A home education intervention plan generation method and system based on closed-loop feedback

Abstract

This invention relates to the field of data processing technology, and discloses a method and system for generating family education intervention plans based on closed-loop feedback. The method includes: prioritizing a set of family education assessment dimensions according to plan generation instructions to obtain priority assessment dimensions; obtaining an original set of educational tasks based on the priority assessment dimensions; constructing a set of educational plans to be optimized based on the original set of educational tasks and a template for plans to be optimized; calculating the fitness of the educational plans to be optimized using a multidimensional education database to obtain a set of fitness to be optimized; and iterating the educational plans using the set of fitness to be optimized and the multidimensional education database to obtain the optimal educational plan. This invention can improve the efficiency and intelligence of educational intervention plan generation, and enhance the effectiveness of educational intervention.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a method and system for generating family education intervention plans based on closed-loop feedback. Background Technology

[0002] Family education interventions are of great value in promoting children's all-round development and improving family interaction patterns. The key to achieving effective support lies in how to generate precise and adjustable intervention plans based on individual differences and dynamic changes.

[0003] Traditional methods for generating family education intervention plans typically generate plans directly based on fixed assessment dimensions and static task libraries. Their drawback is that the plan generation process lacks quantitative analysis of the user's historical execution environment and real-time feedback, resulting in a low degree of matching between the plan and the user's actual situation, and the inability to dynamically iterate and optimize based on the execution results. Summary of the Invention

[0004] This invention provides a method for generating family education intervention plans based on closed-loop feedback and a computer-readable storage medium. Its main purpose is to improve the efficiency and intelligence of education intervention plan generation and enhance the effectiveness of education intervention.

[0005] To achieve the above objectives, the present invention provides a method for generating a family education intervention plan based on closed-loop feedback, comprising: Identify the family education client, determine the user ID to be intervened based on the family education client, and use the family education client to collect multidimensional data from the user ID to be intervened, thereby obtaining a multidimensional education database; Construct a set of family education assessment dimensions, which includes multiple family education assessment dimensions; Based on the family education client receiving the plan generation instruction, the family education assessment dimension set is prioritized according to the plan generation instruction to obtain the priority assessment dimension; The original educational task set is obtained based on the priority assessment dimensions. The original educational task set includes multiple original educational tasks. An optimization plan set is constructed based on the original set of educational tasks and a pre-built template of the plan to be optimized. The template of the plan to be optimized includes multiple placeholders for educational tasks. Based on the multidimensional education database, the fitness of the education plan set to be optimized is calculated to obtain the fitness set to be optimized. By using the fitness set to be optimized and a multidimensional education database to iterate the education plans to be optimized, the optimal education plan is obtained, and the family education intervention plan based on closed-loop feedback is generated.

[0006] Optionally, the step of using a family education client to collect multidimensional data from the user IDs to be intervened, resulting in a multidimensional education database, includes: Based on the family education client, historical task identification is performed on the user ID to be intervened, resulting in multiple historical education intervention plans; The history education intervention plans are extracted sequentially from multiple history education intervention plans to obtain the history education task set in the extracted history education intervention plans. The history education task set includes multiple history education tasks. User data was collected from the set of historical education tasks to obtain a historical task dataset. The historical task dataset includes multiple historical task data, and the historical task data includes: historical task difficulty value, historical task completion degree, historical task text and historical execution environment vector. By aggregating the historical task datasets corresponding to each historical education intervention program, a multidimensional education database is obtained.

[0007] Optionally, the step of collecting user data from the historical education task set to obtain a historical task dataset includes: For each history education task in the history education task set, the following operations shall be performed: A task query questionnaire is generated based on historical education tasks. The task query questionnaire includes: task execution difficulty item and task execution status item. Obtain historical task data based on the task query questionnaire; By summarizing the historical task data corresponding to each historical education task, we obtain the historical task dataset.

[0008] Optionally, obtaining historical task data based on the task query questionnaire includes: The task query questionnaire was used to query task characteristics and obtain the difficulty value and completion rate of historical tasks. Determine the historical task execution timestamp corresponding to the historical education task, and query the planned execution environment based on the user ID to be intervened using the historical task execution timestamp to obtain the historical execution environment vector; Obtain the historical task text for historical education tasks; By merging historical task difficulty values, historical task completion rates, historical task texts, and historical execution environment vectors, we obtain historical task data.

[0009] Optionally, the step of constructing the set of educational plans to be optimized based on the original set of educational tasks and the pre-built template of the plan to be optimized includes: Multiple candidate educational tasks are obtained by randomly selecting from the original set of educational tasks; Obtain multiple candidate task time periods corresponding to multiple candidate educational tasks, and concatenate the total time periods based on the multiple candidate task time periods to obtain the total task time period; If the total time period of the task is within the preset time period of the education plan, then the multiple candidate education tasks are sorted to obtain a candidate education task sequence. The candidate educational task sequence is used to fill in multiple placeholder educational tasks in the template of the plan to be optimized, thus obtaining the educational plan to be optimized. Determine the current plan size; If the current plan size is not equal to the preset standard population size, then return to the step of randomly selecting the original educational task set until the current plan size is equal to the standard population size. If the current planned size is equal to the standard population size, then the educational plans to be optimized are summarized to obtain a set of educational plans to be optimized.

[0010] Optionally, the step of calculating the fitness of the educational plans to be optimized based on the multidimensional education database to obtain the fitness set to be optimized includes: For each educational plan to be optimized in the set of educational plans to be optimized, perform the following operations: Identify multiple educational tasks to be optimized within the educational plan to be optimized, and extract the educational tasks to be optimized sequentially from these multiple educational tasks; Based on the extracted educational task to be optimized, obtain the text of the task to be optimized, extract text features from the text of the task to be optimized, and obtain the encoding vector of the task to be optimized. By using the encoding vectors of the task to be optimized in a multidimensional educational database, similar tasks can be identified, resulting in multiple similar educational tasks. Obtain multiple similar task data corresponding to multiple similar educational tasks. Each similar task data includes: the difficulty value of the similar task and the completion rate of the similar task. The matching degree of the current task is obtained by weighting and calculating multiple similar task data. Summarize the current task matching degree for each educational task to be optimized to obtain multiple current task matching degrees; The fitness to be optimized is obtained by summing the matching degrees of multiple current tasks. The fitness scores for each education plan to be optimized are summarized to obtain the fitness score set.

