Database task configuration method

Abstract

The invention relates to the technical field of data processing, in particular to a database task configuration method, which comprises the following steps of: acquiring execution time, completion degree and execution frequency of a plurality of users for completing each learning task, and generating execution data of each learning task; constructing a plurality of learning classification models, and selecting an optimal learning classification model; dividing users into a single type and a mixed type; configuring a plurality of learning tasks for single learning of the user, generating a task forgetting curve, and executing a task completion model to generate a task execution log; adjusting the task forgetting curve to adjust the time ratio of the newly added learning task to the previous learning task; learning tasks are divided into special learning tasks and general learning tasks, the general learning tasks are divided into hot tasks and cold tasks, and a task completion model is generated in combination with the previous learning tasks and the hot tasks; according to the method and the system, the accuracy of database task configuration is improved while learning tasks are effectively distributed to different students in a targeted manner.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a database task configuration method. Background Art

[0002] Every student is different in learning ability, learning style, knowledge base, etc. Some students have strong comprehension and learn quickly; some students need more time and practice to master knowledge; some students are good at visual learning, some prefer auditory learning, and some students are better at taking notes, writing and reading. By pushing learning tasks in a targeted manner, we can meet the personalized needs of different students and improve learning effects. Traditional education often adopts unified teaching content and progress, which makes it difficult to take into account the specific situation of each student. Targeted learning task push can avoid the waste of educational resources.

[0003] Chinese patent application publication number: CN112651678A discloses a teaching management system and method based on big data. The invention discloses a teaching management system and method based on big data. The system includes: a teaching task management module, a cloud module, a grade management module, and a teaching task planning module. The teaching task management module is used to send the teaching task to the cloud module and send the student information to the grade management module; the cloud module is used to combine the teaching task with the historical data in the teaching task database, fragment the teaching task, and obtain the time and grade corresponding to the fragmented task; the grade management module is used to obtain the student grade according to the student information, and determine the student type according to the student grade, and the student type corresponding to the student information; the learning task planning module is used to recommend the corresponding time and grade of the fragmented task according to the student type, and generate the corresponding teaching plan according to the fragmented task. The system can accurately plan the student's learning tasks, save the student's learning time, and improve the student's learning efficiency.

[0004] Chinese Patent Application Publication No.: CN118333582A discloses a teaching equipment management control system and method. The invention provides a teaching equipment management control system and method. The system includes: an equipment control unit for receiving, parsing, and outputting teaching instructions, collecting learning data, processing input feedback of the equipment, and transmitting it to the central control unit through a communication control unit; a central control unit for receiving data from the equipment control unit and generating control instructions and state analysis of the equipment; a detection and alarm unit for receiving equipment parameters from the central control unit, real-time monitoring the operating state of teaching equipment, and performing fault analysis; a communication control unit for data transmission and communication between units; and a power control unit for supplying power to each unit. Through the integrated multi-module design and the collaborative work of each unit, the invention improves the adaptability and accuracy of teaching strategies, and significantly improves teaching efficiency, learning experience, equipment operation and maintenance level, and energy utilization efficiency.

[0005] It can be seen that in the prior art, simply assigning tasks to students based on their academic performance does not take into account the learning process and learning habits of students, and fails to reasonably allocate learning tasks by adjusting the learning forgetting curve of students. Summary of the Invention

[0006] Therefore, the present invention provides a database task configuration method to overcome the problems in the prior art that simply assigning tasks to students based on their academic performance does not take into account the learning process and learning habits of students, and fails to reasonably allocate learning tasks by adjusting the learning forgetting curve of students.

[0007] To achieve the above object, the present invention provides a database task configuration method, including: Collect the execution time, completion degree, and execution times of several users for each learning task, generate execution data for each learning task, and the execution data includes the average execution time, average completion degree, and average execution times of each learning task; Call several learning tasks in combination with the average completion degree and the average execution times to construct several learning classification models, and select the optimal learning classification model; Classify the users into single-type and mixed-type according to the execution situation of the users for the optimal learning classification model; Based on the single learning time of the users and the average execution time, configure several learning tasks for each single learning of the users, count the single completion degree of each learning task to generate a task forgetting curve, and combine the different classifications of the users and the average completion degree to match and execute a task completion model for the users to generate a task execution log; Adjust the task forgetting curve using a time series analysis algorithm based on the task execution log to adjust the time ratio between newly added learning tasks and past learning tasks; Divide the learning tasks into special learning tasks and general learning tasks according to the completion status of the learning tasks, divide the general learning tasks into hot tasks and cold tasks, and generate the task completion model by combining the past learning tasks and the hot tasks.

