Learning planning method and device and computer storage medium
By dynamically updating the memory strength S based on the time interval of learning events and the test accuracy, and combining this with the preset memory retention rate R to plan learning time, the problem of poor review effect caused by fixed memory strength in the forgetting curve model is solved, and more efficient learning planning is achieved.
Patent Information
- Application Number
- CN202511086195.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-08-04
- Publication Date
- 2025-11-28
AI Technical Summary
In existing forgetting curve models, the memory intensity S is treated as a fixed value without considering individual changes in the user, resulting in a gradual decline in the effectiveness of review.
By obtaining the time interval and test accuracy between the two most recent learning events, the memory strength S is dynamically updated, and the timing of the next learning event is planned based on the preset memory retention rate R, thus achieving adaptive learning time planning.
It improves the efficiency of learning and review, making the review more in line with the user's current memory ability and memory effect expectations, and solves the problem of poor review effect caused by the forgetting curve model deviating from the memory situation.
Smart Images
Figure CN121034152A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of education, and in particular to a learning planning method, device and computer storage medium. BACKGROUND
[0002] Learning is very important, especially for students, learning can achieve knowledge reserve, ability shaping. However, students need to learn a lot of knowledge points, and it is easy to forget the previous one after learning the latter one. For this problem, people often use / repeat the knowledge points to consolidate the memory effect; in order to more efficiently combat forgetting, the forgetting curve is gradually introduced to arrange the review time more scientifically at present, so as to achieve more efficient review effect, especially in the education industry.
[0003] With the deepening of research, more and more psychologists have put forward more mathematical models for fitting the forgetting curve. These mathematical models are often associated with memory strength S. In actual application process, memory strength S is a constant that needs to be defined according to individual memory ability. It has individual differences, and the memory strength S of different users is different. Therefore, the forgetting curve model constructed based on individual memory strength S can better meet the differentiated needs of user ability.
[0004] However, the inventors of the present application found that as the review frequency increases, people become more familiar with the knowledge points, and at this time the memory ability of related knowledge points also changes. However, in the current multiple forgetting curve models, the memory strength S is a constant and often remains unchanged. If the above factors are not considered and the forgetting curve models are simply applied, even the same user, the review effect will gradually decrease due to deviation from his own memory situation. SUMMARY
[0005] Therefore, the embodiments of the present application provide a learning planning method to solve the problem that the review effect gradually decreases due to the deviation of the memory strength from the user's own memory situation when the memory strength is simply applied as a fixed value to the forgetting curve model.
[0006] In a first aspect, the embodiments of the present application provide a learning planning method, which comprises:
[0007] obtaining a time interval t between the last two learning events and a test accuracy A corresponding to the last learning event, wherein the learning event includes a first learning knowledge event and at least one review test event;
[0008] performing memory ability analysis processing based on the test accuracy A and the time interval t to obtain the current memory strength S;
[0009] Obtain a preset memory retention rate R, wherein the preset memory retention rate R is a user-defined target value for memory retention rate;
[0010] Based on the preset memory retention rate R and the current memory strength S, time information conversion processing is performed to obtain the suggested interval duration for the next learning event;
[0011] Based on the end time of the most recent learning event and the suggested interval for the next learning event, the start time of the next learning event is planned.
[0012] Optionally, the step of performing memory capacity analysis based on the test accuracy A and the time interval t to obtain the current memory strength S includes:
[0013] The memory curve model is invoked, which is a function of memory retention rate, memory strength, and time.
[0014] Using the test accuracy A and the time interval t as inputs, memory capacity is analyzed in the memory curve model to obtain the current memory strength S; wherein, the test accuracy A represents the current memory retention rate in the memory curve model.
[0015] Optionally, the step of performing time information conversion processing based on the preset memory retention rate R and the current memory strength S to obtain the suggested interval for the next learning event includes:
[0016] Based on the memory curve model, the preset memory retention rate R and the current memory strength S are used as inputs to perform time information parsing in the memory curve model to obtain the suggested interval duration for the next learning event.
[0017] Optionally, after the step of planning the start time of the next learning event based on the end time of the most recent learning event and the suggested interval of the next learning event, the method further includes:
[0018] Monitor whether a new round of learning events from the user has been triggered;
[0019] When a trigger signal is received, the start time corresponding to the user's new round of learning event is obtained;
[0020] Determine whether the start time of the new round of learning event matches the start time of the planned next learning event;
[0021] When there is a mismatch, retrieve the time difference between the start time of the new round of learning event and the end time of the most recent learning event.
[0022] Based on the time difference, the current memory strength S, the number of questions corresponding to the most recent learning event, the test accuracy A corresponding to the most recent learning event, and the preset memory retention rate R, the number of questions corresponding to the new round of learning events is adjusted.
[0023] Optionally, the step of adjusting the number of questions corresponding to the new round of learning events based on the time difference, the current memory strength S, the number of questions corresponding to the most recent learning event, the test accuracy A corresponding to the most recent learning event, and the preset memory retention rate R includes:
[0024] Based on the time difference and the current memory intensity S, the degree of memory retention is analyzed to obtain the theoretical memory retention rate corresponding to the new round of learning events, wherein the theoretical memory retention rate represents the theoretical accuracy.
[0025] Retrieve the number of questions corresponding to the most recent learning event;
[0026] Based on the theoretical memory retention rate corresponding to the new round of learning events, the test accuracy A corresponding to the most recent learning event, and the number of questions corresponding to the most recent learning event, the number of questions required to restore the test accuracy A is obtained.
[0027] The number of questions corresponding to the new round of learning events is adjusted based on the number of questions required to restore the test accuracy A and the preset memory retention rate R.