[0011] Optionally, the step of using the encoding vector of the task to be optimized to identify similar tasks in a multidimensional educational database yields multiple similar educational tasks, including: The tasks to be identified are extracted sequentially from the multidimensional education database to obtain the corresponding task data. The task data includes: task difficulty value, task completion degree, task text, and task execution environment vector. Construct the encoding vector of the task to be identified based on the text of the task to be identified in the task data; A comprehensive similarity assessment is performed using the encoding vectors of the task to be identified and the encoding vectors of the task to be optimized to obtain the comprehensive task similarity. The overall task similarity scores for each task to be identified are aggregated to obtain the overall task similarity set; Calculate the similarity threshold based on the task comprehensive similarity set, and use the similarity threshold to divide the task comprehensive similarity set to obtain multiple target similarities, where the target similarity is greater than the similarity threshold. Multiple similar educational tasks are identified based on the similarity of multiple objectives.

[0012] Optionally, the step of using the encoding vector of the task to be identified and the encoding vector of the task to be optimized to perform a comprehensive similarity evaluation to obtain a comprehensive task similarity includes: The task text similarity is obtained by calculating the vector cosine based on the encoding vector of the task to be identified and the encoding vector of the task to be optimized. The system receives the current home status parameters according to the planned generation instructions, and constructs the current execution environment vector based on the current home status parameters and the preset current environment status parameters. The task environment similarity is obtained by calculating the vector cosine based on the current execution environment vector and the execution environment vector to be identified in the task data. The overall task similarity is calculated based on the similarity of the task text and the similarity of the task environment.

[0013] Optionally, the step of calculating the current task matching degree by weighting multiple similar task data includes: The current task matching degree is calculated using the following formula:

[0014] in, Indicates the current task matching degree. This indicates the number of similar task data points in a dataset of multiple similar tasks. and Each represents the first of several similar educational tasks. The similarity of objectives corresponding to the first similar educational task and the first The similarity of objectives corresponding to similar educational tasks Represents the first among multiple similar task data. The completion rate of similar tasks contained in a set of similar task data. This represents an exponential function with the natural constant as its base. Indicates the first The difficulty values ​​of similar tasks included in a dataset of similar tasks.

[0015] To achieve the above objectives, the present invention also provides a family education intervention plan generation system based on closed-loop feedback, comprising: The assessment dimension construction module is used to identify the family education client, determine the user ID to be intervened based on the family education client, collect multi-dimensional data using the family education client to obtain a multi-dimensional education database, and construct a family education assessment dimension set, which includes multiple family education assessment dimensions. The education task acquisition module is used to receive plan generation instructions based on the family education client, prioritize the family education assessment dimension set according to the plan generation instructions to obtain the priority assessment dimensions, and obtain the original education task set based on the priority assessment dimensions. The original education task set includes multiple original education tasks. The education plan generation module is used to construct an education plan set to be optimized based on the original education task set and the pre-built plan template to be optimized. The plan template to be optimized includes multiple education task placeholders. The fitness of the plan set to be optimized is calculated based on the multidimensional education database to obtain the fitness set to be optimized. The optimal plan iteration module is used to iterate the educational plans using the fitness set to be optimized and the multidimensional educational database to obtain the optimal educational plan.

[0016] To address the above problems, the present invention also provides an electronic device, the electronic device comprising: Memory, storing at least one instruction; The processor executes the instructions stored in the memory to implement the family education intervention plan generation method based on closed-loop feedback described above.

[0017] To address the aforementioned problems, the present invention also provides a computer-readable storage medium storing at least one instruction, which is executed by a processor in an electronic device to implement the aforementioned method for generating a family education intervention plan based on closed-loop feedback.

[0018] To address the problems described in the background art, this invention utilizes a family education client to collect multidimensional data from the user IDs of those to be intervened, resulting in a multidimensional education database. This step, when collecting historical task data, not only records task content and user feedback but also introduces a historical execution environment vector that quantifies the execution environment. Compared to traditional methods that only focus on the task itself, this provides crucial evidence for matching tasks in similar contexts, enhancing the intelligence of plan development. Based on plan generation instructions, the family education assessment dimension set is prioritized to obtain priority assessment dimensions. This priority assessment can be dynamically determined based on the real-time dialogue topics between the user and the client. Compared to methods statically set by experts or using fixed rules, this more accurately captures the user's most pressing intervention needs, making plan generation more timely and targeted. Furthermore, this solution calculates the fitness of the educational plans to be optimized based on a multi-dimensional educational database, resulting in a fitness set to be optimized. This step, when identifying similar historical tasks, not only calculates task text similarity but also integrates task environment similarity, making the matching more accurate. Secondly, the fitness calculation comprehensively weighs the completion rate, difficulty, and overall similarity to the current task of historical tasks, enabling more accurate prediction of the potential effects of users performing new tasks. This provides a reliable quantitative evaluation indicator for plan optimization. Finally, using the fitness calculated based on multi-dimensional historical feedback, the plan is iteratively optimized using an optimization algorithm. This step achieves closed-loop feedback, ultimately generating an optimal plan that dynamically adapts to the user's historical behavior and context. Compared to static general plan recommendations, this improves the intelligence level of intervention. Therefore, this invention can improve the efficiency and intelligence of educational intervention plan generation, enhancing the effectiveness of educational intervention. Attached Figure Description

[0019] Figure 1 This is a flowchart illustrating a method for generating a family education intervention plan based on closed-loop feedback, according to an embodiment of the present invention. Figure 2 This is a functional module diagram of a family education intervention plan generation system based on closed-loop feedback provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of the structure of an electronic device that implements the family education intervention plan generation method based on closed-loop feedback, according to an embodiment of the present invention.

[0020] Explanation of reference numerals in the attached figures: 10. Electronic device; 11. Processor; 12. Memory; 13. Bus.

[0021] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0022] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0023] This application provides a method for generating family education intervention plans based on closed-loop feedback. The executing entity of this method includes, but is not limited to, at least one electronic device that can be configured to execute the method provided in this application, such as a server or a terminal. In other words, the method can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster.

[0024] Reference Figure 1 The diagram shown is a flowchart illustrating a method for generating a family education intervention plan based on closed-loop feedback, according to an embodiment of the present invention. In this embodiment, the method for generating a family education intervention plan based on closed-loop feedback includes: S1. Determine the family education client, determine the user ID to be intervened based on the family education client, and use the family education client to collect multi-dimensional data on the user ID to be intervened to obtain a multi-dimensional education database.

