Teaching plan generation method, electronic device, and computer program product

Abstract

This disclosure relates to a method for generating lesson plans, as well as an electronic device and computer program product, belonging to the field of intelligent education technology. The method includes: acquiring student assessment data, which includes baseline data for all students and target group assessment data corresponding to the target student group; comparing and analyzing the baseline data for all students and the target group assessment data to construct a teaching deviation feature vector corresponding to the target student group; performing attribution diagnosis processing on the teaching deviation feature vector to obtain lesson plan optimization strategy instructions; and iteratively updating the original lesson plan corresponding to the target student group based on the lesson plan optimization strategy instructions to obtain an updated lesson plan. This disclosure can automatically convert multi-dimensional assessment data into lesson plan optimization strategy instructions and can verify the effect, significantly improving the personalization of teaching content and the iteration efficiency of lesson plans.

Description

Technical Field

[0001] This disclosure relates to the field of intelligent education technology, and more specifically, to a lesson plan generation method, a lesson plan generation device, an electronic device, and a computer program product. Background Technology

[0002] With the rapid development of artificial intelligence (AI) technology, it has been widely applied in the education field. In the teaching process, although teachers possess a large amount of student test data, it is often difficult to translate student assessment scores into concrete teaching improvement actions. For example, in some related solutions, lesson plans are mostly static text, unable to be dynamically adjusted according to students' actual understanding. Furthermore, general AI lesson preparation tools, such as those using Chat Generative Pre-trained Transformer (ChatGPT) to directly generate lessons, lack a deep understanding of the specific learning situation of a particular class, resulting in generic lesson plans that fail to address real-world problems.

[0003] It should be noted that the information disclosed in the background section above is only used to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0004] The purpose of this disclosure is to provide a method, device, electronic equipment, computer-readable storage medium, and computer program product for generating lesson plans, thereby overcoming, to at least a certain extent, the problems of disconnect between assessment data analysis and lesson plan optimization, and the inability to automatically perform precise local iterations of lesson plans in related solutions.

[0005] Other features and advantages of this disclosure will become apparent from the following detailed description, or may be learned in part by practice of the invention.

[0006] According to a first aspect of this disclosure, a method for generating lesson plans is provided, comprising: acquiring student assessment data, wherein the student assessment data includes baseline data for all students and target group assessment data corresponding to a target student group; constructing a teaching deviation feature vector corresponding to the target student group based on a comparative analysis result of the baseline data for all students and the target group assessment data; performing attribution diagnosis processing on the teaching deviation feature vector to obtain lesson plan optimization strategy instructions; and iteratively updating the original lesson plan corresponding to the target student group based on the lesson plan optimization strategy instructions to obtain an updated lesson plan.

[0007] In one exemplary embodiment of this disclosure, the target group assessment data includes historical assessment data and current assessment data; the step of constructing a teaching deviation feature vector corresponding to the target student group based on the comparative analysis results of the overall baseline data and the target group assessment data includes: performing a horizontal analysis and comparison of the overall baseline data and the target group assessment data, and using the divergence value between the error distribution of the target knowledge point in the target student group and the global error distribution as the horizontal deviation value; performing a vertical analysis and comparison of the historical assessment data and the current assessment data to determine the vertical trend value of the target student group for the target knowledge point; and constructing the teaching deviation feature vector based on the horizontal deviation value and the vertical trend value.

[0008] In one exemplary embodiment of this disclosure, constructing the teaching deviation feature vector based on the horizontal deviation value and the vertical trend value includes: obtaining candidate error patterns based on the baseline data of all participants; performing similarity matching between the error distribution of the target student group for the target knowledge point and the candidate error patterns, and determining the error pattern identifier of the target student group for the target knowledge point based on the matching result; and constructing the teaching deviation feature vector according to the horizontal deviation value, the vertical trend value and the error pattern identifier.

[0009] In one exemplary embodiment of this disclosure, the step of performing attribution diagnosis processing on the teaching deviation feature vector to obtain lesson plan optimization strategy instructions includes: inputting the teaching deviation feature vector into a pre-constructed attribution diagnosis model, the attribution diagnosis model including a rule engine and a feature mapping classifier; using the rule engine to preprocess the teaching deviation feature vector; inputting the preprocessed teaching deviation feature vector into the feature mapping classifier, and having the feature mapping classifier output the teaching defect label and lesson plan optimization strategy instructions corresponding to the teaching deviation feature vector.

