refers to similar forgetting rates. This refers to the user's fitted forgetting rate. This refers to the typical user parameter weight, taking 0.7 as an example.

[0080] Step S302: Substitute the similar memory stability coefficient into the preset memory stability coefficient correction formula for calculation to generate the target user's initial memory stability coefficient.

[0081] Among them, the target user's initial memory stability coefficient refers to the data predicted based on similar typical users and the results of memory assessment, which is used to quantify the strength of users' memory of knowledge points. The processing terminal substitutes the similar memory stability coefficient into the memory stability coefficient correction formula to calculate and generate the target user's initial memory stability coefficient.

[0082] The formula for correcting the memory stability coefficient is: .

[0083] In the formula, This refers to the initial memory stability coefficient of the target user. This refers to the similarity memory stability coefficient. This refers to the default base memory stability coefficient, in order to Taking =1 as an example, This refers to the typical user parameter weight, taking 0.7 as an example.

[0084] Step S303: Replace the similar forgetting curves with the target user's initial forgetting rate and the target user's initial memory stability coefficient to generate the user's forgetting curve.

[0085] In this step, the user forgetting curve is consistent with the user forgetting curve in step S10111 above. The processing terminal replaces the similar memory stability coefficient and similar forgetting rate in the similar forgetting curve with the target user's initial memory stability coefficient and the target user's initial forgetting rate to obtain the user forgetting curve.

[0086] Reference Figure 4 The steps for generating user forgetting curves by refining historical user forgetting curves based on user review data and adjacent review time intervals include: Step S400: Obtain historical user memory depth score, historical user forgetting rate, and historical user memory stability coefficient.

[0087] Among them, the historical user memory depth score refers to the evaluation value of the target user's behavioral performance data and the correctness of the answers during the most recent review; the historical user forgetting rate refers to the forgetting rate generated during the most recent review; and the historical user memory stability coefficient refers to the memory stability coefficient generated during the most recent review. The processing terminal can obtain the historical user memory depth score, historical user forgetting rate, and historical user memory stability coefficient by searching in the user learning database.

[0088] Step S401: Find the user's review accuracy, review time, and number of corrections in the user's review data.

[0089] In this step, the user review accuracy rate, user review time, and number of user corrections are the same as those in step S1012 above. The processing terminal can obtain the user review accuracy rate, user review time, and number of user corrections by searching in the user review data.

[0090] Step S402: Analyze the time interval between adjacent review sessions, the user's review accuracy, the user's review duration, and the number of corrections made by the user to generate a user memory depth score.

[0091] The user memory depth score refers to the evaluation value used to quantify the target user's behavioral performance data and the correctness of their answers during the question-answering process. The processing terminal analyzes adjacent review time intervals, user review accuracy, user review duration, and the number of corrections made by the user to obtain the user memory depth score. For specific methods, please refer to [link / reference needed]. Figure 5 The steps.

[0092] Step S403: Substitute the user's memory depth score, historical user memory depth score, historical user forgetting rate, and adjacent review time interval into the preset prediction review effect error formula to calculate the user's prediction review effect error.

[0093] The user's predicted review effect error refers to the average deviation between the target user's actual memory depth score and the predicted memory depth score. The processing terminal substitutes the user's memory depth score, historical user memory depth score, historical user forgetting rate, and adjacent review time interval into the predicted review effect error formula to calculate the user's predicted review effect error.

[0094] The formula for predicting the error of review effectiveness is: .

[0095] In the formula, This refers to the error in users' predictions of review effectiveness. This refers to the number of knowledge points that the target user needs to review in this study session. The processing terminal retrieves the user's review plan from the user's learning database, identifies the knowledge points that need to be reviewed within the plan, and finally counts these knowledge points to obtain the total number of knowledge points. It refers to the first The user's memory depth score corresponds to the knowledge points reviewed by each target user. It refers to the first The ideal user memory depth score corresponding to the knowledge points reviewed by each target user. This refers to the historical user memory depth score corresponding to the knowledge points reviewed by the target user. This refers to the historical forgetting rate of the knowledge points reviewed by the target user. This refers to the time interval between adjacent review sessions.

[0096] Step S404: Calculate the quotient between the user's predicted review effect error and the preset full-score predicted review effect score to generate the user's review effect error percentage.