[0025] It is understood that the aforementioned family education client refers to a software application used to provide family education intervention services, such as a mobile app or WeChat mini-program. Users can use this family education client to perform various educational tasks (i.e., subsequent historical educational tasks), such as parent-child communication scenario simulation exercises and parent emotion management training. Simultaneously, the family education client also embeds large language models, such as ChatGPT, Wenxin Yiyan, and ChatGLM, which are used to engage in dialogue with users, thereby resolving their questions during the execution of various educational tasks. Furthermore, to better understand the user's performance status in performing various educational tasks, the family education client also pushes a questionnaire for that educational task to the user after its completion, such as a subsequent task query questionnaire, to understand the user's perceived subjective difficulty (i.e., the difficulty value of subsequent historical tasks) and the actual completion status of the task (i.e., the completion rate of subsequent historical tasks). The user ID to be intervened refers to the user ID corresponding to the family education client. The multidimensional education database refers to the database of user IDs to be intervened in the past educational task execution process, obtained after multidimensional data collection.

[0026] In detail, the process of using a family education client to collect multidimensional data from the user IDs to be intervened, resulting in a multidimensional education database, includes: Based on the family education client, historical task identification is performed on the user ID to be intervened, resulting in multiple historical education intervention plans; The history education intervention plans are extracted sequentially from multiple history education intervention plans to obtain the history education task set in the extracted history education intervention plans. The history education task set includes multiple history education tasks. User data was collected from the set of historical education tasks to obtain a historical task dataset. The historical task dataset includes multiple historical task data, and the historical task data includes: historical task difficulty value, historical task completion degree, historical task text and historical execution environment vector. By aggregating the historical task datasets corresponding to each historical education intervention program, a multidimensional education database is obtained.

[0027] It should be explained that the historical educational intervention plan refers to the educational intervention plan executed by the user corresponding to the user ID to be intervened in the past period, as recorded by the family education client. This educational intervention plan is automatically generated by the family education client and is used to intervene and improve a specific dimension of family education assessment. Each educational intervention plan consists of multiple educational tasks, which are specific educational activity units that constitute the educational intervention plan and require the user to perform, such as "conducting a nonviolent communication exercise with the child" or "completing the daily parent-child reading check-in." The generation method of the educational intervention plan here is the same as the generation method of the optimal educational plan in this solution. The historical educational task set refers to a collection of multiple historical educational tasks, where a historical educational task refers to the educational task included in the historical educational intervention plan. The historical task dataset refers to a collection of multiple historical task data, where the historical task data refers to a data set of historical task difficulty values, historical task completion rates, historical task text, and historical execution environment vectors.

[0028] In detail, the process of collecting user data from the historical education task set to obtain a historical task dataset includes: For each history education task in the history education task set, the following operations shall be performed: A task query questionnaire is generated based on historical education tasks. The task query questionnaire includes: task execution difficulty item and task execution status item. Obtain historical task data based on the task query questionnaire; By summarizing the historical task data corresponding to each historical education task, we obtain the historical task dataset.

[0029] It should be explained that the task query questionnaire refers to a pre-built questionnaire used to query the user's performance during the execution of a historical education task. After a user completes a historical education task, the family education client will receive the task query questionnaire, which the user will then fill out. The task execution difficulty item refers to a quantitative assessment item used to collect the user's perception of the execution difficulty when performing the historical education task, and the task execution status item refers to a quantitative assessment item used to collect the user's perception of the task completion status after performing the historical education task. Both the task execution difficulty item and the task execution status item in this task query questionnaire include quantitative options. For example, the task execution status item is: "To what extent do you think you have completed the goal of this task?" Both the task execution difficulty item and the task execution status item in this task query questionnaire include quantitative options, such as providing a level option from 1 to 5 for the user to choose from. The higher the value, the higher the user's perceived task difficulty, or the better the user's perception of task completion.

[0030] Specifically, obtaining historical task data based on the task query questionnaire includes: The task query questionnaire was used to query task characteristics and obtain the difficulty value and completion rate of historical tasks. Determine the historical task execution timestamp corresponding to the historical education task, and query the planned execution environment based on the user ID to be intervened using the historical task execution timestamp to obtain the historical execution environment vector; Obtain the historical task text for historical education tasks; By merging historical task difficulty values, historical task completion rates, historical task texts, and historical execution environment vectors, we obtain historical task data.

[0031] Furthermore, the historical task difficulty value refers to the quantified value selected by the user in the task execution difficulty item of the task query questionnaire. The historical task completion rate refers to the quantified value selected by the user in the task execution status item of the task query questionnaire. The historical task execution timestamp refers to the timestamp when the historical educational task was generated. The historical execution environment vector refers to a vector representing the environmental information when the historical educational task was generated. Since different environmental information will affect the feasibility and effectiveness of educational task execution, for example, the willingness to execute outdoor activity-type educational tasks will decrease in hot weather, and family time will be longer during holidays, which is suitable for arranging interactive tasks. Therefore, it is necessary to introduce this historical execution environment vector to improve the accuracy of similar educational task identification in subsequent similar task identification. The above-mentioned planned execution environment query includes: automatically obtaining device location information, weather data, date attributes, such as whether it is a weekend or holiday, or being filled in by the user. The historical task text refers to the descriptive text of the historical educational task, which is used to describe the execution method of the historical educational task. This historical task text can be formulated by relevant experts.

[0032] S2. Construct a set of family education assessment dimensions, which includes multiple family education assessment dimensions.

[0033] Understandably, the aforementioned family education assessment dimension set refers to a collection of multiple family education assessment dimensions, where each family education assessment dimension refers to an indicator used to quantitatively assess parents' behaviors and abilities in different aspects of family education. Optional dimensions for family education assessment include: parent-child communication, rule establishment, habit cultivation, and emotion management. For the parent-child communication dimension, educational tasks such as "nonviolent communication scenario practice" and "active listening and feedback" can be set. Rule establishment refers to parents setting clear and reasonable behavioral boundaries and norms within the family and assisting children in understanding and abiding by them. For this dimension, educational tasks such as "holding family meetings to jointly formulate rules," "implementing screen time management agreements," and "practicing allowance usage rules" can be set. Habit cultivation refers to parents guiding and demonstrating to help children develop good living, learning, and behavioral habits. For this dimension, tasks such as "21-day bedtime reading challenge" and "sharing fixed household chores" can be set. Emotion management refers to parents recognizing, understanding, and regulating their own emotions and effectively accepting and guiding their children's emotions. For this dimension, tasks such as "emotional diary recording and reflection" and "identifying and naming emotion cards" can be set.