[0028] Optionally, the step of analyzing memory retention based on the time difference and the current memory intensity S to obtain the theoretical memory retention rate corresponding to the new round of learning events includes:
[0029] Call the memory curve model;
[0030] Using the time difference and the current memory strength S as inputs, the memory retention rate is analyzed in the memory curve model to obtain the theoretical memory retention rate corresponding to the new round of learning events.
[0031] Optionally, the step of obtaining the number of questions required to restore the test accuracy A based on the theoretical memory retention rate corresponding to the new round of learning events, the test accuracy A corresponding to the most recent learning event, and the number of questions corresponding to the most recent learning event includes:
[0032] Obtain the difference between the test accuracy A corresponding to the most recent learning event and the theoretical memory retention rate corresponding to the new round of learning events;
[0033] Based on the difference and the number of questions corresponding to the most recent learning event, the change in the number of questions matching the difference is obtained, wherein the change in the number of questions is used to fill the memory loss corresponding to the difference;
[0034] Based on the number of questions corresponding to the most recent learning event and the change in the number of questions matching the difference, the number of questions required to restore the test accuracy A is obtained.
[0035] Optionally, the step of adjusting the number of questions corresponding to the new round of learning events based on the number of questions required to restore the test accuracy A and the preset memory retention rate R includes:
[0036] Based on the preset memory retention rate R and the test accuracy A corresponding to the most recent learning event, the ratio of the preset memory retention rate R to the test accuracy A corresponding to the most recent learning event is obtained;
[0037] The number of questions required to restore the test accuracy A is multiplied by the ratio to obtain the number of questions corresponding to the new round of learning events, and then adjusted.
[0038] Secondly, embodiments of this application provide a learning planning device, the device comprising:
[0039] The first acquisition module is used to obtain the time interval t between the two most recent learning events and the test accuracy A corresponding to the most recent learning event. The learning events include the first knowledge learning event and at least one review test event.
[0040] The parsing module is used to perform memory ability parsing processing based on the test accuracy A and the time interval t to obtain the current memory strength S;
[0041] The second acquisition module is used to acquire a preset memory retention rate R, wherein the preset memory retention rate R is a user-set target value for memory retention rate;
[0042] The conversion module is used to perform time information conversion processing based on the preset memory retention rate R and the current memory strength S to obtain the suggested interval duration for the next learning event;
[0043] The planning module is used to plan the start time of the next learning event based on the end time of the most recent learning event and the suggested interval of the next learning event.
[0044] Thirdly, embodiments of this application provide a computer storage medium storing a computer program, which, when executed by a processor, implements the steps in the learning planning method provided in the embodiments of the invention.
[0045] The learning planning method provided in this application has the following technical effects: By introducing a test accuracy rate A, the user's current memory status is analyzed based on the test accuracy rate A, and memory ability is analyzed based on the test accuracy rate A and the corresponding time interval t for each test to obtain the memory intensity S corresponding to each review test, that is, to obtain the memory intensity S after each repeated use / memorization of knowledge points, thus realizing adaptive updating of memory intensity S; then, based on the updated memory intensity S and the user's pre-set memory retention rate R, time information is converted to obtain the time limit corresponding to the next review event to achieve / maintain the user's expected memory retention rate R under the current memory intensity S, and the start time of the next learning event is planned accordingly. In this way, this solution uses the test accuracy rate A to obtain the updated memory intensity S, and based on the updated memory intensity S and the user's expected memory retention rate R, realizes the learning time planning that conforms to the user's current memory ability and expected memory effect, solving the problem that the current method of simply applying the forgetting curve model as a fixed value deviates from the user's actual memory status and the review effect gradually decreases, making the user's review more efficient. Attached Figure Description
[0046] Figure 1 This is a flowchart of a learning planning method provided in an embodiment of the present invention;
[0047] Figure 2 This is a data diagram corresponding to a learning planning method provided in an embodiment of the present invention;
[0048] Figure 3 This is a schematic diagram of the structure of a learning planning device provided in an embodiment of the present invention;
[0049] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation
[0050] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0051] like Figure 1 As shown, Figure 1 This is a flowchart of a learning planning method provided by an embodiment of the present invention. The learning planning method includes the following steps:
[0052] S101: Obtain the time interval t between the two most recent learning events, and the test accuracy A corresponding to the most recent learning event.
[0053] There can be multiple learning events. In this embodiment of the invention, at least two learning events will occur. Specifically, after learning a knowledge point, an initial test will be conducted, which includes two learning events: the first knowledge learning event and the first review test event. To consolidate the learning results, after the first review test event, a second review test event, a third review test event, and so on will be conducted sequentially, thereby forming repeated use / memorization of the knowledge point. This process can be simply described as: first learning event, review test event 1, review test event 2, review test event 3, etc. Therefore, learning events include the first knowledge learning event and at least one review test event.
[0054] The term "two most recent learning events" has two meanings: first, it refers to two consecutive learning events; second, among completed / finished learning events, it refers to the two learning events that are closest to the current time point on the timeline. In this scheme, "two most recent learning events" is defined by these two dimensions.
[0055] Regarding the time interval between two learning events: After the first learning event, the initial review test can be conducted within half a day, in which case the time interval between the two events is 0.5 days; alternatively, the initial review test can be conducted on the second day, in which case the time interval is 1 day. The time interval between the first two events can be determined based on the user's actual situation and is not limited here. Subsequent review test events will also have a time interval with the previous review test event, which could be 3 days or 5 days.
[0056] Furthermore, as explained above, in this embodiment, all learning events except the first one (which is the initial learning event) are review and test events. Therefore, when at least two learning events occur, the subsequent learning events are all review and test events; that is, the most recent learning event corresponds to a review and test event. In this embodiment, it is necessary to obtain the test accuracy A corresponding to the most recent learning event.
[0057] S102: Based on the test accuracy A and the time interval t, perform memory ability analysis to obtain the current memory strength S.