[0097] Among them, the user review effect error ratio refers to the proportion of the user's predicted review effect error to the full score. It is used to determine the magnitude of the correction of the forgetting curve. The processing terminal divides the user's predicted review effect error by the full score predicted review effect to obtain the user review effect error ratio.

[0098] The perfect score for predicting the review effect refers to the value when the user's memory depth score is perfect. In one embodiment, the perfect score for predicting the review effect is 100.

[0099] Step S405: Based on the error ratio of user review effect, historical user forgetting rate and historical user memory stability coefficient, the historical user forgetting curve is corrected to generate the user forgetting curve.

[0100] In this step, the user forgetting curve is consistent with the user forgetting curve in step S10111 above. The processing terminal corrects the historical user forgetting curve based on the user's review effect error ratio, historical user forgetting rate, and historical user memory stability coefficient to generate the user forgetting curve. The specific method is as follows: Figure 6 The steps.

[0101] Reference Figure 5 The steps to generate a user memory depth score by analyzing adjacent review time intervals, user review accuracy, user review duration, and number of corrections include: Step S500: Substitute the user's review accuracy, user's review time, and user's number of corrections into the preset review difficulty index formula to calculate and generate the user's review difficulty index.

[0102] The User Review Difficulty Index is a data point used to quantify the mental effort required for a target user to review each knowledge point. The higher the value of the User Review Difficulty Index, the more difficult it is for the target user to review. The processing terminal calculates the User Review Difficulty Index by substituting the user's review accuracy, review time, and number of corrections into the review difficulty index formula.

[0103] The formula for the difficulty index of review is: .

[0104] In the formula, This refers to the user's difficulty index in reviewing the material. This refers to the user reviewing the first... User review time for each knowledge point This refers to the user reviewing the first... The user's correct review rate for each knowledge point This refers to the weighting of user review accuracy. When the user review accuracy for each knowledge point is not less than 50%, =1, when the user's correct review rate for each knowledge point is less than 50%. =0, It refers to the first In one embodiment, the standard time taken for each knowledge point is as follows: when the target user's subject is English, and the English words are long words with more than 9 letters... =7s, for English words that are medium to long, consisting of 6-8 letters. =5s, when the English word is a short word of 3-5 letters. =3s, This refers to the user reviewing the first... Number of times a user has edited a knowledge point This refers to the maximum number of times a knowledge point can be corrected; for example, 5.

[0105] Step S501: Substitute the user's review difficulty index into the preset memory depth scoring formula to generate an initial user memory depth score.

[0106] The initial user memory depth score refers to the data on the target user's behavior and the evaluation of the correctness of their answers when reviewing each knowledge point, without considering the error caused by the user's possible interruption of review. The processing terminal substitutes the user's review difficulty index into the memory depth score formula to calculate the initial user memory depth score.

[0107] The formula for memory depth scoring is: .

[0108] In the formula, This refers to the initial user memory depth score. This refers to the historical user memory depth score corresponding to the knowledge points reviewed by the target user. It refers to the first The user's review difficulty index corresponding to each knowledge point This refers to the user's review difficulty threshold. For example, a threshold of 1.0 indicates that when... If the user's accuracy rate on all test questions related to that knowledge point is less than 30%, it indicates that the user has not mastered that knowledge point.

[0109] Step S502: Determine whether the time interval between adjacent review sessions is not less than the preset review interruption time interval.

[0110] The review interruption time interval refers to the time interval threshold used to determine whether the target user has interrupted their review. In one embodiment, the review interruption time interval is 48 hours.

[0111] The processing terminal determines whether the adjacent review time interval is not less than the review interruption time interval, thereby determining whether the target user has interrupted their review.

[0112] Step S5021: If it is less than, then the initial user memory depth score is determined as the user memory depth score.

[0113] If the processing terminal determines that the adjacent review time interval is less than the review interruption time interval, it means that the target user has not interrupted the review. Therefore, the processing terminal determines the initial user memory depth score as the user memory depth score.

[0114] Step S5022: If it is not less than, then substitute the historical user forgetting rate and the historical user memory stability coefficient into the preset memory decay coefficient formula to calculate the interruption review decay coefficient.

[0115] If the processing terminal determines that the adjacent review time interval is not less than the review interruption time interval, it indicates that the target user has interrupted the review. Therefore, the processing terminal needs to determine the interruption review decay coefficient to provide data support for subsequently determining the user's memory depth score.