[0034] S3. Receive the plan generation instruction from the family education client, and prioritize the family education assessment dimension set according to the plan generation instruction to obtain the priority assessment dimension.

[0035] It is clear that the plan generation instruction refers to an instruction initiated by the user to generate an educational intervention plan. The priority assessment dimension refers to the family education assessment dimension that requires priority intervention. The specific steps for prioritizing the family education assessment dimension set based on the plan generation instruction include: first, determining the generation time of the plan generation instruction; reading the dialogue content between the user and the family education client's large language model within a period prior to that generation time; extracting multiple dialogue topics from the dialogue content; and using the family education assessment dimension with the most dialogue topics as the priority assessment dimension, or having the user specify a family education assessment dimension as the priority assessment dimension.

[0036] S4. Obtain the original educational task set based on the priority assessment dimensions. The original educational task set includes multiple original educational tasks.

[0037] Understandably, the original educational task set refers to a collection of multiple original educational tasks, wherein the original educational tasks refer to educational tasks formulated by relevant educational experts for priority assessment dimensions.

[0038] S5. Construct an education plan set to be optimized based on the original education task set and the pre-built plan template to be optimized. The plan template to be optimized includes multiple education task placeholders.

[0039] It is clear that the "plan template to be optimized" refers to a pre-designed plan framework with a fixed structure and placeholders, and the "educational task placeholders" refer to the positions reserved in the plan template to be optimized for filling in specific educational task content (such as subsequent educational tasks to be optimized). The "set of educational plans to be optimized" refers to a collection of multiple educational plans to be optimized, wherein each educational plan to be optimized refers to a plan template for filling in specific educational task content.

[0040] In detail, the construction of the set of educational plans to be optimized based on the original set of educational tasks and the pre-built template of the plan to be optimized includes: Multiple candidate educational tasks are obtained by randomly selecting from the original set of educational tasks; Obtain multiple candidate task time periods corresponding to multiple candidate educational tasks, and concatenate the total time periods based on the multiple candidate task time periods to obtain the total task time period; If the total time period of the task is within the preset time period of the education plan, then the multiple candidate education tasks are sorted to obtain a candidate education task sequence. The candidate educational task sequence is used to fill in multiple placeholder educational tasks in the template of the plan to be optimized, thus obtaining the educational plan to be optimized. Determine the current plan size; If the current plan size is not equal to the preset standard population size, then return to the step of randomly selecting the original educational task set until the current plan size is equal to the standard population size. If the current planned size is equal to the standard population size, then the educational plans to be optimized are summarized to obtain a set of educational plans to be optimized.

[0041] It should be explained that a candidate educational task refers to a specific original educational task selected randomly from the original set of educational tasks. The candidate task time period refers to the execution time of a candidate educational task, for example, 19:00-19:30 in the evening. The total task time period refers to the union of the time periods of multiple candidate task time periods. The educational plan execution time period refers to the time period input by the user that allows the educational task to be executed, for example, for weekdays, the user inputs an educational plan execution time period of 20:00-21:00 in the evening; for holidays, the user inputs an educational plan execution time period of 9:00-11:00 or 15:00-17:00 in the evening. If the total task time period is not within the educational plan execution time period, it indicates that the time arrangement of the randomly generated multiple candidate task time periods conflicts with the user's available time.

[0042] Furthermore, the candidate educational task sequence refers to an ordered set of multiple candidate educational tasks after sorting. This sorting order can be executed from simplest to most difficult. Optionally, the sorting steps include: identifying the task difficulty value corresponding to each candidate educational task from a multidimensional educational database (the task difficulty value is obtained in the same way as the difficulty values ​​of subsequent similar tasks), and sorting the candidate educational tasks in ascending order of task difficulty value to obtain the candidate educational task sequence. The standard population size refers to the total number of educational plans to be optimized, which is set manually. This current plan size is the iteration size for subsequent educational plan iterations. Optionally, this current plan size can be set for different optimization algorithms. For example, when using a genetic algorithm, the standard population size can be set to 20 to 100; when using a particle swarm optimization algorithm, the standard population size can be set to 20 to 50. The current plan size refers to the current number of educational plans to be optimized.

[0043] S6. Calculate the fitness of the educational plans to be optimized based on the multidimensional education database to obtain the fitness set to be optimized.

[0044] It is clear that the set of fitness to be optimized refers to a collection of multiple fitness values ​​to be optimized. Each fitness value corresponds one-to-one with an educational plan to be optimized. The fitness value indicates the quality of the corresponding educational plan to be optimized. The larger the fitness value, the better the corresponding educational plan to be optimized is for the user. That is, the less difficult it is for the user to execute each educational task in the educational plan to be optimized and the higher the completion rate.

[0045] In detail, the step of calculating the fitness of the educational plans to be optimized based on the multidimensional education database to obtain the fitness set to be optimized includes: For each educational plan to be optimized in the set of educational plans to be optimized, perform the following operations: Identify multiple educational tasks to be optimized within the educational plan to be optimized, and extract the educational tasks to be optimized sequentially from these multiple educational tasks; Based on the extracted educational task to be optimized, obtain the text of the task to be optimized, extract text features from the text of the task to be optimized, and obtain the encoding vector of the task to be optimized. By using the encoding vectors of the task to be optimized in a multidimensional educational database, similar tasks can be identified, resulting in multiple similar educational tasks. Obtain multiple similar task data corresponding to multiple similar educational tasks. Each similar task data includes: the difficulty value of the similar task and the completion rate of the similar task. The matching degree of the current task is obtained by weighting and calculating multiple similar task data. Summarize the current task matching degree for each educational task to be optimized to obtain multiple current task matching degrees; The fitness to be optimized is obtained by summing the matching degrees of multiple current tasks. The fitness scores for each education plan to be optimized are summarized to obtain the fitness score set.