[0008] Furthermore, the process of calling several of the learning tasks to construct several learning classification models in combination with the average completion rate and the average number of executions, and selecting the optimal learning classification model includes: Call several of the learning tasks to construct several learning classification models in combination with the average completion rate and the average number of executions; Calculate the estimated execution time of each of the learning classification models in combination with the average execution time; Select the optimal learning classification model based on the user's single learning time and the estimated execution time.

[0009] Furthermore, the process of classifying the user into the single type and the mixed type according to the execution situation of the user for the optimal learning classification model includes: Statistically analyze the memory sizes occupied by text, audio, and video in each learning task in the optimal learning classification model; Allocate each of the learning tasks to different learning modules according to the task classification corresponding to the maximum value of the memory size; Calculate a dimension score based on the user's usage of each learning module in the optimal learning classification model, and classify the user into the single type and the mixed type according to the dimension score; The task classification includes visual type, auditory type, and reading and writing type. The single type is a user with only one of the task classifications, and the mixed type is a user with two or more of the task classifications. The learning modules include a text module, an audio module, and a video module.

[0010] Furthermore, the process of calculating a dimension score based on the user's usage of each learning module in the optimal learning classification model and classifying the user into the single type and the mixed type according to the dimension score includes: Statistically analyze the number of selection times and usage duration of the user using different learning modules in the optimal learning classification model; Calculate the dimension score by combining the number of selection times and the usage duration, and sort the dimension scores; Determine that the user is of the single type based on the difference between the highest dimension score and the second highest dimension score, or compare each of the dimension scores with a preset dimension score.

[0011] Further, the process of configuring several of the learning tasks for the user's single learning session based on the user's single learning time and the average execution time includes: Sum up the average execution times corresponding to each of the learning tasks to obtain the total configuration time; Calculate the configuration time difference between the total configuration time and the single learning time; Compare the configuration time difference with a preset configuration time difference, and configure several of the learning tasks for the user's single learning session according to the comparison result; The preset configuration time difference is positively correlated with the user's single learning time.

[0012] Further, count the several single completion degrees of each of the learning tasks in each of the single learning sessions, and generate the task forgetting curve by combining the time intervals of each of the single learning sessions and the corresponding single completion degrees.

[0013] Further, the process of matching and executing a task completion model for the user to generate a task execution log by combining the different classifications of the user and the average completion degree includes: Collect the classifications of each user and the average completion degree of completing each of the learning tasks in the learning classification model; Match the task completion model that is consistent with the classification of the user and the average completion degree in the learning task database; Record the situation of the user completing the task completion model to generate the task execution log; The task completion model includes the classifications of each user and the corresponding learning tasks completed.

[0014] Further, use the time series analysis algorithm based on the task execution log to predict the completion degree of the past learning tasks to adjust the extreme points of the task forgetting curve. The time series analysis algorithm predicts the extreme points of the task forgetting curve corresponding to the past learning tasks by analyzing the completion degree and execution time of the past learning tasks. The extreme points are the points on the task forgetting curve with the largest curvature.

[0015] Further, according to the adjusted task forgetting curve, reduce the time ratio of the new learning tasks and the past learning tasks at the extreme points. The single learning time includes the time for executing the new learning tasks and the time for executing the past learning tasks. The time ratio is the ratio of the time occupied by the user in executing the new learning tasks and the time occupied by the user in executing the past learning tasks.

[0016] Further, according to the completion situation of the learning tasks, the learning tasks are divided into special learning tasks and general learning tasks, and the general learning tasks are divided into hot tasks and cold tasks. The process of generating the task completion model by combining the past learning tasks and the hot tasks includes: The learning tasks not completed by the user in the single learning are divided into special learning tasks, and the learning task database pushes relevant materials of the special learning tasks to the user according to the classification of the user; The learning tasks completed by the user in the single learning are divided into general learning tasks, and the hot and cold tasks are determined based on the completion duration; A single-type allocation method is adopted for the hot tasks, and a mixed-type allocation method is adopted for the cold tasks; Select the learning tasks that are the same as the past learning tasks and the hot tasks to generate the task completion model.

[0017] Compared with the prior art, the beneficial effects of the present invention are as follows. The present invention classifies the user into visual type, auditory type, and reading and writing type by using the learning classification model. Different types of learners receive and process information in different ways. Visual learners are sensitive to visual information such as images, charts, and videos; Auditory learners are better at learning by listening to explanations and audio; Reading and writing learners prefer to read written materials and take written records. After classifying the users, learning tasks, resources, and methods that match their learning styles can be provided for them, thereby improving the learning efficiency and effect. According to the learning types of students, learning resources are reasonably allocated. After students understand their learning types, they can more clearly understand their learning characteristics and advantages, which helps them give full play to their strengths and improve their deficiencies in the learning process. When the learning tasks and resources match the learning types of students, students are more likely to be engaged in learning and feel the joy of learning, which can effectively reduce the boredom and frustration in the learning process, improve the participation and perseverance of students, and improve the accuracy of database task configuration.