[0116] The interruption review decay coefficient refers to the actual rate at which each knowledge point to be reviewed is forgotten when learning is interrupted. The processing terminal substitutes the historical user forgetting rate and the historical user memory stability coefficient into the memory decay coefficient formula to calculate the interruption review decay coefficient.

[0117] The formula for the memory decay coefficient is: .

[0118] In the formula, It refers to the first The attenuation coefficient of interrupted review of each knowledge point This refers to the historical user forgetting rate. This refers to the historical user memory stability coefficient. This refers to the total number of effective learning days for the target user, which is obtained by the processing terminal by searching the user's learning database.

[0119] Step S50221: Calculate the product of the interrupted review attenuation coefficient and the adjacent review time interval to generate the total interrupted review attenuation.

[0120] The total attenuation of interrupted review refers to the data used to quantify the total amount of forgetting of each knowledge point to be reviewed during the entire interruption period. The processing terminal multiplies the interrupted review attenuation coefficient by the adjacent review time interval to obtain the total attenuation of interrupted review.

[0121] Step S50222: Substitute the initial user memory depth score and the total decay of interrupted review into the preset interrupted memory depth score formula to calculate and generate the user memory depth score.

[0122] In this step, the user memory depth score is the same as the user memory depth score in step S402 above. The processing terminal substitutes the initial user memory depth score and the total attenuation of interrupted review into the interrupted memory depth score formula to calculate the user memory depth score.

[0123] The formula for scoring the depth of interrupted memory is: .

[0124] In the formula, This refers to the user's memory depth rating. This refers to the initial user memory depth score. It refers to the first The attenuation coefficient of interrupted review of each knowledge point This refers to the time interval between adjacent review sessions.

[0125] Reference Figure 6 The steps for generating user forgetting curves by correcting historical user forgetting curves based on the percentage of errors in user review effectiveness, historical user forgetting rates, and historical user memory stability coefficients include: Step S600: Calculate the product of the user's review effect error ratio, the preset forgetting rate decay coefficient, and the historical user forgetting rate to generate the user forgetting rate update.

[0126] The user forgetting rate update refers to the adjustment amount made by adjusting the historical user forgetting rate based on the target user's current review status. The processing terminal multiplies the user review effect error ratio by the forgetting rate decay coefficient and the historical user forgetting rate to obtain the user forgetting rate update.

[0127] The forgetting rate decay coefficient is an adjustment parameter that adjusts the forgetting rate of historical users based on the target user's current review progress. The processing terminal uses a formula... By performing calculations, the forgetting rate decay coefficient can be obtained, where, This refers to the rate of forgetting decay coefficient. This refers to the initial forgetting rate decay coefficient; for example, 0.2. This refers to the number of valid learning days for the target user, which the processing terminal retrieves from the user's learning database.

[0128] Step S601: Calculate the difference between the historical user forgetting rate and the updated user forgetting rate to generate the updated user forgetting rate.

[0129] The user-updated forgetting rate refers to the user forgetting rate after adjusting the historical user forgetting rate based on the target user's current review status. The processing terminal subtracts the user forgetting rate update from the historical user forgetting rate to obtain the user-updated forgetting rate.

[0130] Step S602: Calculate the product of the user's review effect error ratio, the preset stable decay coefficient, and the historical user memory stability coefficient to generate the user memory stability coefficient update amount.

[0131] The update amount of the user memory stability coefficient refers to the adjustment amount of the historical user memory stability coefficient based on the current review status of the target user. The processing terminal multiplies the user review effect error ratio with the stability decay coefficient and the historical user memory stability coefficient to obtain the update amount of the user memory stability coefficient.

[0132] The stability decay coefficient refers to the adjustment parameter used to adjust the historical user's memory stability coefficient based on the target user's current review status. In one embodiment, the stability decay coefficient is half of the forgetting rate decay coefficient.

[0133] Step S603: Calculate the sum of the historical user memory stability coefficient and the user memory stability coefficient update amount to generate the user updated memory stability coefficient.

[0134] The user-updated memory stability coefficient refers to the memory stability coefficient after adjusting the historical user memory stability coefficient based on the target user's current review status. The processing terminal adds the historical user memory stability coefficient to the user memory stability coefficient update amount to obtain the user-updated memory stability coefficient.

[0135] Step S604: Replace the historical user forgetting curve with the user's updated forgetting rate and the user's updated memory stability coefficient to generate a user forgetting curve.