[0058] In this implementation, considering that memory intensity S is an element defined by an individual's memory ability, and that memory intensity S can change for the same person; and that review test results can change as the user's mastery and retention of knowledge points change, this solution introduces test accuracy A to reflect the changes in the memory intensity S of relevant knowledge points for an individual user in different batches of review events.
[0059] It is worth emphasizing that this solution does not simply increase or decrease the value of memory strength S based on the correctness or incorrectness of the test results. Instead, it introduces the value of test accuracy A. The test accuracy A dynamically changes with the degree of mastery and retention of knowledge points each time. It can reflect the user's current retention rate of the target knowledge point, and the current memory strength S is obtained through this retention rate. Simple correct and incorrect results cannot break down this information.
[0060] Furthermore, this implementation method groups two adjacent learning events together and analyzes the memory intensity of the later learning event to obtain the memory intensity S at the end of the group. Specifically, after experiencing a learning event (including review and testing events), people's memory of the knowledge point is consolidated / strengthened, slowing down the rate of forgetting. However, before experiencing another learning event (including review and testing events), it still shows a certain tendency to be forgotten over time until the next learning event forms a memory stimulus. In other words, the memory situation itself changes during the time interval between two learning events, and the re-stimulation from the later learning event will again affect the memory intensity S. This implementation method analyzes the test accuracy A of the later learning event in each group, which essentially obtains the amount of memory retention of that knowledge point after the passage of time and the re-stimulation event, i.e., the current memory intensity S.
[0061] S103: Obtain the preset memory retention rate R, where the preset memory retention rate R is the user-defined target value for memory retention rate.
[0062] In this embodiment, the preset memory retention rate R is the user-set target value for memory retention, which can also be understood as an expected value. For example, the user expects the memory retention rate to reach 80% or 50%, and the subsequent plan will combine the user-set expected value to plan the learning time. Of course, in some solutions, the preset memory retention rate R can also be data set by the user through experimentation. From the perspective of product use, these data are the current target values for memory retention.
[0063] It is worth mentioning that the preset memory retention rate R can be a constant value throughout the multiple review and test event cycles, or it can be adjusted after any review and test event; there is no restriction here. If the preset memory retention rate R is adjusted, the solution in this embodiment can re-plan the learning time based on the adjusted memory retention rate R.
[0064] S104: Based on the preset memory retention rate R and the current memory strength S, perform time information conversion processing to obtain the suggested interval for the next learning event.
[0065] In this embodiment, as described in the preceding steps, the current memory strength S and the preset memory retention rate R have been obtained in this step. In order to achieve / maintain the preset memory retention rate target value with the current memory ability during subsequent review, this step will calculate the optimal interval for the next learning event based on the current user's memory strength S and the preset memory retention rate R, which is the recommended interval here.
[0066] It is worth mentioning that in this implementation, the optimal interval for the next learning event is determined based on the memory strength S corresponding to the test accuracy A of the most recent learning event. Since the next learning event and the most recent learning event are adjacent events and are closest on the current timeline, using the memory strength S corresponding to the most recent learning event to assist in determining the optimal time interval for the next learning event is more in line with the user's memory situation.
[0067] S105: Based on the end time of the most recent learning event and the suggested interval for the next learning event, plan the start time of the next learning event.
[0068] In this embodiment, the start time of the next learning event can be planned by taking the end time of the most recent learning event as the benchmark and combining it with the suggested interval obtained in the above steps.
[0069] It is understandable that using the end time of the most recent learning event as the benchmark includes scenarios that start from the end time of the most recent learning event, as well as scenarios that start from the end time of the most recent learning event plus a certain amount of flexible time. There are no restrictions here.
[0070] In this embodiment of the invention, a test accuracy rate A is introduced, and the user's current memory status is analyzed based on the test accuracy rate A. For example, the memory status can be analyzed based on the direct proportional relationship between accuracy rate A and memory strength, or the test accuracy rate A can be used to characterize the user's current memory status. Memory ability is analyzed based on the test accuracy rate A and the corresponding time interval t for each test, obtaining the memory strength S corresponding to each review test, that is, obtaining the memory strength S after each repeated use / memorization of knowledge points, thus achieving adaptive updating of memory strength S. Then, based on the updated memory strength S and the user's pre-set memory retention rate R, time information is converted to obtain the time limit corresponding to the next review event when the user's desired memory retention rate R is to be reached / maintained under the current memory strength S, and the start time of the next learning event is planned accordingly. Thus, this solution uses the test accuracy A to obtain the updated memory strength S, and based on the updated memory strength S and the user's expected memory retention rate R, it realizes a learning time plan that matches the user's current memory ability and expected memory effect. This solves the problem that simply applying the forgetting curve model deviates from the user's actual memory situation, resulting in a gradual decline in review effectiveness, and makes the user's review more efficient.
[0071] Optionally, the step of performing memory capacity analysis based on the test accuracy A and the time interval t to obtain the current memory strength S includes:
[0072] The memory curve model is invoked; the memory curve model is a function of memory retention rate, memory strength, and time.
[0073] Using the test accuracy A and the time interval t as inputs, memory capacity is analyzed in the memory curve model to obtain the current memory strength S; where the test accuracy A represents the current memory retention rate in the memory curve model.
[0074] In this embodiment, multiple memory curve models are available. This embodiment adopts the indicator function model, which is as follows:
[0075]
[0076] Where R(t) is the retention rate of memory after experiencing a learning event, i.e. the proportion that is remembered; t is time, which can be in hours or days; S is memory strength, a constant that is related to the memory material and the individual's memory ability, and is usually called memory strength or memory stability.
[0077] The test accuracy A represents the current memory retention rate in the memory curve model. By substituting the test accuracy A and the corresponding time interval t into the model, the memory strength S can be calculated.