[0010] In one exemplary embodiment of this disclosure, the step of outputting the teaching defect label and lesson plan optimization strategy instruction corresponding to the teaching deviation feature vector by the feature mapping classifier includes: determining the teaching defect label corresponding to the teaching deviation feature vector through the feature mapping classifier; obtaining a pre-constructed strategy knowledge graph, which is constructed based on the phenomenon-cause-strategy triplet; and performing a retrieval operation on the pre-constructed strategy knowledge graph based on the teaching deviation feature vector by the feature mapping classifier to generate the lesson plan optimization strategy instruction.

[0011] In one exemplary embodiment of this disclosure, the step of iteratively updating the original lesson plan corresponding to the target student group based on the lesson plan optimization strategy instruction to obtain the updated lesson plan includes: performing a structural parsing operation on the original lesson plan to obtain an original lesson plan tree structure, the original lesson plan tree structure including multiple types of nodes; determining the node to be modified corresponding to the original lesson plan based on the lesson plan optimization strategy instruction; and performing a local iterative update process on the node to be modified based on the lesson plan optimization strategy instruction to obtain the updated lesson plan.

[0012] In one exemplary embodiment of this disclosure, the step of performing local iterative update processing on the node to be modified based on the lesson plan optimization strategy instruction to obtain the updated lesson plan includes: rewriting the node to be modified based on the lesson plan optimization strategy instruction and related incorrect question data to obtain an updated modified node; replacing the node to be modified in the original lesson plan tree structure with the updated modified node to obtain an updated lesson plan tree structure; and performing a consistency check operation on the updated lesson plan tree structure based on the node content of the updated modified node to obtain the updated lesson plan.

[0013] In one exemplary embodiment of this disclosure, the method further includes: determining the knowledge point features and modification logic corresponding to the update / modification node; generating isomorphic variation exercises corresponding to the update / modification node based on the knowledge point features and modification logic; and generating a verification exercise set based on the isomorphic variation exercises, wherein the verification exercise set is used to verify the effectiveness of the updated lesson plan.

[0014] According to a second aspect of this disclosure, a lesson plan generation device is provided, comprising: an assessment data acquisition module for acquiring student assessment data, the student assessment data including baseline data for all students and target group assessment data corresponding to a target student group; a deviation vector construction module for constructing a teaching deviation feature vector corresponding to the target student group based on the comparative analysis results of the baseline data for all students and the target group assessment data; an optimization instruction generation module for performing attribution diagnosis processing on the teaching deviation feature vector to obtain a lesson plan optimization strategy instruction; and a lesson plan updating module for iteratively updating the original lesson plan corresponding to the target student group based on the lesson plan optimization strategy instruction to obtain an updated lesson plan.

[0015] In one exemplary embodiment of this disclosure, the target group assessment data includes historical assessment data and current assessment data; the deviation vector construction module includes a deviation vector construction unit, used for: performing a horizontal analysis and comparison between the baseline data of all participants and the target group assessment data, and taking the divergence value between the error distribution of the target knowledge point in the target student group and the global error distribution as the horizontal deviation value; performing a vertical analysis and comparison between the historical assessment data and the current assessment data, and determining the vertical trend value of the target student group for the target knowledge point; and constructing the teaching deviation feature vector based on the horizontal deviation value and the vertical trend value.

[0016] In one exemplary embodiment of this disclosure, the deviation vector construction unit includes a deviation vector construction subunit, used for: obtaining candidate error patterns based on the baseline data of all participants; performing similarity matching between the error distribution of the target student group for the target knowledge point and the candidate error patterns, and determining the error pattern identifier of the target student group for the target knowledge point based on the matching result; and constructing the teaching deviation feature vector according to the horizontal deviation value, the vertical trend value and the error pattern identifier.

[0017] In one exemplary embodiment of this disclosure, the optimization instruction generation module includes an optimization instruction generation unit, configured to: input the teaching deviation feature vector into a pre-constructed attribution diagnostic model, the attribution diagnostic model including a rule engine and a feature mapping classifier; preprocess the teaching deviation feature vector using the rule engine; input the preprocessed teaching deviation feature vector into the feature mapping classifier, and have the feature mapping classifier output the teaching defect label and lesson plan optimization strategy instruction corresponding to the teaching deviation feature vector.