[0136] In this step, the user forgetting curve is consistent with the user forgetting curve in step S10111 above. The processing terminal replaces the historical user forgetting rate and historical user memory stability coefficient in the historical user forgetting curve with the user update forgetting rate and user update memory stability coefficient to obtain the user forgetting curve.

[0137] Reference Figure 7 The steps involved in analyzing learned knowledge points and users' forgetting curves to generate user review plans include: Step S700: Find the user's actual forgetting rate in the user forgetting curve.

[0138] The actual forgetting rate of the user refers to the forgetting rate of the current target user. The processing terminal can find the actual forgetting rate of the user by searching in the user forgetting curve. The actual forgetting rate of the user in this step is consistent with the user update forgetting rate in step S601 above.

[0139] Step S701: Obtain the user's actual memory depth score and the number of times the knowledge points have been reviewed.

[0140] The user's actual memory depth score refers to the evaluation value used to quantify the target user's behavioral performance data and the correctness of their answers during the most recent review process. The user's actual memory depth score in this step is the same as the user memory depth score in step S402 above, and will not be repeated here.

[0141] The number of times a knowledge point is reviewed refers to the total number of times a target user reviews each knowledge point, which is obtained by the processing terminal by searching the user's learning database.

[0142] Step S702: Substitute the user's actual forgetting rate, the user's actual memory depth score, and the number of times the knowledge point was reviewed into the preset knowledge point forgetting degree formula to calculate the knowledge point memory degree.

[0143] Among them, the knowledge point retention level refers to the data used to quantify the estimated retention level of each learned knowledge point by the target user. The processing terminal substitutes the user's actual forgetting rate, the user's actual memory depth score and the number of times the knowledge point is reviewed into the knowledge point forgetting level formula to calculate the knowledge point retention level.

[0144] The formula for the degree of forgetting of knowledge points is: .

[0145] In the formula, It refers to the first The degree of memorization of each knowledge point This refers to the user's actual memory depth score. This refers to the user's actual rate of forgetting. It refers to the first The total effective learning days for each knowledge point are obtained by the processing terminal by searching the user's learning database. It refers to the first Number of times to review each knowledge point.

[0146] Step S703: Based on the learned knowledge points, compare the degree of knowledge point memorization with the preset knowledge point memorization threshold to generate knowledge points to be reviewed and the degree of memorization of the knowledge points to be reviewed.

[0147] Among them, the knowledge points to be reviewed refer to the set of knowledge points that the target user has a relatively low level of memory among the already learned knowledge points; the memory level of the knowledge points to be reviewed refers to the memory level of the knowledge points that the target user has a relatively low level of memory among the already learned knowledge points. The processing terminal compares the memory level of the knowledge points with the memory level threshold of the knowledge points based on the already learned knowledge points, extracts and summarizes the knowledge points that are lower than the memory level threshold, and thus obtains the knowledge points to be reviewed. The memory level of the knowledge points corresponding to these knowledge points is then extracted and summarized to obtain the memory level of the knowledge points to be reviewed.

[0148] The knowledge point memorization threshold refers to the value used to filter out knowledge points that the target user has a lower memorization level from the learned knowledge points based on the memorization level. In one embodiment, the knowledge point memorization threshold is 60.

[0149] Step S704: Calculate the quotient of the degree of memorization of the knowledge points to be reviewed and the preset total degree of memorization of knowledge points to generate the percentage of the degree of memorization of the knowledge points to be reviewed.

[0150] The percentage of knowledge points to be reviewed refers to a set of data used to quantify the urgency of reviewing each knowledge point. The smaller the percentage of knowledge points to be reviewed, the stronger the user's forgetting of that knowledge point, and therefore the higher the urgency of reviewing that knowledge point. The processing terminal divides the percentage of knowledge points to be reviewed by the total knowledge points to be reviewed to obtain the percentage of knowledge points to be reviewed.

[0151] The total knowledge point retention rate refers to the degree to which a user fully masters a knowledge point. In one embodiment, the total knowledge point retention rate is 100.

[0152] Step S705: Sort the knowledge points to be reviewed according to the percentage of memorization of each knowledge point to be reviewed, so as to generate candidate knowledge points to be reviewed.

[0153] Among them, the candidate knowledge points to be reviewed refer to the knowledge points to be reviewed after being sorted by their urgency. The processing terminal sorts the knowledge points to be reviewed in ascending order based on the percentage of memorization of the knowledge points to be reviewed. The generated knowledge points to be reviewed, sorted by the percentage of memorization of the knowledge points to be reviewed, are the candidate knowledge points to be reviewed.