[0078] Here, we take the memory intensity S at the i-th learning event as an example. i For example, the output of memory strength is shown as follows:
[0079] S i =-t i-1 / ln(A i-1 ); where: i is the i-th event; S i A represents the memory intensity corresponding to the i-th learning event; i-1 Let t be the test accuracy rate for the (i-1)th review test event, where the (i-1)th review test event is the i-th learning event. (It should be noted that the first learning event is the first time the knowledge is learned, and the review test event occurs only in the second learning event; therefore, the second learning event is the first review test event.) i-1 For the (i-1)th review test event, the time interval between the (i-2)th review test event and the (i-1)th review test event. (It should be noted that the time interval between two learning events only begins from the second learning event. The second learning event is the same as the first review test event. Therefore, both the time interval t and the test accuracy A are understood and explained in terms of the number of review test events.)
[0080] Regarding the above formula, here is an example: when i = 3, S3 = -t2 / ln(A2), which means that the memory strength of the third learning event is determined by the test accuracy during the second review test event and the corresponding time interval during the second review test event (i.e., the time interval between the second review test event and the first review test event).
[0081] Optionally, the step of performing time information conversion processing based on a preset memory retention rate R and the current memory strength S to obtain the suggested interval for the next learning event includes:
[0082] Based on the memory curve model, the preset memory retention rate R and the current memory strength S are used as inputs. The time information is analyzed in the memory curve model to obtain the suggested interval for the next learning event.
[0083] It is understood that the same memory curve model as in the above scheme can be used, or a different memory curve model can be used depending on the needs of the scenario. In this embodiment, it is preferred to use the same memory curve model as the aforementioned scheme, that is:
[0084]
[0085] By inputting the preset memory retention rate R and the current memory strength S obtained according to the aforementioned scheme into the memory curve model, the suggested interval for the next learning event can be output. Here, we will still use the i-th learning event as an example for explanation, and input the aforementioned S... i =-t i-1 / ln(A i-1 After inputting the preset memory retention rate R into the above model formula, the following output can be obtained:
[0086]
[0087] Where: t i For the i-th review test event, t represents the time interval between the (i-1)-th review test event and the i-th review test event; i-1 For the (i-1)th review test event, the time interval between the (i-2)th and (i-1)th review test events; R is the preset memory retention rate; A i-1 Let be the test accuracy of the (i-1)th review test event.
[0088] Optionally, after planning the start time of the next learning event based on the end time of the most recent learning event and the suggested interval for the next learning event, the steps include:
[0089] Monitor whether a new round of learning events from the user has been triggered;
[0090] When a trigger signal is received, the start time corresponding to the user's new round of learning event is obtained;
[0091] Determine whether the start time of the new round of learning events matches the start time of the planned next learning event;
[0092] When there is a mismatch, retrieve the time difference between the start time of the new round of learning events and the end time of the most recent learning event.
[0093] Based on the time difference, current memory strength S, number of questions corresponding to the most recent learning event, test accuracy A corresponding to the most recent learning event, and preset memory retention rate R, adjust the number of questions corresponding to the new round of learning events.
[0094] In this embodiment, when the user reviews according to the start time of the next learning event as planned above, the above scheme can be used cyclically based on the test accuracy A of each latest learning event and the actual interval time t.
[0095] However, when users do not review according to the planned schedule, it can be understood that they are not studying at the suggested optimal review time (or within the time interval of the optimal review time). This may weaken the review effect due to memory loss, or result in a waste of resources and time. For example, if a user delays their study by two days, their memory may have already deteriorated significantly; or if they start studying two days earlier, although they retain good memory, studying ahead of time can lead to an unnecessary increase in study frequency and a waste of time. However, for students with heavy study loads or workers with tight schedules, their study sessions often cannot start on time as planned, and are more likely to be initiated earlier or later than scheduled.
[0096] In response, the inventors of this solution proposed the following design: To compensate for the forgetting caused by the time delay, the number of questions can be adjusted to achieve memory reinforcement / forgetting countermeasures to a certain extent; if the practical time is brought forward, the number of questions can also be adjusted to reduce the burden of questions and reduce time waste.
[0097] Therefore, in this embodiment, when the trigger signal of a new round of learning event for a user is detected, the time information is first compared to determine whether the user is learning on time. If not, the time difference between the new round of learning event and the most recent learning event is further obtained, and the number of questions in the new round of learning event is adjusted in combination with the user's preset memory retention rate R.
[0098] It is worth mentioning that in this implementation, the word "match" in "determining whether the start time of the new round of learning events matches the start time of the planned next learning event" can be understood as "consistent" or "within a tolerable flexible time difference", such as 2 hours or 0.5 days, etc., and there are no restrictions here.
[0099] It should be noted that after the new round of learning events ends, the original plan can be returned to calculate the recommended review interval for the next learning event based on the test accuracy and the analyzed memory strength of the new round of learning events.
[0100] In some scenarios, where the user's learning goal is solely focused on maintaining review efficiency, and the user has ample free time, the number of questions can be adjusted only when the new round of learning events is delayed, and not when the new round of learning events is brought forward. For example, if the original number of questions is 5, and the number of questions is adjusted to 7 when the new round of learning events is delayed, and the number of questions remains 5 when the new round of learning events is brought forward, then it is not necessary to execute the "time difference between the start time of the new round of learning events and the end time of the most recent learning events" to adjust the number of questions.