[0046] It should be explained that the "educational task to be optimized" refers to the candidate educational tasks in the educational plan to be optimized. The "text of the task to be optimized" refers to the descriptive text of the educational task to be optimized. The "encoding vector of the task to be optimized" refers to a fixed-dimensional vector obtained by transforming the text of the task to be optimized through feature extraction methods. Specifically, the text feature extraction method for the text of the task to be optimized is as follows: the text of the task to be optimized is input into word embedding models such as Word2Vec and GloVe, or sentence encoding models such as BERT and Sentence-BERT; the vector output by the model is the encoding vector of the task to be optimized. The "similar educational tasks" refers to educational tasks in the multi-dimensional educational database that have similar features to the encoding vector of the task to be optimized. The "similar task data" refers to the dataset of similar task difficulty values ​​and similar task completion degrees. The "similar task difficulty value" refers to the historical task difficulty values ​​included in the similar task data. The "similar task completion degree" refers to the historical task completion degree included in the similar task data. The "current task matching degree" refers to a weighted calculation value representing the degree of matching between the educational task to be optimized and a certain similar educational task; the higher the current task matching degree, the higher the degree of matching between the educational task to be optimized and the similar educational task. The fitness to be optimized can be expressed as the sum of the matching degrees of multiple current tasks.

[0047] In detail, the process involves using the encoding vector of the task to be optimized to identify similar tasks in a multi-dimensional educational database, resulting in multiple similar educational tasks, including: The tasks to be identified are extracted sequentially from the multidimensional education database to obtain the corresponding task data. The task data includes: task difficulty value, task completion degree, task text, and task execution environment vector. Construct the encoding vector of the task to be identified based on the text of the task to be identified in the task data; A comprehensive similarity assessment is performed using the encoding vectors of the task to be identified and the encoding vectors of the task to be optimized to obtain the comprehensive task similarity. The overall task similarity scores for each task to be identified are aggregated to obtain the overall task similarity set; Calculate the similarity threshold based on the task comprehensive similarity set, and use the similarity threshold to divide the task comprehensive similarity set to obtain multiple target similarities, where the target similarity is greater than the similarity threshold. Multiple similar educational tasks are identified based on the similarity of multiple objectives.

[0048] It should be explained that the task to be identified refers to historical educational tasks recorded in the multidimensional education database. The task data to be identified refers to the historical task data corresponding to the task to be identified, wherein the task difficulty value, task completion degree, task text, and execution environment vector refer to the historical task difficulty value, historical task completion degree, historical task text, and historical execution environment vector corresponding to the task to be identified, respectively. The task encoding vector refers to a fixed-dimensional vector obtained by converting the task text through feature extraction methods. The method for obtaining this task encoding vector is the same as the method for obtaining the task encoding vector to be optimized mentioned above, and will not be repeated here. The task comprehensive similarity refers to the similarity obtained after comprehensive similarity evaluation, used to represent the comprehensive similarity between the task to be identified and the educational task to be optimized in terms of text and environment. The similarity threshold refers to the threshold used to determine whether the task comprehensive similarity is high. The similarity threshold is calculated by calculating the mean and standard deviation of the task comprehensive similarity set, then the similarity threshold is: ,in, Indicates the similarity threshold. This represents the mean. The standard deviation is represented by the target similarity, which is the overall task similarity score greater than the similarity threshold. When the overall task similarity score is greater than the similarity threshold, it indicates that the overall similarity score is relatively high, meaning that the overall similarity between the task to be identified and the educational task to be optimized is relatively high.

[0049] In detail, the step of using the encoding vector of the task to be identified and the encoding vector of the task to be optimized to perform a comprehensive similarity evaluation to obtain a comprehensive task similarity includes: The task text similarity is obtained by calculating the vector cosine based on the encoding vector of the task to be identified and the encoding vector of the task to be optimized. The system receives the current home status parameters according to the planned generation instructions, and constructs the current execution environment vector based on the current home status parameters and the preset current environment status parameters. The task environment similarity is obtained by calculating the vector cosine based on the current execution environment vector and the execution environment vector to be identified in the task data. The overall task similarity is calculated based on the similarity of the task text and the similarity of the task environment.

[0050] It should be explained that the task text similarity refers to the vector cosine value between the encoding vector of the task to be identified and the encoding vector of the task to be optimized. The greater the task text similarity, the greater the similarity between the text description of the task to be identified and the educational task to be optimized. The current family status parameter represents the combined encoding of the idle status of children and parents in the family. For example, if both the user's children and parents are idle at the time the plan generation instruction is generated, the current family status parameter is recorded as 1; if the children are in school and the parents are idle, the current family status parameter is recorded as 2; if the parents are in work and the children are idle, the current family status parameter is recorded as 3; and if the parents are in work and the children are in school, the current family status parameter is recorded as 4. The current environment status parameter refers to the environmental data such as weather, season, and date attributes at the time the plan generation instruction is queried by the family education client. The current execution environment vector refers to a vector composed of the current home status parameter and the current environment status parameter. For example, if the current home status parameter is 2 and the current environment status parameters are weather (sunny, coded as 1), (season: spring, coded as 2), (date attribute: weekday, coded as 0), then the current execution environment vector is represented as [2, 1, 2, 0]. In addition to the above parameters, the current execution environment vector can also introduce specific environmental parameters such as temperature and humidity.

[0051] Furthermore, the task environment similarity refers to the vector cosine value between the current execution environment vector and the execution environment vector to be identified. The greater the task environment similarity, the greater the similarity between the task to be identified and the educational task to be optimized in terms of their external environments. The specific calculation formula for calculating the comprehensive task similarity based on task text similarity and task environment similarity is as follows: ,in, Indicates the overall similarity of tasks. and These represent the weights for calculating task text similarity and task environment similarity, respectively. and This can be set by the user, and the sum of the two values ​​is 1; the default value is 0.5. and These represent task text similarity and task environment similarity, respectively. Compared to traditional technologies that only focus on the semantic similarity of the task content itself when recommending or matching tasks (i.e., only calculating task text similarity), this solution introduces task environment similarity and performs weighted synthesis, so that the identified similar educational tasks are not only educational tasks with similar historical content, but also educational tasks with similar environmental conditions. This improves the accuracy of similar task matching.

[0052] Specifically, the step of calculating the current task matching degree by weighting multiple similar task data includes: The current task matching degree is calculated using the following formula:

[0053] in, Indicates the current task matching degree. This indicates the number of similar task data points in a dataset of multiple similar tasks. and Each represents the first of several similar educational tasks. The similarity of objectives corresponding to the first similar educational task and the first The similarity of objectives corresponding to similar educational tasks Represents the first among multiple similar task data. The completion rate of similar tasks contained in a set of similar task data. This represents an exponential function with the natural constant as its base. Indicates the first The difficulty values ​​of similar tasks included in a dataset of similar tasks.