[0018] In one exemplary embodiment of this disclosure, the optimization instruction generation unit includes an optimization instruction generation subunit, configured to: determine the teaching defect label corresponding to the teaching deviation feature vector through the feature mapping classifier; obtain a pre-constructed strategy knowledge graph, which is constructed based on the phenomenon-cause-strategy triplet; and generate the lesson plan optimization strategy instruction by having the feature mapping classifier perform a retrieval operation on the pre-constructed strategy knowledge graph based on the teaching deviation feature vector.

[0019] In one exemplary embodiment of this disclosure, the lesson plan update module includes a local update unit, configured to: perform a structural parsing operation on the original lesson plan to obtain an original lesson plan tree structure, the original lesson plan tree structure including multiple types of nodes; determine the node to be modified corresponding to the original lesson plan based on the lesson plan optimization strategy instruction; and perform local iterative update processing on the node to be modified based on the lesson plan optimization strategy instruction to obtain the updated lesson plan.

[0020] In one exemplary embodiment of this disclosure, the local update unit includes a local update subunit, configured to: rewrite the node to be modified based on the lesson plan optimization strategy instruction and related incorrect question data to obtain an updated modified node; replace the node to be modified in the original lesson plan tree structure with the updated modified node to obtain an updated lesson plan tree structure; and perform a consistency check operation on the updated lesson plan tree structure based on the node content of the updated modified node to obtain the updated lesson plan.

[0021] In one exemplary embodiment of this disclosure, the lesson plan generation device further includes an exercise set generation module, configured to: determine the knowledge point features and modification logic corresponding to the update / modification node; generate isomorphic variation exercises corresponding to the update / modification node based on the knowledge point features and modification logic; and generate a verification exercise set based on the isomorphic variation exercises, wherein the verification exercise set is used to verify the effectiveness of the updated lesson plan.

[0022] According to a third aspect of this disclosure, an electronic device is provided, comprising: a processor; and a memory storing computer-readable instructions that, when executed by the processor, implement the lesson plan generation method according to any one of the preceding claims.

[0023] According to a fourth aspect of this disclosure, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the lesson plan generation method according to any one of the preceding claims.

[0024] According to a fifth aspect of this disclosure, a computer program product is provided, comprising a computer program that, when executed by a processor, implements the lesson plan generation method described in any of the preceding claims.

[0025] The technical solution provided in this disclosure may include the following beneficial effects: The lesson plan generation method in the exemplary embodiments of this disclosure, on the one hand, integrates benchmark data of all participants with assessment data of the target group to construct a teaching deviation feature vector that reflects the learning characteristics of the target student group. Based on this vector, targeted lesson plan optimization strategy instructions are generated, and the lesson plans are automatically iteratively updated. On the other hand, iteratively updating the lesson plans based on the lesson plan optimization strategy instructions solves the problems of a lack of specificity in related lesson plan generation schemes and a disconnect between data analysis and teaching implementation. This enables precise iterative updates of lesson plans, significantly improving the personalization and iteration efficiency of teaching content.

[0026] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0027] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure. It is obvious that the drawings described below are merely some embodiments of this disclosure, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort. In the drawings: Figure 1 A flowchart illustrating an exemplary embodiment of the lesson plan generation method of this disclosure is shown schematically. Figure 2 This schematically illustrates an overall flowchart of a lesson plan generation method according to an exemplary embodiment of the present disclosure; Figure 3 The flowchart illustrating the construction of a multidimensional deviation feature vector according to an exemplary embodiment of the present disclosure is shown in the schematic diagram. Figure 4 A flowchart illustrating an exemplary embodiment of the present disclosure of a lesson plan optimization strategy instruction is shown. Figure 5 This illustration schematically shows a flowchart of an exemplary embodiment of the present disclosure for iteratively updating an original lesson plan based on a lesson plan optimization strategy instruction; Figure 6 A flowchart illustrating the generation of a verification exercise set according to an exemplary embodiment of this disclosure is shown schematically; Figure 7 The diagram illustrates the system architecture of a lesson plan generation method according to an exemplary embodiment of this disclosure. Figure 8 A block diagram of a lesson plan generation apparatus according to an exemplary embodiment of the present disclosure is shown schematically; Figure 9 The schematic diagram illustrates a computer-readable storage medium according to an exemplary embodiment of the present disclosure; Figure 10A block diagram of an electronic device according to an exemplary embodiment of the present disclosure is shown schematically. Detailed Implementation

[0028] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the embodiments set forth herein; rather, they are provided so that this disclosure will be thorough and complete, and will fully convey the concept of the exemplary embodiments to those skilled in the art. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted.