[0101] To address this, this implementation method can employ the following approach: When a mismatch occurs, the step of retrieving the time difference between the start time of the new learning event and the end time of the most recent learning event includes: when a mismatch occurs, confirming whether the start time of the new learning event is later than the planned start time of the next learning event; if the start time of the new learning event is later than the planned start time of the next learning event, then retrieving the time difference between the start time of the new learning event and the end time of the most recent learning event; if the start time of the new learning event is not later than the planned start time of the next learning event, then not retrieving the time difference between the start time of the new learning event and the end time of the most recent learning event. As a preferred approach, subsequent solutions in this implementation method can be understood as scenarios where the new learning event is delayed, and no further explanation is provided for these subsequent solutions. However, it is understood that these subsequent solutions are applicable to both early and delayed new learning events, and are not limited to scenarios where the new learning event is delayed.
[0102] Optionally, the steps for adjusting the number of questions for a new round of learning events based on the time difference, current memory strength S, the number of questions corresponding to the most recent learning event, the test accuracy A corresponding to the most recent learning event, and the preset memory retention rate R include:
[0103] Based on the time difference and the current memory intensity S, the degree of memory retention is analyzed to obtain the theoretical memory retention rate corresponding to the new round of learning events. The theoretical memory retention rate represents the theoretical accuracy.
[0104] Retrieve the number of questions corresponding to the most recent learning event;
[0105] Based on the theoretical memory retention rate corresponding to the new round of learning events, the test accuracy A corresponding to the most recent learning event, and the number of questions corresponding to the most recent learning event, the number of questions required to restore the test accuracy A is obtained.
[0106] Based on the number of questions required to restore the test accuracy A corresponding to the most recent learning event and the preset memory retention rate R, adjust the number of questions corresponding to the new round of learning events.
[0107] In this embodiment, the theoretical memory retention rate for the new learning event is estimated by combining the time difference between the new learning event and the most recent learning event. Simultaneously, using the accuracy rate of the previous learning event as a primary baseline target, and combining the test accuracy A of the previous learning event with the number of questions in the previous learning event, the initial number of questions that should be achieved to reach the memory strength S required to achieve the test accuracy A of the previous learning event is obtained. Furthermore, a preset memory retention rate R is used to obtain the final target number of questions.
[0108] In a preferred embodiment, the following function model is designed:
[0109] In the i-th learning event, the number of questions is N. i For: N i =(A i-1 -R(t i )+1)*N i-1 *R / A i-1 ;
[0110] User's current memory retention R(t) i )for:
[0111]
[0112] User's current memory strength S i For: S i =-t i-1 / ln(A i-1 ).
[0113] i represents the i-th event, where i ≥ 3, meaning the new round of learning events will only occur from the 3rd learning event; N i N is the number of questions in the i-th learning event; i-1 A is the number of questions in the (i-1)th learning event; i-1 t is the test accuracy of the (i-1)th learning event; i For the i-th review test event, R(t) represents the time interval between the (i-1)-th review test event and the i-th review test event; i ) is the amount of memory retained during the i-th review test event; R is the preset memory retention rate.
[0114] The function model will be introduced in subsequent sections.
[0115] Optionally, the steps to analyze memory retention based on the time difference and the current memory strength S, and to obtain the theoretical memory retention rate for the new round of learning events, include:
[0116] Call the memory curve model;
[0117] Using the time difference and the current memory strength S as inputs, the memory retention rate is analyzed in the memory curve model to obtain the theoretical memory retention rate corresponding to the new round of learning events.
[0118] In the preferred design scheme, the design idea is to compensate for the memory loss caused by the time difference to a certain extent. In this embodiment, the theoretical memory retention rate corresponding to the new round of learning event is obtained based on the time difference between the new round of learning event and the most recent learning event and the previously calculated current memory strength S, which is the preliminary preparation for the formation of subsequent memory loss compensation scheme.
[0119] Regarding the use of the memory curve model, the same memory curve model as in the aforementioned scheme can be used, or a different memory curve model can be used depending on the needs of the scenario. In this embodiment, it is preferred to use the same memory curve model as in the aforementioned scheme, namely:
[0120]
[0121] Input the time difference and the previously calculated current memory strength S into the memory curve model. The output value obtained is the theoretical memory retention rate of the new round of learning events.
[0122] Optionally, the steps to determine the number of questions required to restore the test accuracy to A, based on the theoretical memory retention rate corresponding to the new round of learning events, the test accuracy A corresponding to the most recent learning event, and the number of questions corresponding to the most recent learning event, include:
[0123] Obtain the difference between the test accuracy A corresponding to the most recent learning event and the theoretical memory retention rate corresponding to the new round of learning events;
[0124] Based on the difference and the number of questions corresponding to the most recent learning event, the change in the number of questions matching the difference is obtained, where the change in the number of questions is used to fill in the memory loss corresponding to the difference;
[0125] Based on the number of questions corresponding to the most recent learning event and the change in the number of questions with matching differences, the number of questions required to restore the test accuracy A is obtained.
[0126] In this implementation, using the test accuracy A and the corresponding number of questions from the most recent learning event as a primary benchmark, the analysis determines how many questions need to be added after the aforementioned time difference to deepen and compensate for the loss of knowledge points, restoring memory capacity to the test accuracy A of the most recent learning event. Where: the aforementioned difference is A. i-1 -R(t i ); Changes in the number of questions above: (A) i-1 -R(ti ))*N i-1 The above "based on the change in the number of questions corresponding to the most recent learning event and the number of questions with matching differences, to obtain the number of questions needed to restore the test accuracy A" is: (A i-1 -R(t i ))*N i-1 +N i-1 After sorting, we get: (A) i-1 -R(t i )+1)*N i-1 .
[0127] Optionally, the step of adjusting the number of questions for the new round of learning events based on the number of questions required to restore the test accuracy A corresponding to the most recent learning event and the preset memory retention rate R includes:
[0128] Based on the preset memory retention rate R and the test accuracy A corresponding to the most recent learning event, the ratio of the preset memory retention rate R to the test accuracy A corresponding to the most recent learning event is obtained.