[0054] It needs to be explained that in the above formula for calculating the current task matching degree, The larger the value, the higher the user's completion rate for the similar educational task, indicating a greater task matching degree. The larger the value, the greater the difficulty of performing the similar educational task relative to the user, meaning a higher degree of task matching. The larger the value, the more similar the similar educational task is to the task to be optimized. Item and The higher the percentage of the calculation for the item, the higher the percentage of the denominator. This term is used to normalize molecules.

[0055] S7. Using the fitness set to be optimized and the multidimensional education database, iterate the education plans to obtain the optimal education plan and complete the generation of a family education intervention plan based on closed-loop feedback.

[0056] It is clear that the optimal education plan refers to the educational intervention plan obtained after iterative education planning. The educational tasks included in this optimal education plan have relatively low execution difficulty and a high completion rate for users of the family education client. The above-mentioned education plan iteration can be implemented using intelligent optimization algorithms such as particle swarm optimization and genetic optimization. The specific calculation process is existing technology and will not be elaborated here.

[0057] To address the problems described in the background art, this invention utilizes a family education client to collect multidimensional data from the user IDs of those to be intervened, resulting in a multidimensional education database. This step, when collecting historical task data, not only records task content and user feedback but also introduces a historical execution environment vector that quantifies the execution environment. Compared to traditional methods that only focus on the task itself, this provides crucial evidence for matching tasks in similar contexts, enhancing the intelligence of plan development. Based on plan generation instructions, the family education assessment dimension set is prioritized to obtain priority assessment dimensions. This priority assessment can be dynamically determined based on the real-time dialogue topics between the user and the client. Compared to methods statically set by experts or using fixed rules, this more accurately captures the user's most pressing intervention needs, making plan generation more timely and targeted. Furthermore, this solution calculates the fitness of the educational plans to be optimized based on a multi-dimensional educational database, resulting in a fitness set to be optimized. This step, when identifying similar historical tasks, not only calculates task text similarity but also integrates task environment similarity, making the matching more accurate. Secondly, the fitness calculation comprehensively weighs the completion rate, difficulty, and overall similarity to the current task of historical tasks, enabling more accurate prediction of the potential effects of users performing new tasks. This provides a reliable quantitative evaluation indicator for plan optimization. Finally, using the fitness calculated based on multi-dimensional historical feedback, the plan is iteratively optimized using an optimization algorithm. This step achieves closed-loop feedback, ultimately generating an optimal plan that dynamically adapts to the user's historical behavior and context. Compared to static general plan recommendations, this improves the intelligence level of intervention. Therefore, this invention can improve the efficiency and intelligence of educational intervention plan generation, enhancing the effectiveness of educational intervention.

[0058] like Figure 2 The diagram shown is a functional block diagram of a family education intervention plan generation system based on closed-loop feedback provided in an embodiment of the present invention.

[0059] The closed-loop feedback-based family education intervention plan generation system 100 of this invention can be installed in an electronic device. Depending on the functions implemented, the closed-loop feedback-based family education intervention plan generation system 100 may include an assessment dimension construction module 101, an education task acquisition module 102, an education plan generation module 103, and an optimal plan iteration module 104. The module described in this invention can also be called a unit, which refers to a series of computer program segments that can be executed by the processor of an electronic device and can perform a fixed function, and which are stored in the memory of the electronic device. The assessment dimension construction module 101 is used to determine the family education client, determine the user ID to be intervened based on the family education client, collect multi-dimensional data using the family education client to obtain a multi-dimensional education database, and construct a family education assessment dimension set, wherein the family education assessment dimension set includes multiple family education assessment dimensions. The education task acquisition module 102 is used to receive a plan generation instruction based on the family education client, perform priority evaluation on the family education assessment dimension set according to the plan generation instruction to obtain the priority evaluation dimension, and obtain the original education task set according to the priority evaluation dimension, wherein the original education task set includes multiple original education tasks. The education plan generation module 103 is used to construct an education plan set to be optimized based on the original education task set and the pre-built plan template to be optimized. The plan template to be optimized includes multiple education task placeholders. The fitness of the plan set to be optimized is calculated according to the multidimensional education database to obtain the fitness set to be optimized. The optimal plan iteration module 104 is used to iterate the educational plans using the fitness set to be optimized and the multidimensional educational database to obtain the optimal educational plan.

[0060] In detail, the modules in the family education intervention plan generation system 100 based on closed-loop feedback described in this embodiment of the invention employ the same methods as described above. Figure 1 The method used is the same as the closed-loop feedback-based family education intervention plan generation method described in the article, and can produce the same technical effect, so it will not be elaborated here.

[0061] like Figure 3 The diagram shown is a schematic representation of an electronic device that implements a method for generating family education intervention plans based on closed-loop feedback, according to an embodiment of the present invention.

[0062] The electronic device 1 may include a processor 10, a memory 11 and a bus 12, and may also include a computer program stored in the memory 11 and capable of running on the processor 10, such as a method program for generating family education intervention plans based on closed-loop feedback.

[0063] The memory 11 includes at least one type of readable storage medium, such as flash memory, portable hard drive, multimedia card, card-type memory (e.g., SD or DX memory), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the memory 11 can be an internal storage unit of the electronic device 1, such as the portable hard drive of the electronic device 1. In other embodiments, the memory 11 can be an external storage device of the electronic device 1, such as a plug-in portable hard drive, smart media card (SMC), secure digital card (SD), flash card, etc., equipped on the electronic device 1. Furthermore, the memory 11 includes both internal storage units and external storage devices of the electronic device 1. The memory 11 can be used not only to store application software and various types of data installed on the electronic device 1, such as the code of a family education intervention plan generation method program based on closed-loop feedback, but also to temporarily store data that has been output or will be output.

[0064] In some embodiments, the processor 10 may be composed of integrated circuits, such as a single packaged integrated circuit or multiple integrated circuits with the same or different functions, including combinations of one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and various control chips. The processor 10 is the control unit of the electronic device, connecting various components of the entire electronic device through various interfaces and lines. It executes programs or modules stored in the memory 11 (e.g., a method for generating family education intervention plans based on closed-loop feedback) and calls data stored in the memory 11 to perform various functions of the electronic device 1 and process data.