[0018] Furthermore, in the present invention, by combining the different classifications and average completion rates of users, a task completion model completed by past users with the same classification and average completion rate as the user is matched in the learning task database. The learning experience of past users is used to guide current users. Each user has unique learning characteristics and levels. By classifying users and calculating the average completion rate, past users similar to the current user can be found more accurately. The task completion models of these past users are formed based on their actual learning experiences and can provide personalized learning paths and method suggestions for current users, meeting the specific needs of current users, improving the pertinence and effectiveness of learning. Past users have accumulated successful experiences and methods during the task completion process, and current users can directly draw on these experiences to avoid detours. By referring to the task completion models of past users, a learning method suitable for themselves can be quickly found, which can save a lot of time and energy, enable current users to make more efficient use of learning resources, focus on knowledge learning and ability improvement, and reduce time waste. By analyzing the task completion models of past users and the feedback of current users, the needs and adaptation situations of different types of users for learning tasks can be understood, which helps to optimize the design of learning tasks. Different users may adopt different methods and strategies when facing the same task. By drawing on the experiences of multiple past users, current users can broaden their learning ideas, learn to view and solve problems from different perspectives, and further improve the accuracy of database task configuration.

[0019] Furthermore, in the present invention, the task forgetting curve of the user is adjusted by collecting parameters during the user's learning process, and the learning tasks are dynamically adjusted for the user. The learning rhythm, forgetting speed, and knowledge mastery level of each user are different. By collecting parameters during the learning process, such as learning time, answer accuracy rate, review times, weak points in knowledge mastery, etc., the learning characteristics and status of the user can be deeply understood. Based on these parameters, the task forgetting curve is adjusted, and the forgetting law of the user's knowledge can be grasped more accurately, so as to dynamically adjust the learning tasks to accurately match the user's current learning needs, improve the pertinence and effectiveness of learning. Arranging learning tasks according to the adjusted task forgetting curve can timely review and consolidate knowledge when the user is about to forget it, avoiding a large amount of knowledge forgetting. A reasonable task arrangement can reduce the time for the user to repeatedly learn the mastered knowledge. Scientifically predicting the task forgetting curve and arranging targeted review and learning tasks according to the user's forgetting law helps to transform short-term memory into long-term memory. By reviewing and applying knowledge points at appropriate time points multiple times, the user can master knowledge more firmly, reduce forgetting, and improve the knowledge retention rate, laying a solid foundation for subsequent learning and application. Dynamically adjusting learning tasks according to the user's learning situation can avoid waste of learning resources. For the content that the user has mastered proficiently, the investment of relevant resources is reduced, and for the parts that the user needs to strengthen, more high-quality resources are provided. This can improve the utilization efficiency of learning resources, make the limited resources more reasonably configured, and further improve the accuracy of database task configuration.

[0020] Furthermore, in the present invention, the learning tasks are marked according to the completion status of the learning tasks, and different types of distribution methods of relevant materials are adopted for learning tasks with different popularity levels. Through diverse task distribution methods, users can effectively handle different learning tasks. Learning tasks with different popularity levels may have different learning focuses and difficulties. Through marking and targeted material distribution, users can more clearly understand the characteristics of the tasks and obtain learning materials suitable for the tasks. The diverse task distribution methods can adapt to the learning styles and preferences of different users. Some users like to learn by reading text materials, while others prefer to watch videos or participate in practical operations. For tasks with different popularity levels, different types of materials are distributed. After marking the task popularity according to the task completion status, the learning task order can be arranged according to the popularity level and the knowledge logic relationship, and the corresponding materials can be distributed. For popular basic tasks, basic knowledge materials are provided first to guide users to lay a good foundation; for subsequent popular advanced tasks, more in-depth and comprehensive materials are provided to help users gradually build a complete knowledge system and achieve systematic learning. Appropriate material distribution and diverse task distribution methods can make users feel the orderliness and convenience of learning during the learning process. Users do not need to spend a lot of time searching for and screening materials, can be more focused on the learning content itself, reduce the troubles and obstacles during the learning process, improve the overall learning experience, make the learning process smoother and more efficient, and further improve the accuracy of database task configuration. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 is a flowchart of the method for configuring database tasks of the present invention; Figure 2 is a flowchart of selecting the optimal learning classification model in an embodiment of the present invention; Figure 3 is a logic diagram for determining user types based on the difference between the highest-dimensional score and the second-highest-dimensional score in an embodiment of the present invention; Figure 4 is a logic diagram for determining user types based on the scores of each dimension in an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0022] In order to make the objectives and advantages of the present invention clearer, the present invention will be further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0023] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are only used to explain the technical principles of the present invention and do not limit the protection scope of the present invention.