[0154] Step S706: Based on the preset maximum daily review volume, the candidate knowledge points to be reviewed are screened to generate a user review plan.

[0155] In this step, the user review plan is consistent with the user review plan in step S10122 above. When the processing terminal determines that the number of candidate knowledge points to be reviewed is greater than the maximum daily review volume, the candidate knowledge points to be reviewed are filtered from front to back according to the maximum daily review volume. The filtered knowledge point sequence with order is the user review plan. If the number of candidate knowledge points to be reviewed is not greater than the maximum daily review volume, then the candidate knowledge points to be reviewed are the user review plan.

[0156] Based on the same inventive concept, embodiments of this application provide a learning management system, including: The acquisition module is used to acquire learned knowledge points, user evaluation data, target user basic information, user review data, adjacent review time intervals, historical user forgetting curves, historical user memory depth scores, historical user forgetting rates, historical user memory stability coefficients, user actual memory depth scores, and the number of times knowledge points have been reviewed. Memory used to store a program for a learning management method; The processor is a system that allows programs in memory to be loaded and executed by the processor, enabling a learning management method.

[0157] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional modules is used as an example. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device, and unit described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0158] This application provides a computer-readable storage medium storing a computer program that can be loaded by a processor and executed as a learning management method.

[0159] Computer storage media include, for example, USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, optical disks, and other media that can store program code.

[0160] Based on the same inventive concept, embodiments of this application provide a smart terminal, including a memory and a processor, wherein the memory stores a computer program that can be loaded and executed by the processor to perform a learning management method.

[0161] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional modules is used as an example. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device, and unit described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0162] The above are all preferred embodiments of this application and are not intended to limit the scope of protection of this application. Any feature disclosed in this specification (including the abstract and drawings) may be replaced by other equivalent or similar features unless specifically stated otherwise. That is, unless specifically stated otherwise, each feature is only one example of a series of equivalent or similar features.

Claims

1. A learning management method, characterized in that, include: Obtain the learned knowledge points of the target users; Determine whether the learned knowledge points meet the preset learning requirements for existing users; If not, obtain user evaluation data and basic information of the target user; The system analyzes user evaluation data, target user basic information, and a pre-set user group forgetting parameter database to generate user forgetting curves and continues to acquire learned knowledge points for iterative judgment. If the conditions are met, then obtain the adjacent review time interval, the historical user forgetting curve, and the page identifier of the preset paper learning medium; Obtain user review data based on page identifiers; The historical user forgetting curves are corrected based on user review data and adjacent review time intervals to generate user forgetting curves; Analyze learned knowledge points and user forgetting curves to generate user review plans; Based on the user's forgetting curve and review plan, the preset adaptive review system performs multi-device synchronous operation.

2. The learning management method according to claim 1, characterized in that, The steps for generating user forgetting curves include analyzing user evaluation data, target user basic information, and a pre-defined user group forgetting parameter database: Find basic information about the user group in the user group forgetting parameter database; Substitute the target user's basic information and the user group's basic information into the preset similarity calculation formula to calculate the target user's forgotten matching similarity. The target user forgets the similarity of the match and sorts them to generate the minimum similarity of the match; Based on the minimum matching similarity, find the similar forgetting rate, similar memory stability coefficient and similar forgetting curve in the user group forgetting parameter database; Find the accuracy rate at each test time point in the user evaluation data; The similar forgetting rate, similar memory stability coefficient, test time point accuracy, and similar forgetting curve are analyzed to generate user forgetting curves.

3. The learning management method according to claim 2, characterized in that, The steps for generating user forgetting curves by analyzing similar forgetting rates, similar memory stability coefficients, test time point accuracy, and similar forgetting curves include: The accuracy rate at the test time points is substituted into the preset least squares formula for calculation to generate the user-fitted forgetting rate. The similar forgetting rate and the user-fitted forgetting rate are substituted into a preset forgetting rate correction formula for calculation to generate the target user's initial forgetting rate. The similar memory stability coefficient is substituted into the preset memory stability coefficient correction formula for calculation to generate the initial memory stability coefficient of the target user; The similar forgetting curves are replaced based on the target user's initial forgetting rate and the target user's initial memory stability coefficient to generate the user's forgetting curve.