[0065] The bus 12 can be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus 12 can be divided into an address bus, a data bus, a control bus, etc. The bus 12 is configured to realize the connection and communication between the memory 11 and at least one processor 10, etc.

[0066] Figure 3 Only electronic devices with components are shown; it will be understood by those skilled in the art that... Figure 3The structure shown does not constitute a limitation on the electronic device 1, and may include fewer or more components than shown, or combine certain components, or have different component arrangements.

[0067] For example, although not shown, the electronic device 1 may also include a power supply (such as a battery) to power the various components. Preferably, the power supply can be logically connected to the at least one processor 10 through a power management device, thereby enabling functions such as charging management, discharging management, and power consumption management. The power supply may also include one or more DC or AC power supplies, recharging devices, power fault detection circuits, power converters or inverters, power status indicators, and other arbitrary components. The electronic device 1 may also include various sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be described in detail here.

[0068] Furthermore, the electronic device 1 may also include a network interface. Optionally, the network interface may include a wired interface and / or a wireless interface (such as a Wi-Fi interface, a Bluetooth interface, etc.), which is typically used to establish communication connections between the electronic device 1 and other electronic devices.

[0069] Optionally, the electronic device 1 may further include a user interface, which may be a display, an input unit (such as a keyboard), and optionally, a standard wired interface or a wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen, etc. The display may also be appropriately referred to as a screen or display unit, used to display information processed in the electronic device 1 and to display a visual user interface.

[0070] The program for generating a family education intervention plan based on closed-loop feedback, stored in the memory 11 of the electronic device 1, is a combination of multiple instructions. When run in the processor 10, it can achieve the following: Identify the family education client, determine the user ID to be intervened based on the family education client, and use the family education client to collect multidimensional data from the user ID to be intervened, thereby obtaining a multidimensional education database; Construct a set of family education assessment dimensions, which includes multiple family education assessment dimensions; Based on the family education client receiving the plan generation instruction, the family education assessment dimension set is prioritized according to the plan generation instruction to obtain the priority assessment dimension; The original educational task set is obtained based on the priority assessment dimensions. The original educational task set includes multiple original educational tasks. An optimization plan set is constructed based on the original set of educational tasks and a pre-built template of the plan to be optimized. The template of the plan to be optimized includes multiple placeholders for educational tasks. Based on the multidimensional education database, the fitness of the education plan set to be optimized is calculated to obtain the fitness set to be optimized. By using the fitness set to be optimized and a multidimensional education database to iterate the education plans to be optimized, the optimal education plan is obtained, and the family education intervention plan based on closed-loop feedback is generated.

[0071] Specifically, the processor 10's implementation method for the above instructions can be found in [reference needed]. Figures 1 to 3 The descriptions of the relevant steps in the corresponding embodiments are not repeated here.

[0072] Furthermore, if the modules / units integrated in the electronic device 1 are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. The computer-readable storage medium can be volatile or non-volatile. For example, the computer-readable medium may include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, or a read-only memory (ROM).

[0073] The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor of an electronic device, can perform the following: Identify the family education client, determine the user ID to be intervened based on the family education client, and use the family education client to collect multidimensional data from the user ID to be intervened, thereby obtaining a multidimensional education database; Construct a set of family education assessment dimensions, which includes multiple family education assessment dimensions; Based on the family education client receiving the plan generation instruction, the family education assessment dimension set is prioritized according to the plan generation instruction to obtain the priority assessment dimension; The original educational task set is obtained based on the priority assessment dimensions. The original educational task set includes multiple original educational tasks. An optimization plan set is constructed based on the original set of educational tasks and a pre-built template of the plan to be optimized. The template of the plan to be optimized includes multiple placeholders for educational tasks. Based on the multidimensional education database, the fitness of the education plan set to be optimized is calculated to obtain the fitness set to be optimized. By using the fitness set to be optimized and a multidimensional education database to iterate the education plans to be optimized, the optimal education plan is obtained, and the family education intervention plan based on closed-loop feedback is generated.

[0074] In the embodiments provided by this invention, it should be understood that the disclosed devices, systems, and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative, and actual implementations may have other classification methods.

[0075] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0076] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0077] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0078] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for generating family education intervention plans based on closed-loop feedback, characterized in that, The method includes: Identify the family education client, determine the user ID to be intervened based on the family education client, and use the family education client to collect multidimensional data from the user ID to be intervened, thereby obtaining a multidimensional education database; Construct a set of family education assessment dimensions, which includes multiple family education assessment dimensions; Based on the family education client receiving the plan generation instruction, the family education assessment dimension set is prioritized according to the plan generation instruction to obtain the priority assessment dimension; The original educational task set is obtained based on the priority assessment dimensions. The original educational task set includes multiple original educational tasks. An optimization plan set is constructed based on the original set of educational tasks and a pre-built template of the plan to be optimized. The template of the plan to be optimized includes multiple placeholders for educational tasks. Based on the multidimensional education database, the fitness of the education plan set to be optimized is calculated to obtain the fitness set to be optimized. By using the fitness set to be optimized and a multidimensional education database to iterate the education plans to be optimized, the optimal education plan is obtained, and the family education intervention plan based on closed-loop feedback is generated.

2. The method for generating a family education intervention plan based on closed-loop feedback as described in claim 1, characterized in that, The process involves using a family education client to collect multidimensional data from the user IDs of those to be intervened, resulting in a multidimensional education database, including: Based on the family education client, historical task identification is performed on the user ID to be intervened, resulting in multiple historical education intervention plans; The history education intervention plans are extracted sequentially from multiple history education intervention plans to obtain the history education task set in the extracted history education intervention plans. The history education task set includes multiple history education tasks. User data was collected from the set of historical education tasks to obtain a historical task dataset. The historical task dataset includes multiple historical task data, and the historical task data includes: historical task difficulty value, historical task completion degree, historical task text and historical execution environment vector. By aggregating the historical task datasets corresponding to each historical education intervention program, a multidimensional education database is obtained.

3. The method for generating a family education intervention plan based on closed-loop feedback as described in claim 2, characterized in that, The process of collecting user data from the historical education task set to obtain the historical task dataset includes: For each history education task in the history education task set, the following operations shall be performed: A task query questionnaire is generated based on historical education tasks. The task query questionnaire includes: task execution difficulty item and task execution status item. Obtain historical task data based on the task query questionnaire; By summarizing the historical task data corresponding to each historical education task, we obtain the historical task dataset.