[0129] The number of questions required to restore the test accuracy A corresponding to the most recent learning event is multiplied by the ratio to obtain the number of questions corresponding to the new round of learning events, and then adjusted accordingly.
[0130] In this implementation, in addition to using the test accuracy A and the corresponding number of questions from the most recent learning event as the initial benchmark, it also incorporates the user-preset memory retention rate R as the desired target value. Using the user-preset memory retention rate R as the final target, it determines how many questions need to be added to sufficiently deepen and compensate for the memory loss of the corresponding knowledge points, thus obtaining the final target number of questions, i.e., N. i =(A i-1 -R(t i )+1)*N i-1 *R / A i-1 In this way, the learning plan can be implemented with the user's expectations in mind, which is more in line with the user's needs. Among them, when the number of questions calculated is not an integer, it can be designed to carry over to the next integer, for example, if the result of the calculated number of questions is 5.6, then it can be designed to carry over to 6 questions; or it can be designed to retain only the current integer part as the current number of questions, for example, if the result of the calculated number of questions is 5.6, then it can be extracted into 5 questions.
[0131] To better illustrate the above-described solutions of the embodiments of the present invention, the following are also provided: Figure 2 The specific data is illustrated below. The gray-filled items represent situations where users did not review at the optimal interval; the number of questions will be adjusted based on the actual interval.
[0132] like Figure 3As shown, an embodiment of the present invention also provides a learning planning device, the device comprising:
[0133] The first acquisition module 101 is used to obtain the time interval t between the two most recent learning events and the test accuracy A corresponding to the most recent learning event, wherein the learning events include the first knowledge learning event and at least one review test event;
[0134] The parsing module 102 is used to perform memory capacity parsing based on the test accuracy A and the time interval t to obtain the current memory strength S;
[0135] The second acquisition module 103 is used to acquire a preset memory retention rate R, wherein the preset memory retention rate R is a user-defined target value for memory retention rate;
[0136] The conversion module 104 is used to perform time information conversion processing based on the preset memory retention rate R and the current memory strength S to obtain the suggested interval duration for the next learning event;
[0137] The planning module 105 is used to plan the start time of the next learning event based on the end time of the most recent learning event and the suggested interval of the next learning event.
[0138] Optionally, the parsing module 102 includes:
[0139] The first calling submodule is used to call the memory curve model, which is a function relating memory retention rate, memory strength, and time.
[0140] The first processing submodule is used to analyze memory capacity in the memory curve model by taking the test accuracy A and the time interval t as inputs, and obtain the current memory strength S; where the test accuracy A represents the current memory retention rate in the memory curve model.
[0141] Optionally, the conversion module 104 includes:
[0142] The second processing submodule is used to analyze time information in the memory curve model based on the preset memory retention rate R and the current memory strength S, and obtain the suggested interval for the next learning event.
[0143] Optionally, the learning planning device also includes:
[0144] Monitoring module: Used to monitor whether a trigger signal for a new round of learning events from the user is received;
[0145] The third acquisition module is used to acquire the start time corresponding to the user's new round of learning event when a trigger signal is received.
[0146] Judgment module: Used to determine whether the start time of the new round of learning event matches the start time of the planned next learning event;
[0147] Retrieval module: Used to retrieve the time difference between the start time of the new round of learning events and the end time of the most recent learning event when there is a mismatch;
[0148] Adjustment module: Used to adjust the number of questions for a new round of learning events based on the time difference, current memory strength S, number of questions for the most recent learning event, test accuracy A for the most recent learning event, and preset memory retention rate R.
[0149] Optional, the adjustment modules include:
[0150] The third processing submodule is used to analyze the degree of memory retention based on the time difference and the current memory strength S, and obtain the theoretical memory retention rate corresponding to the new round of learning events. The theoretical memory retention rate represents the theoretical accuracy.
[0151] The first retrieval submodule is used to retrieve the number of questions corresponding to the most recent learning event.
[0152] The fourth processing submodule is used to obtain the number of questions required to restore the test accuracy to A based on the theoretical memory retention rate corresponding to the new round of learning events, the test accuracy A corresponding to the most recent learning event, and the number of questions corresponding to the most recent learning event.
[0153] The fifth processing submodule is used to adjust the number of questions for the new round of learning events based on the number of questions required to restore the test accuracy A corresponding to the most recent learning event and the preset memory retention rate R.
[0154] Optionally, the third processing submodule includes:
[0155] The second submodule for invoking the memory curve model;
[0156] The sixth processing submodule is used to analyze the degree of memory retention in the memory curve model by taking the time difference and the current memory strength S as input, and to obtain the theoretical memory retention rate corresponding to the new round of learning events.
[0157] Optionally, the fourth processing submodule includes:
[0158] First Difference Acquisition Module: Used to obtain the difference between the test accuracy A corresponding to the most recent learning event and the theoretical memory retention rate corresponding to the new round of learning events;
[0159] The seventh processing submodule is used to obtain the change in the number of questions matching the difference based on the difference and the number of questions corresponding to the most recent learning event. The change in the number of questions is used to fill in the memory loss corresponding to the difference.
[0160] The eighth processing submodule is used to obtain the number of questions needed to restore the test accuracy A based on the number of questions corresponding to the most recent learning event and the change in the number of questions with matching differences.
[0161] Optionally, the fifth processing submodule includes:
[0162] The ninth processing submodule is used to obtain the ratio of the preset memory retention rate R to the test accuracy A corresponding to the most recent learning event, based on the preset memory retention rate R and the test accuracy A corresponding to the most recent learning event.
[0163] The tenth processing submodule is used to obtain the number of questions for the new round of learning events by multiplying the number of questions required to restore the test accuracy A corresponding to the most recent learning event by the ratio, and then making adjustments.