4. The learning management method according to claim 1, characterized in that, The steps for refining historical user forgetting curves based on user review data and adjacent review time intervals to generate user forgetting curves include: Obtain historical user memory depth scores, historical user forgetting rates, and historical user memory stability coefficients; Find the user's review accuracy rate, review time, and number of corrections in the user's review data; The system analyzes the time interval between adjacent review sessions, the user's review accuracy, the user's review duration, and the number of corrections made by the user to generate a user memory depth score. The user's memory depth score, historical user memory depth score, historical user forgetting rate, and adjacent review time interval are substituted into the preset prediction review effect error formula to calculate the user's prediction review effect error. Calculate the ratio between the user's predicted review effect error and the preset full-score predicted review effect score to generate the user's review effect error percentage; The historical user forgetting curve is corrected based on the percentage of error in user review effectiveness, the historical user forgetting rate, and the historical user memory stability coefficient to generate a user forgetting curve.

5. The learning management method according to claim 4, characterized in that, The steps to generate a user memory depth score by analyzing adjacent review time intervals, user review accuracy, user review duration, and number of corrections include: The user's review accuracy, review time, and number of corrections are substituted into a preset review difficulty index formula to generate a user review difficulty index. The user's review difficulty index is substituted into the preset memory depth scoring formula for calculation to generate an initial user memory depth score. Determine whether the time interval between adjacent review sessions is not less than the preset review interruption time interval; If it is less than, then the initial user memory depth score will be determined as the user memory depth score; If it is not less than, then the historical user forgetting rate and historical user memory stability coefficient will be substituted into the preset memory decay coefficient formula for calculation to generate the interruption review decay coefficient. Calculate the product of the interrupted review decay coefficient and the adjacent review time interval to generate the total interrupted review decay; The initial user memory depth score and the total decay of interrupted review are substituted into the preset interrupted memory depth score formula to calculate the user memory depth score.

6. The learning management method according to claim 4, characterized in that, The steps for generating a user forgetting curve by correcting the historical user forgetting curve based on the percentage of errors in user review effectiveness, the historical user forgetting rate, and the historical user memory stability coefficient include: Calculate the product of the user's review effect error ratio, the preset forgetting rate decay coefficient, and the historical user forgetting rate to generate the user forgetting rate update. Calculate the difference between the historical user forgetting rate and the updated user forgetting rate to generate the updated user forgetting rate; Calculate the product of the user's review effect error ratio, the preset stable decay coefficient, and the historical user memory stability coefficient to generate the user memory stability coefficient update amount; Calculate the sum of the historical user memory stability coefficient and the user memory stability coefficient update amount to generate the user updated memory stability coefficient; The historical user forgetting curves are replaced with user-updated forgetting rates and user-updated memory stability coefficients to generate user forgetting curves.

7. The learning management method according to claim 1, characterized in that, The steps involved in analyzing learned knowledge points and users' forgetting curves to generate user review plans include: Find the user's actual forgetting rate from the user's forgetting curve; Obtain the user's actual memory depth score and the number of times the knowledge point was reviewed; The user's actual forgetting rate, the user's actual memory depth score, and the number of times the knowledge point was reviewed are substituted into a preset formula for the degree of forgetting of the knowledge point to calculate the degree of memory of the knowledge point. Based on the knowledge points already learned, the degree of knowledge point memorization is compared with the preset knowledge point memorization threshold to generate knowledge points to be reviewed and the degree of memorization of the knowledge points to be reviewed; Calculate the quotient between the degree of memorization of the knowledge points to be reviewed and the preset total degree of memorization of knowledge points, in order to generate the percentage of memorization of the knowledge points to be reviewed; The candidate knowledge points to be reviewed are ranked according to the degree of memorization of each knowledge point. The system filters candidate knowledge points to be reviewed based on a preset maximum daily review volume to generate a review plan for the user.

8. A learning management system, characterized in that, include: The acquisition module is used to acquire learned knowledge points, user evaluation data, target user basic information, user review data, time intervals between adjacent reviews, and historical user forgetting curves. A memory for storing a program of a learning management method as described in any one of claims 1 to 7; The processor and the program in the memory can be loaded and executed by the processor to implement a learning management method as described in any one of claims 1 to 7.

9. A smart terminal, characterized in that, It includes a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and executed as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer program is stored that can be loaded by a processor and executed as any one of claims 1 to 7.

Images