[0024] It should be noted that in the description of the present invention, the terms indicating the direction or positional relationship such as "upper", "lower", "left", "right", "inner", "outer", etc. are based on the direction or positional relationship shown in the drawings. This is only for convenience of description and does not indicate or imply that the device or element must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation to the present invention.

[0025] In addition, it should be noted that in the description of the present invention, unless otherwise clearly specified and limited, the terms "installation", "connection", and "connection" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium, and it can be the communication inside two elements. For those skilled in the art, the specific meanings of the above terms in the present invention can be understood according to specific situations.

[0026] Please refer to Figure 1 as shown in the figure, which is a flowchart of the database task configuration method of the present invention. The embodiments of the present invention provide a database task configuration method, including: Step S1, collect the execution time, completion degree, and execution times of several users for each learning task, generate the execution data of each learning task, and the execution data includes the average execution time, average completion degree, and average execution times of each learning task; Step S2, call several learning tasks to construct several learning classification models in combination with the average completion degree and average execution times, and select the optimal learning classification model; Step S3, classify users into single-type and mixed-type according to the execution situation of users for the optimal learning classification model; Step S4, configure several learning tasks for each user's single learning based on the user's single learning time and average execution time, count the single completion degree of each learning task to generate a task forgetting curve, and combine the different classifications and average completion degrees of users to match and execute a task completion model for the user to generate a task execution log; Step S5, use a time series analysis algorithm based on the task execution log to adjust the task forgetting curve to adjust the time ratio of newly added learning tasks and past learning tasks; Step S6, divide the learning tasks into special learning tasks and general learning tasks according to the learning task completion situation, divide the general learning tasks into hot tasks and cold tasks, and generate a task completion model in combination with past learning tasks and hot tasks.

[0027] Please refer to Figure 2As shown, it is a flowchart for selecting the optimal learning classification model in an embodiment of the present invention. In step S2, several learning tasks are called in combination with the average completion degree and the average number of executions to construct several learning classification models, and the process of selecting the optimal learning classification model includes: Step S201, call several learning tasks in combination with the average completion degree and the average number of executions to construct several learning classification models; Step S202, calculate the estimated execution time of each learning classification model in combination with the average execution time; Step S203, select the optimal learning classification model based on the user's single learning time and the estimated execution time.

[0028] In implementation, several learning tasks with an average completion degree greater than or equal to the preset completion degree and an average number of executions greater than or equal to the preset number of executions are called to construct several learning classification models; the average execution time of each learning task in the learning classification model is added up to calculate the estimated execution time of the learning classification model; the learning classification model with the same single learning time and the estimated execution time of the user is selected as the optimal learning classification model; It can be understood that if several optimal learning classification models are selected, the learning classification model with the fewest learning tasks is selected as the optimal learning classification model.

[0029] The preset completion degree and the preset number of executions are respectively positively correlated with the user's single learning time.

[0030] It can be understood that if the user's single learning time is longer, the more learning tasks are included, and the higher requirements for the integrity and the number of uses of the learning tasks are, so the preset completion degree and the preset number of executions are respectively positively correlated with the user's single learning time.

[0031] Optionally, the user's single learning time is 1 hour, the preset completion degree is 90%, and the preset number of executions is 10 times; The user's single learning time is 1.5 hours, the preset completion degree is 92%, and the preset number of executions is 15 times; The user's single learning time is 2 hours, the preset completion degree is 94%, and the preset number of executions is 20 times; Specifically, in step S3, the process of classifying the user into a single type and a mixed type according to the execution situation of the user for the optimal learning classification model includes: Statistically analyze the memory sizes occupied by text, audio, and video respectively in each learning task in the optimal learning classification model; Allocate each learning task to different learning modules according to the task classification corresponding to the maximum value of the memory size; Calculate the dimension score based on the user's use of each learning module in the optimal learning classification model, and classify the user into a single type and a mixed type according to the dimension score; Task classifications include visual, auditory, and reading / writing types. A single-type user has only one task classification, and a mixed-type user has two or more task classifications. The learning module includes a text module, an audio module, and a video module.

[0032] It can be understood that the mixed types are divided into visual-auditory mixed type, auditory-reading / writing mixed type, visual-reading / writing mixed type, and visual-auditory-reading / writing mixed type.

[0033] It can be understood that the audio information contained in the video is not counted in the audio memory occupancy, and the image information is counted in the video memory occupancy.

[0034] It can be understood that the dimension score of the text module corresponds to the usage frequency of the reading / writing type, the dimension score of the audio module corresponds to the usage frequency of the auditory type, and the dimension score of the video module corresponds to the usage frequency of the visual type.