[0164] The learning planning device provided in this embodiment of the invention can implement the various processes in the above method embodiments and achieve the same technical effect. To avoid repetition, the technical analysis and technical effects will not be described again here.
[0165] This invention also provides a computer-readable storage medium storing a computer program. When executed by a processor, the computer program implements the steps in the learning planning method provided in this invention, including:
[0166] Obtain the time interval t between the two most recent learning events, and the test accuracy A corresponding to the most recent learning event. The learning events include the first knowledge learning event and at least one review test event.
[0167] Based on the test accuracy A and the time interval t, memory capacity is analyzed and processed to obtain the current memory strength S;
[0168] Obtain the preset memory retention rate R, where the preset memory retention rate R is the user-defined target value for memory retention rate;
[0169] Based on the preset memory retention rate R and the current memory strength S, time information conversion processing is performed to obtain the suggested interval for the next learning event;
[0170] Based on the end time of the most recent learning event and the suggested interval for the next learning event, the start time of the next learning event is planned.
[0171] Optionally, the step of performing memory capacity analysis based on the test accuracy A and the time interval t to obtain the current memory strength S includes:
[0172] The memory curve model is invoked; the memory curve model is a function of memory retention rate, memory strength, and time.
[0173] Using the test accuracy A and the time interval t as inputs, memory capacity is analyzed in the memory curve model to obtain the current memory strength S; where the test accuracy A represents the current memory retention rate in the memory curve model.
[0174] Optionally, the step of performing time information conversion processing based on a preset memory retention rate R and the current memory strength S to obtain the suggested interval for the next learning event includes:
[0175] Based on the memory curve model, the preset memory retention rate R and the current memory strength S are used as inputs. The time information is analyzed in the memory curve model to obtain the suggested interval for the next learning event.
[0176] Optionally, after planning the start time of the next learning event based on the end time of the most recent learning event and the suggested interval for the next learning event, the steps include:
[0177] Monitor whether a new round of learning events from the user has been triggered;
[0178] When a trigger signal is received, the start time corresponding to the user's new round of learning event is obtained;
[0179] Determine whether the start time of the new round of learning events matches the start time of the planned next learning event;
[0180] When there is a mismatch, retrieve the time difference between the start time of the new round of learning events and the end time of the most recent learning event.
[0181] Based on the time difference, current memory strength S, number of questions corresponding to the most recent learning event, test accuracy A corresponding to the most recent learning event, and preset memory retention rate R, adjust the number of questions corresponding to the new round of learning events.
[0182] Optionally, the steps for adjusting the number of questions corresponding to a new round of learning events based on the time difference and a preset memory retention rate R include:
[0183] Based on the time difference, current memory strength S, number of questions corresponding to the most recent learning event, test accuracy A corresponding to the most recent learning event, and current memory strength S, the degree of memory retention is analyzed to obtain the theoretical memory retention rate corresponding to the new round of learning events. The theoretical memory retention rate represents the theoretical accuracy.
[0184] Retrieve the number of questions corresponding to the most recent learning event;
[0185] Based on the theoretical memory retention rate corresponding to the new round of learning events, the test accuracy A corresponding to the most recent learning event, and the number of questions corresponding to the most recent learning event, the number of questions required to restore the test accuracy A is obtained.
[0186] Based on the number of questions required to restore the test accuracy A corresponding to the most recent learning event and the preset memory retention rate R, adjust the number of questions corresponding to the new round of learning events.
[0187] Optionally, the steps to analyze memory retention based on the time difference and the current memory strength S, and to obtain the theoretical memory retention rate for the new round of learning events, include:
[0188] Call the memory curve model;
[0189] Using the time difference and the current memory strength S as inputs, the memory retention rate is analyzed in the memory curve model to obtain the theoretical memory retention rate corresponding to the new round of learning events.
[0190] Optionally, the steps to determine the number of questions required to restore the test accuracy to A, based on the theoretical memory retention rate corresponding to the new round of learning events, the test accuracy A corresponding to the most recent learning event, and the number of questions corresponding to the most recent learning event, include:
[0191] Obtain the difference between the test accuracy A corresponding to the most recent learning event and the theoretical memory retention rate corresponding to the new round of learning events;
[0192] Based on the difference and the number of questions corresponding to the most recent learning event, the change in the number of questions matching the difference is obtained, where the change in the number of questions is used to fill in the memory loss corresponding to the difference;
[0193] Based on the number of questions corresponding to the most recent learning event and the change in the number of questions with matching differences, the number of questions required to restore the test accuracy A is obtained.
[0194] Optionally, the step of adjusting the number of questions for the new round of learning events based on the number of questions required to restore the test accuracy A corresponding to the most recent learning event and the preset memory retention rate R includes:
[0195] Based on the preset memory retention rate R and the test accuracy A corresponding to the most recent learning event, the ratio of the preset memory retention rate R to the test accuracy A corresponding to the most recent learning event is obtained.
[0196] The number of questions required to restore the test accuracy A corresponding to the most recent learning event is multiplied by the ratio to obtain the number of questions corresponding to the new round of learning events, and then adjusted accordingly.
[0197] The embodiments of the present invention can implement the various processes in the above method embodiments and achieve the same technical effects. To avoid repetition, the technical analysis and technical effects will not be described again here.
[0198] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The computer-readable storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.
[0199] like Figure 4 As shown, this application embodiment also relates to an electronic device 10, including a processor 12, which can execute any of the above-described learning planning methods.
[0200] Specifically, it includes a processor 12 and a memory 11, as well as a computer program for the learning planning method stored on the memory 11 and capable of running on the processor 12, wherein the processor 12 runs the computer program for the learning planning method in the memory 11.