4. The method for generating a family education intervention plan based on closed-loop feedback as described in claim 3, characterized in that, The step of obtaining historical task data based on the task query questionnaire includes: The task query questionnaire was used to query task characteristics and obtain the difficulty value and completion rate of historical tasks. Determine the historical task execution timestamp corresponding to the historical education task, and query the planned execution environment based on the user ID to be intervened using the historical task execution timestamp to obtain the historical execution environment vector; Obtain the historical task text for historical education tasks; By merging historical task difficulty values, historical task completion rates, historical task texts, and historical execution environment vectors, we obtain historical task data.

5. The method for generating a family education intervention plan based on closed-loop feedback as described in claim 4, characterized in that, The process of constructing the set of educational plans to be optimized based on the original set of educational tasks and a pre-built template of the plan to be optimized includes: Multiple candidate educational tasks are obtained by randomly selecting from the original set of educational tasks; Obtain multiple candidate task time periods corresponding to multiple candidate educational tasks, and concatenate the total time periods based on the multiple candidate task time periods to obtain the total task time period; If the total time period of the task is within the preset time period of the education plan, then the multiple candidate education tasks are sorted to obtain a candidate education task sequence. The candidate educational task sequence is used to fill in multiple placeholder educational tasks in the template of the plan to be optimized, thus obtaining the educational plan to be optimized. Determine the current plan size; If the current plan size is not equal to the preset standard population size, then return to the step of randomly selecting the original educational task set until the current plan size is equal to the standard population size. If the current planned size is equal to the standard population size, then the educational plans to be optimized are summarized to obtain a set of educational plans to be optimized.

6. The method for generating a family education intervention plan based on closed-loop feedback as described in claim 5, characterized in that, The step of calculating the fitness of the educational plans to be optimized based on the multidimensional education database yields a fitness set to be optimized, including: For each educational plan to be optimized in the set of educational plans to be optimized, perform the following operations: Identify multiple educational tasks to be optimized within the educational plan to be optimized, and extract the educational tasks to be optimized sequentially from these multiple educational tasks; Based on the extracted educational task to be optimized, obtain the text of the task to be optimized, extract text features from the text of the task to be optimized, and obtain the encoding vector of the task to be optimized. By using the encoding vectors of the task to be optimized in a multidimensional educational database, similar tasks can be identified, resulting in multiple similar educational tasks. Obtain multiple similar task data corresponding to multiple similar educational tasks. Each similar task data includes: the difficulty value of the similar task and the completion rate of the similar task. The matching degree of the current task is obtained by weighting and calculating multiple similar task data. Summarize the current task matching degree for each educational task to be optimized to obtain multiple current task matching degrees; The fitness to be optimized is obtained by summing the matching degrees of multiple current tasks. The fitness scores for each education plan to be optimized are summarized to obtain the fitness score set.

7. The method for generating a family education intervention plan based on closed-loop feedback as described in claim 6, characterized in that, The process involves using the encoding vector of the task to be optimized to identify similar tasks in a multidimensional educational database, resulting in multiple similar educational tasks, including: The tasks to be identified are extracted sequentially from the multidimensional education database to obtain the corresponding task data. The task data includes: task difficulty value, task completion degree, task text, and task execution environment vector. Construct the encoding vector of the task to be identified based on the text of the task to be identified in the task data; A comprehensive similarity assessment is performed using the encoding vectors of the task to be identified and the encoding vectors of the task to be optimized to obtain the comprehensive task similarity. The overall task similarity scores for each task to be identified are aggregated to obtain the overall task similarity set; Calculate the similarity threshold based on the task comprehensive similarity set, and use the similarity threshold to divide the task comprehensive similarity set to obtain multiple target similarities, where the target similarity is greater than the similarity threshold. Multiple similar educational tasks are identified based on the similarity of multiple objectives.

8. The method for generating a family education intervention plan based on closed-loop feedback as described in claim 7, characterized in that, The comprehensive similarity evaluation using the encoding vector of the task to be identified and the encoding vector of the task to be optimized, to obtain the comprehensive task similarity, includes: The task text similarity is obtained by calculating the vector cosine based on the encoding vector of the task to be identified and the encoding vector of the task to be optimized. The system receives the current home status parameters according to the planned generation instructions, and constructs the current execution environment vector based on the current home status parameters and the preset current environment status parameters. The task environment similarity is obtained by calculating the vector cosine based on the current execution environment vector and the execution environment vector to be identified in the task data. The overall task similarity is calculated based on the similarity of the task text and the similarity of the task environment.

9. The method for generating a family education intervention plan based on closed-loop feedback as described in claim 8, characterized in that, The step of calculating the current task matching degree by weighting multiple similar task data includes: The current task matching degree is calculated using the following formula: ； in, Indicates the current task matching degree. This indicates the number of similar task data points in a dataset of multiple similar tasks. and Each represents the first of several similar educational tasks. The similarity of objectives corresponding to the first similar educational task and the first The similarity of objectives corresponding to similar educational tasks Represents the first among multiple similar task data. The completion rate of similar tasks contained in a set of similar task data. This represents an exponential function with the natural constant as its base. Indicates the first The difficulty values ​​of similar tasks included in a dataset of similar tasks.

10. A family education intervention plan generation system based on closed-loop feedback, characterized in that, The system includes: The assessment dimension construction module is used to identify the family education client, determine the user ID to be intervened based on the family education client, collect multi-dimensional data using the family education client to obtain a multi-dimensional education database, and construct a family education assessment dimension set, which includes multiple family education assessment dimensions. The education task acquisition module is used to receive plan generation instructions based on the family education client, prioritize the family education assessment dimension set according to the plan generation instructions to obtain the priority assessment dimensions, and obtain the original education task set based on the priority assessment dimensions. The original education task set includes multiple original education tasks. The education plan generation module is used to construct an education plan set to be optimized based on the original education task set and the pre-built plan template to be optimized. The plan template to be optimized includes multiple education task placeholders. The fitness of the plan set to be optimized is calculated based on the multidimensional education database to obtain the fitness set to be optimized. The optimal plan iteration module is used to iterate the educational plans using the fitness set to be optimized and the multidimensional educational database to obtain the optimal educational plan.

Images