[0035] Specifically, the present invention classifies the user into visual, auditory, and reading / writing types by using the learning classification model. Different types of learners receive and process information differently. Visual learners are sensitive to visual information such as images, charts, and videos; auditory learners are better at learning by listening to explanations and audio; reading / writing learners prefer to read written materials and take written records. After classifying the users, learning tasks, resources, and methods that match their learning styles can be provided for them, thereby improving learning efficiency and effect. According to the learning types of students, learning resources can be reasonably allocated. After students understand their learning types, they can more clearly recognize their learning characteristics and advantages, which helps them give full play to their strengths and improve their deficiencies in the learning process. When learning tasks and resources match the learning types of students, students are more likely to be engaged in learning and feel the joy of learning, which can effectively reduce the boredom and frustration in the learning process, improve students' participation and perseverance, and improve the accuracy of database task configuration.

[0036] Please refer to Figure 3 As shown, it is a logic diagram for determining the user type based on the difference between the highest dimension score and the second-highest dimension score in an embodiment of the present invention. In step S3, the dimension score is calculated based on the usage of each learning module by the user in the optimal learning classification model. The process of classifying the user into single-type and mixed-type according to the dimension score includes: Count the selection times and usage duration of the user using different learning modules in the optimal learning classification model; Calculate the dimension score by combining the selection times and usage duration, and sort the dimension scores; Determine that the user is single-type based on the difference between the highest dimension score and the second-highest dimension score, or compare each dimension score with a preset dimension score, where If the difference between the highest dimension score and the second-highest dimension score is greater than or equal to the preset difference, the user is determined to be of the single type. If the difference between the highest dimension score and the second-highest dimension score is less than the preset difference, then calculate the scores of each dimension and compare the scores of each dimension with the preset dimension score. In implementation, dimension score = number of selections + usage duration × 0.01. Only the numerical values of each item are selected for calculation in the above formula.

[0037] In a specific embodiment, it is set that the number of selections of the text module used by the user in a single learning session is 2 times, the usage duration is 240 seconds, the number of selections of the audio module is 4 times, the usage duration is 560 seconds, and the number of selections of the video module is 1 time, and the usage duration is 140 seconds. Then the text dimension score is 4.4, the audio dimension score is 9.6, and the video dimension score is 2.4. The preset difference is set to 2. Sort the scores of each dimension from largest to smallest to get the highest dimension score of 9.6, the second-highest dimension score of 4.4, and the lowest dimension score of 2.4. Calculate the difference between the highest dimension score and the second-highest dimension score to be 5.2, which is greater than the preset difference of 2. Then the user is determined to be of the single type. If the difference between the highest dimension score and the second-highest dimension score is 1.6, which is less than the preset difference of 2, then determine to compare the scores of each dimension with the preset dimension score. The preset difference is positively correlated with the total learning duration of the user.

[0038] It can be understood that as the total learning duration of the user increases, the selected modules better highlight their own characteristics. In the state of preferentially selecting their favorite learning modules, the time and number of selections increase. Therefore, the preset difference is positively correlated with the total learning duration of the user.

[0039] Optionally, the total learning duration of the user is 1 hour, and the preset difference is 2. The total learning duration of the user is 1.5 hours, and the preset difference is 3. The total learning duration of the user is 2 hours, and the preset difference is 4.

[0040] Please refer to Figure 4 As shown, it is the logic diagram for determining the user type based on the scores of each dimension in the embodiment of the present invention. Compare the scores of each dimension with the preset dimension score. Among them, If the scores of all dimensions are less than the preset dimension score, the user is determined to be of the mixed type. If there is a single dimension score greater than or equal to the preset dimension score, the user is determined to be of the single type: If there are two dimension scores greater than or equal to the preset dimension score, the user is determined to be of the mixed type. If the scores of all dimensions are greater than or equal to the preset dimension score, the user is determined to be a mixed type: In a specific embodiment, the preset dimension score is set to 5. If the text dimension score is 4.4, the audio dimension score is 2.3, and the video dimension score is 2.4, all of which are less than the preset dimension score, the user is determined to be a mixed type; If the text dimension score is 6.4, the audio dimension score is 2.3, the video dimension score is 2.4, and the text dimension score of 6.4 is greater than the preset dimension score, the user is determined to be a single type: If the text dimension score is 6.4, the audio dimension score is 7.3, the video dimension score is 2.4, and the text dimension score and the audio dimension score are greater than the preset dimension score, the user is determined to be a mixed type; If the text dimension score is 6.4, the audio dimension score is 7.3, and the video dimension score is 6.4, all of which are greater than the preset dimension score, the user is determined to be a mixed type: The preset dimension score is positively correlated with the total learning duration of the user.