[0029] Furthermore, the described features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. Numerous specific details are provided in the following description to give a thorough understanding of embodiments of this disclosure. However, those skilled in the art will recognize that the technical solutions of this disclosure can be practiced without one or more of the specific details described, or other methods, components, apparatuses, steps, etc., can be employed. In other instances, well-known structures, methods, apparatuses, implementations, materials, or operations are not shown or described in detail to avoid obscuring various aspects of this disclosure.

[0030] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities can be implemented in software, or in one or more software-hardened modules, or in different network and / or processor devices and / or microcontroller devices.

[0031] According to embodiments of this disclosure, a lesson plan generation method, a lesson plan generation device, a computer-readable storage medium, an electronic device, and a computer program product are proposed.

[0032] The principles and spirit of this disclosure will be explained in detail below with reference to several representative embodiments.

[0033] In some related solutions, educational data analysis and AI-assisted lesson preparation are mainly divided into two categories: data dashboards and general generation methods. Data dashboards focus on statistically displaying exam scores and knowledge point mastery rates (e.g., a class achieving a 60% score on quadratic functions), limiting their functionality to "diagnosis" and failing to directly provide "treatment plans for improving teacher instruction." General generation methods utilize Large Language Models (LLMs) to directly generate the entire lesson plan based on input keywords. This approach is a one-way output, lacking the utilization of historical student data and cross-sectional comparative data.

[0034] The two solutions described above have the following shortcomings: 1) Disconnect between data and content: Data analysis results (diagnosis) are separated from lesson plan writing (execution). Teachers need to manually modify lesson plans based on the data, which is inefficient and relies heavily on their personal experience. 2) Lack of attribution logic: General AI cannot understand the specific reasons behind low scores (whether it's due to unclear concepts or overly complicated calculation methods), resulting in lesson plans that cannot address the root cause. 3) "Full rewriting" rather than "partial iteration": Existing tools typically generate a completely new lesson plan, disrupting the teacher's original teaching rhythm and logical framework, and lacking the ability to "surgically" remedy specific weaknesses. 4) Lack of verification loop: After lesson plan modifications, there is a lack of immediate verification methods to confirm the effectiveness of the modifications.

[0035] To achieve truly precise and targeted teaching, there is an urgent need for a technological solution that can automatically convert multi-dimensional student assessment data into instructions for modifying specific chapters of lesson plans, and can also verify the effectiveness of these instructions. This will not only improve teaching efficiency but is also key to achieving large-scale personalized education.

[0036] Based on this, in this example embodiment, a lesson plan generation method is first provided. The lesson plan generation method of this disclosure can be implemented using a server or using a terminal device. The terminal described in this disclosure can include mobile terminals such as mobile phones, tablets, laptops, handheld computers, and personal digital assistants (PDAs), as well as fixed terminals such as desktop computers. Figure 1 The illustration shows a schematic diagram of a lesson plan generation method flow according to some embodiments of the present disclosure. Reference Figure 1 The method for generating lesson plans may include the following steps: Step S110: Obtain student assessment data, which includes baseline data for all students and target group assessment data corresponding to the target student group. Step S120: Compare and analyze the baseline data of all participants with the assessment data of the target group to construct the teaching deviation feature vector corresponding to the target student group. Step S130: Perform attribution diagnosis processing on the teaching deviation feature vector to obtain lesson plan optimization strategy instructions; Step S140: Based on the lesson plan optimization strategy instructions, iteratively update the original lesson plan corresponding to the target student group to obtain the updated lesson plan.

[0037] According to the lesson plan generation method in this example embodiment, on the one hand, by integrating benchmark data of all participants and assessment data of the target group, a teaching deviation feature vector that reflects the learning characteristics of the target student group is constructed. Based on this vector, targeted lesson plan optimization strategy instructions are generated, automatically updating the lesson plan iteratively. On the other hand, iteratively updating the lesson plan based on the lesson plan optimization strategy instructions solves the problems of a lack of specificity in related lesson plan generation schemes and a disconnect between data analysis and teaching implementation. This enables precise iterative updates of lesson plans, significantly improving the personalization and iteration efficiency of teaching content.

[0038] The lesson plan generation method in this example embodiment will be further explained below.