[0201] The above description discloses only preferred embodiments of the present invention and should not be construed as limiting the scope of the present invention. Therefore, equivalent variations made in accordance with the claims of the present invention are still within the scope of the present invention.
Claims
1. A learning planning method, characterized in that, The method includes: Obtain the time interval t between the two most recent learning events, and the test accuracy A corresponding to the most recent learning event. The learning events include the first knowledge learning event and at least one review test event. Based on the test accuracy A and the time interval t, memory ability analysis is performed to obtain the current memory strength S; Obtain a preset memory retention rate R, wherein the preset memory retention rate R is a user-defined target value for memory retention rate; Based on the preset memory retention rate R and the current memory strength S, time information conversion processing is performed to obtain the suggested interval duration for the next learning event; Based on the end time of the most recent learning event and the suggested interval for the next learning event, the start time of the next learning event is planned.
2. The learning planning method as described in claim 1, characterized in that, The step of performing memory ability analysis based on the test accuracy A and the time interval t to obtain the current memory strength S includes: The memory curve model is invoked, which is a function of memory retention rate, memory strength, and time. Using the test accuracy A and the time interval t as inputs, memory capacity is analyzed in the memory curve model to obtain the current memory strength S; wherein, the test accuracy A represents the current memory retention rate in the memory curve model.
3. The learning planning method as described in claim 2, characterized in that, The step of performing time information conversion processing based on the preset memory retention rate R and the current memory strength S to obtain the suggested interval for the next learning event includes: Based on the memory curve model, the preset memory retention rate R and the current memory strength S are used as inputs to perform time information parsing in the memory curve model to obtain the suggested interval duration for the next learning event.
4. The learning planning method as described in claim 2 or 3, characterized in that, After the step of planning the start time of the next learning event based on the end time of the most recent learning event and the suggested interval of the next learning event, the following steps are included: Monitor whether a new round of learning events from the user has been triggered; When a trigger signal is received, the start time corresponding to the user's new round of learning event is obtained; Determine whether the start time of the new round of learning event matches the start time of the planned next learning event; When there is a mismatch, retrieve the time difference between the start time of the new round of learning event and the end time of the most recent learning event. Based on the time difference, the current memory strength S, the number of questions corresponding to the most recent learning event, the test accuracy A corresponding to the most recent learning event, and the preset memory retention rate R, the number of questions corresponding to the new round of learning events is adjusted.
5. The learning planning method as described in claim 4, characterized in that, The step of adjusting the number of questions corresponding to the new round of learning events based on the time difference, the current memory strength S, the number of questions corresponding to the most recent learning event, the test accuracy A corresponding to the most recent learning event, and the preset memory retention rate R includes: Based on the time difference and the current memory intensity S, the degree of memory retention is analyzed to obtain the theoretical memory retention rate corresponding to the new round of learning events, wherein the theoretical memory retention rate represents the theoretical accuracy. Retrieve the number of questions corresponding to the most recent learning event; Based on the theoretical memory retention rate corresponding to the new round of learning events, the test accuracy A corresponding to the most recent learning event, and the number of questions corresponding to the most recent learning event, the number of questions required to restore the test accuracy A is obtained. The number of questions corresponding to the new round of learning events is adjusted based on the number of questions required to restore the test accuracy A and the preset memory retention rate R.
6. The learning planning method as described in claim 5, characterized in that, The step of analyzing memory retention based on the time difference and the current memory strength S to obtain the theoretical memory retention rate corresponding to the new round of learning events includes: Call the memory curve model; Using the time difference and the current memory strength S as inputs, the memory retention rate is analyzed in the memory curve model to obtain the theoretical memory retention rate corresponding to the new round of learning events.
7. The learning planning method as described in claim 5, characterized in that, The step of obtaining the number of questions required to restore the test accuracy A based on the theoretical memory retention rate corresponding to the new round of learning events, the test accuracy A corresponding to the most recent learning event, and the number of questions corresponding to the most recent learning event includes: Obtain the difference between the test accuracy A corresponding to the most recent learning event and the theoretical memory retention rate corresponding to the new round of learning events; Based on the difference and the number of questions corresponding to the most recent learning event, the change in the number of questions matching the difference is obtained, wherein the change in the number of questions is used to fill the memory loss corresponding to the difference; Based on the number of questions corresponding to the most recent learning event and the change in the number of questions matching the difference, the number of questions required to restore the test accuracy A is obtained.
8. The learning planning method as described in claim 7, characterized in that, The step of adjusting the number of questions corresponding to the new round of learning events based on the number of questions required to restore the test accuracy A and the preset memory retention rate R includes: Based on the preset memory retention rate R and the test accuracy A corresponding to the most recent learning event, the ratio of the preset memory retention rate R to the test accuracy A corresponding to the most recent learning event is obtained; The number of questions required to restore the test accuracy A is multiplied by the ratio to obtain the number of questions corresponding to the new round of learning events, and then adjusted.
9. A learning planning device, characterized in that, The device includes: The first acquisition module is used to obtain the time interval t between the two most recent learning events and the test accuracy A corresponding to the most recent learning event. The learning events include the first knowledge learning event and at least one review test event. The parsing module is used to perform memory ability parsing processing based on the test accuracy A and the time interval t to obtain the current memory strength S; The second acquisition module is used to acquire a preset memory retention rate R, wherein the preset memory retention rate R is a user-set target value for memory retention rate; The conversion module is used to perform time information conversion processing based on the preset memory retention rate R and the current memory strength S to obtain the suggested interval duration for the next learning event; The planning module is used to plan the start time of the next learning event based on the end time of the most recent learning event and the suggested interval of the next learning event.
10. A computer storage medium, characterized in that, The computer storage medium stores a computer program, which, when executed by a processor, implements the steps of the learning planning method as described in any one of claims 1 to 8.