[0041] It can be understood that as the total learning duration of the user increases, the number of times and the usage duration of the modules that meet the user's preferences increase, and the corresponding dimension scores increase. Therefore, the preset dimension score is positively correlated with the total learning duration of the user.

[0042] Optionally, the total learning duration of the user is 1 hour, and the preset dimension score is 5; The total learning duration of the user is 1.5 hours, and the preset dimension score is 8; The total learning duration of the user is 2 hours, and the preset dimension score is 11.

[0043] Specifically, in step S4, the process of configuring several learning tasks for the user's single learning based on the user's single learning time and average execution time includes: Adding up the average execution times corresponding to each learning task to obtain the configured total time; Calculating the configured time difference between the configured total time and the single learning time; Comparing the configured time difference with the preset configured time difference, and configuring several learning tasks for the user's single learning according to the comparison result; In a specific embodiment, the preset configured time difference is set to 1 minute. If the configured time difference is 0.6 minutes, which is less than the preset time difference, it is determined that the configuration is successful, If the time difference is 1.3 minutes, which is greater than the preset time difference, it is determined to reconfigure.

[0044] The preset configured time difference is positively correlated with the user's single learning time.

[0045] It can be understood that the longer the user's single learning time is and the more learning tasks are executed, the greater the difference between the combination of each learning task and the single learning time will be. Therefore, the preset configuration time difference is positively correlated with the user's single learning time.

[0046] Optionally, the user's single learning time is 1 hour and the preset configuration time difference is 1 minute; The user's single learning time is 1.5 hours and the preset configuration time difference is 3 minutes; The user's single learning time is 2 hours and the preset configuration time difference is 5 minutes.

[0047] Specifically, in step S4, the process of combining the time intervals of each single learning and the corresponding single completion degrees to generate a task forgetting curve includes:

[0048] It can be understood that a task forgetting curve is generated with the time intervals of each single learning as the horizontal axis and the corresponding single completion degrees as the vertical axis.

[0049] Specifically, in step S4, the process of matching and executing a task completion model for the user to generate a task execution log in combination with the user's different classifications and average completion degrees includes: Collect the classifications of each user and the average completion degrees of each learning task in the learning classification model; Match a task completion model in the learning task database that is consistent with the user's classification and average completion degree; Record the situation of the user completing the task completion model to generate a task execution log; The task completion model includes the classifications of each user and the corresponding learning tasks completed.

[0050] Specifically, in the present invention, by combining the different classifications and average completion degrees of users, a task completion model completed by past users with the same classification and average completion degree as the user is matched in the learning task database. The learning experience of past users is used to guide current users. Each user has unique learning characteristics and levels. By classifying users and calculating the average completion degree, past users similar to the current user can be found more accurately. The task completion models of these past users are formed based on their actual learning experiences and can provide personalized learning paths and method suggestions for current users, meeting the specific needs of current users, improving the pertinence and effectiveness of learning. Past users have accumulated successful experiences and methods during the task completion process, and current users can directly draw on these experiences to avoid detours. By referring to the task completion models of past users, a suitable learning method can be quickly found, which can save a lot of time and energy, enable current users to make more efficient use of learning resources, focus on knowledge learning and ability improvement, and reduce time waste. By analyzing the task completion models of past users and the feedback of current users, the needs and adaptation situations of different types of users for learning tasks can be understood, which helps to optimize the design of learning tasks. Different users may adopt different methods and strategies when facing the same task. By drawing on the experiences of multiple past users, current users can broaden their learning ideas, learn to view and solve problems from different perspectives, and further improve the accuracy of database task configuration.

[0051] Specifically, in step S5, a time series analysis algorithm is used based on the task execution log to predict the completion degree of past learning tasks to adjust the extreme point of the task forgetting curve. The time series analysis algorithm predicts the extreme point of the task forgetting curve corresponding to the past learning task by analyzing the completion degree and execution time of the past learning task. The extreme point is the point with the largest curvature on the task forgetting curve.

[0052] Specifically, in step S5, according to the adjusted task forgetting curve, the time ratio of new learning tasks and past learning tasks is reduced at the extreme point. The single learning time includes the time for executing new learning tasks and the time for executing past learning tasks. The time ratio is the ratio of the time occupied by the user for executing new learning tasks to the time occupied by the user for executing past learning tasks.

[0053] It can be understood that at the time point corresponding to the extreme point, by reducing new learning tasks and increasing past learning tasks in a single learning, the time ratio of new learning tasks and past learning tasks is reduced.

[0054] Specifically, in the present invention, the task forgetting curve of the user is adjusted by collecting parameters during the user's learning process, and the learning tasks are dynamically adjusted for the user. The learning rhythm, forgetting speed, and knowledge mastery level of each user are different. By collecting parameters during the learning process, such as learning time, answer accuracy rate, review times, weak points in knowledge mastery, etc., the learning characteristics and status of the user can be deeply understood. Based on these parameters, the task forgetting curve is adjusted, and the forgetting law of the user's knowledge can be grasped more accurately, so as to dynamically adjust the learning tasks to accurately match the user's current learning needs, improve the pertinence and effectiveness of learning. Arranging learning tasks according to the adjusted task forgetting curve can review and consolidate in time when the user is about to forget knowledge points, avoiding a large amount of knowledge forgetting. A reasonable task arrangement can reduce the time for the user to repeatedly learn the mastered knowledge. Scientifically predicting the task forgetting curve and making targeted review and learning task arrangements according to the user's forgetting law helps to transform short-term memory into long-term memory. By reviewing and applying knowledge points at appropriate time points multiple times, the user can master knowledge more firmly, reduce forgetting, and improve the knowledge retention rate, laying a solid foundation for subsequent learning and application. Dynamically adjusting learning tasks according to the user's learning situation can avoid waste of learning resources. For the content that the user has mastered proficiently, the input of relevant resources is reduced, and for the part that the user needs to strengthen, more high-quality resources are provided. In this way, the utilization efficiency of learning resources can be improved, and limited resources can be more reasonably configured, further improving the accuracy of database task configuration.

[0055] Specifically, in step S6, the learning tasks are divided into special learning tasks and general learning tasks according to the learning task completion situation. The general learning tasks are divided into hot tasks and cold tasks. The process of generating a task completion model by combining past learning tasks and hot tasks includes: The learning tasks not completed by the user in a single learning are divided into special learning tasks, and the learning task database pushes relevant materials of the special learning tasks to the user according to the classification. The learning tasks completed by the user in a single learning are divided into general learning tasks, and the hot and cold tasks of the general learning tasks are determined based on the completion duration. A single-type allocation method is adopted for hot tasks, and a mixed-type allocation method is adopted for cold tasks. Select learning tasks with the same past learning tasks and hot tasks to generate a task completion model.

[0056] It can be understood that when the user's classification is single-type, relevant materials of learning tasks are pushed according to the corresponding classification. When the user's classification is mixed-type, relevant materials of learning tasks are pushed according to the corresponding mixed classification.

[0057] It can be understood that when the user is a hybrid type, a single-type allocation method is adopted for hot tasks, that is, the single type with the highest dimension score is selected to push learning task-related materials. When the user is a single type, a hybrid allocation method is adopted for cold tasks, that is, vision, audition, and reading and writing hybrid are used to push learning task-related materials.

[0058] Specifically, in the present invention, the learning tasks are marked according to the completion situation of the learning tasks. Different types of allocation methods of relevant materials are adopted for learning tasks with different heat levels. Through diverse task allocation methods, users can effectively handle different learning tasks. Learning tasks with different heat levels may have different learning focuses and difficulties. Through marking and targeted material allocation, users can more clearly understand the task characteristics and obtain learning materials suitable for the task. Diverse task allocation methods can adapt to the learning styles and preferences of different users. Some users like to learn through reading text materials, while others are more inclined to watch videos or participate in practical operations. For tasks with different heat levels, different types of material allocation are adopted. After marking the task heat according to the task completion situation, the learning task order can be arranged according to the heat level and the knowledge logic relationship, and the corresponding materials can be allocated. For popular basic tasks, basic knowledge materials are provided first to guide users to lay a good foundation. For subsequent popular advanced tasks, more in-depth and comprehensive materials are provided to help users gradually build a complete knowledge system and achieve systematic learning. Appropriate material allocation and diverse task allocation methods can make users feel the orderliness and convenience of learning during the learning process. Users do not need to spend a lot of time looking for and screening materials, can focus more on the learning content itself, reduce the troubles and obstacles during the learning process, improve the overall learning experience, make the learning process smoother and more efficient, and further improve the accuracy of database task configuration.

[0059] So far, the technical solution of the present invention has been described in combination with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

Claims

1. A method for configuring database tasks, characterized in that, Including: Collect the execution time, completion degree, and execution times of several users for completing each learning task, generate execution data for each of the learning tasks, where the execution data includes the average execution time, average completion degree, and average execution times of each of the learning tasks; Call several learning tasks in combination with the average completion degree and the average execution times to construct several learning classification models, and select the optimal learning classification model; Classify the users into single-type and mixed-type according to the execution situation of the users for the optimal learning classification model; Configure several of the learning tasks for a single user learning session based on the user's single learning time and the average execution time, count the single completion degree of each of the learning tasks to generate a task forgetting curve, and in combination with the different classifications of the users and the average completion degree, match and execute a task completion model for the users to generate a task execution log; Based on the task execution log, use a time series analysis algorithm to adjust the task forgetting curve to adjust the time ratio of newly added learning tasks and past learning tasks; Divide the learning tasks into special learning tasks and general learning tasks according to the completion situation of the learning tasks, divide the general learning tasks into hot tasks and cold tasks, and generate the task completion model in combination with the past learning tasks and the hot tasks; 2. The database task configuration method according to claim 1, wherein The process of calling several of the learning tasks in combination with the average completion degree and the average execution times to construct several learning classification models, and selecting the optimal learning classification model includes: Call several of the learning tasks in combination with the average completion degree and the average execution times to construct several learning classification models; Calculate the expected execution time of each of the learning classification models in combination with the average execution time; Select the optimal learning classification model based on the user's single learning time and the expected execution time; 3. The database task configuration method according to claim 2, wherein The process of classifying the users into the single-type and the mixed-type according to the execution situation of the users for the optimal learning classification model includes: Statistically calculate the memory sizes occupied by text, audio, and video respectively in each of the learning tasks in the optimal learning classification model; Allocate each of the learning tasks to different learning modules according to the task classification corresponding to the maximum value of the memory size; Calculate a dimension score based on the user's usage of each of the learning modules in the optimal learning classification model, and classify the users into the single-type and the mixed-type according to the dimension score; The task classifications include visual type, auditory type, and reading and writing type. The single-type is a user who only has one of the task classifications, and the mixed-type is a user who has two or more of the task classifications. The learning modules include a text module, an audio module, and a video module; 4. The database task configuration method according to claim 3, wherein, The process of calculating a dimension score based on the user's usage of each of the learning modules in the optimal learning classification model, and classifying the users into the single-type and the mixed-type according to the dimension score includes: Statistically calculate the selection times and usage durations of the user using different learning modules in the optimal learning classification model; Calculate the dimension score in combination with the selection times and the usage durations, and sort the dimension scores; Determine that the user is the single type based on the difference between the highest dimension score and the second highest dimension score, or compare each of the dimension scores with a preset dimension score.

5. The database task configuration method according to claim 4, wherein The process of configuring several of the learning tasks for the user's single learning session based on the user's single learning time and the average execution time includes: Sum up the average execution times corresponding to each of the learning tasks to obtain the total configured time; Calculate the configured time difference between the total configured time and the single learning time; Compare the configured time difference with a preset configured time difference, and configure several of the learning tasks for the user's single learning session according to the comparison result; The preset configured time difference is positively correlated with the user's single learning time.

6. The database task configuration method according to claim 5, wherein Statistically calculate the several single completion degrees of each of the learning tasks in each of the single learning sessions, and generate the task forgetting curve by combining the time intervals and the corresponding single completion degrees of each of the single learning sessions.

7. The database task configuration method according to claim 6, wherein The process of matching and executing a task completion model for the user to generate a task execution log by combining the different classifications of the user and the average completion degree includes: Collect the classifications of each user and the average completion degree of each of the learning tasks in the learning classification model; Match the task completion model that is consistent with the classification of the user and the average completion degree in the learning task database; Record the situation of the user completing the task completion model to generate the task execution log; The task completion model includes the classifications of each user and the corresponding learning tasks completed.

8. The database task configuration method according to claim 7, wherein Based on the task execution log, use the time series analysis algorithm to predict the completion degree of the past learning tasks to adjust the extreme points of the task forgetting curve. The time series analysis algorithm predicts the extreme points of the task forgetting curve corresponding to the past learning tasks through the analysis of the completion degree and execution time of the past learning tasks. The extreme points are the points on the task forgetting curve with the largest curvature.

9. The database task configuration method according to claim 8, wherein, According to the adjusted task forgetting curve, reduce the time ratio of the new learning tasks and the past learning tasks at the extreme points. The single learning time includes the time for executing the new learning tasks and the time for executing the past learning tasks. The time ratio is the ratio of the time occupied by the user for executing the new learning tasks to the time occupied by the user for executing the past learning tasks.

10. The database task configuration method according to claim 9, wherein Divide the learning tasks into special learning tasks and general learning tasks according to the learning task completion situation, divide the general learning tasks into hot tasks and cold tasks. The process of generating the task completion model by combining the past learning tasks and the hot tasks includes: Divide the learning tasks that the user has not completed in the single learning session into special learning tasks, and the learning task database pushes relevant materials of the special learning tasks to the user according to the classification of the user; Divide the learning tasks that the user has completed in the single learning session into general learning tasks, and determine the hot and cold tasks based on the completion duration; Adopt a single type of allocation method for the hot tasks and a mixed type of allocation method for the cold tasks; Generate the task completion model by selecting learning tasks that are the same as the past learning tasks and the hot tasks.

Images