[0039] In this disclosure, a teaching deviation feature vector is constructed by integrating and analyzing student assessment data during the teaching process. Before constructing the teaching deviation feature vector for the target student group, student assessment data is obtained. This student assessment data includes baseline data for all students and target group assessment data corresponding to the target student group. The student assessment data can be generated based on assessments of student learning performance using teaching tasks, or it can be assessment data for all students in the school / region; for example, it can include all assessment data for different grades and subjects across the school / region. This data is used to compare with the assessment data of the target student group to determine the learning characteristics of the target student group. The student assessment data may include, but is not limited to, student performance data, student performance distribution, baseline data for all students, and statistical data on incorrect knowledge points.

[0040] Benchmark data for all students can be statistical distribution data used as a benchmark for data comparison. For example, benchmark data for all students can be benchmark data determined by statistical analysis of the assessment data of all students in the school. Benchmark data for all students can include, but is not limited to, data such as knowledge point-level option distribution, knowledge point-level score rate benchmark, and typical error pattern database. Knowledge point-level option distribution can be the proportion of each option selected by all students in the school / region on a specific question or knowledge point (it can be used to calculate the overall distribution of the target knowledge point across grades). Knowledge point-level score rate benchmark can be the average score rate of all students in the school / region on each knowledge point, which can be used for horizontal benchmarking. Typical error pattern database can be frequently occurring incorrect problem-solving paths or pitfalls across the school, which can be used as a priori reference for attribution diagnosis.

[0041] The target student group can be the group of students to be analyzed. For example, the target student group can be determined by class, with students from the same class forming one target student group. To enhance the personalization of teaching tasks, this embodiment can perform targeted data analysis on the assessment data of different target student groups, thereby updating the lesson plans for the target student groups accordingly. The target group assessment data can be the assessment data corresponding to the target student group.

[0042] refer to Figure 2 , Figure 2 The diagram illustrates the overall flowchart of a lesson plan generation method according to an exemplary embodiment of this disclosure. In step S210, data collection and cleaning are performed. Taking a target class as an example, the current assessment data, historical assessment data, and baseline data for all students in the target class are collected. After collecting the aforementioned test data, the assessment data is cleaned and standardized for subsequent analysis.

[0043] In one exemplary embodiment of this disclosure, for step S120, based on the comparative analysis results of the baseline data of all participants and the assessment data of the target group, a teaching deviation feature vector corresponding to the target student group is constructed, including: performing a horizontal analysis and comparison of the baseline data of all participants and the assessment data of the target group, and taking the divergence value between the error distribution of the target knowledge point in the target student group and the global error distribution as the horizontal deviation value; performing a vertical analysis and comparison of the historical assessment data and the current assessment data to determine the vertical trend value of the target student group for the target knowledge point; and constructing a teaching deviation feature vector based on the horizontal deviation value and the vertical trend value.

[0044] After obtaining the baseline data for all staff and the target group assessment data corresponding to the target learner group, continue to refer to... Figure 2 In step S220, a teaching deviation feature vector is constructed based on the above data. The target group assessment data is obtained, which may include historical assessment data and current assessment data. Historical assessment data may be the detailed answer data of students in the target class from the past N exams. Current assessment data may be the detailed answer data of students in the target class in this assessment, and this data may include, but is not limited to, the knowledge point score rate and error distribution of the target student group.

[0045] refer to Figure 3 , Figure 3The flowchart illustrating the construction of a multidimensional deviation feature vector according to an exemplary embodiment of this disclosure is shown. After obtaining the aforementioned evaluation data, in step S310, the obtained evaluation data is used as input. In step S321, the evaluation data is compared horizontally. For example, a horizontal analysis and comparison is performed between the student benchmark data and the target group evaluation data to determine the divergence value between the error distribution of the target knowledge point in the target student group and the global error distribution.

[0046] Lateral deviation can be calculated by comparing the error distribution differences between the target class and the grade level on a specific knowledge point. The comparison result of lateral deviation can be denoted as... Calculate the error distribution of this class on a specific knowledge point. Global distribution of grades The relative entropy between them (i.e., Kullback-Leibler divergence, or KL divergence for short) is shown in Equation 1.

[0047] (Formula 1)

[0048] in, It could be the divergence value of the target learner group on the target knowledge point k; It can be the error distribution of the target student group (class i) on the target knowledge point k; i can represent different target student groups (different classes i). It can be the global error distribution of the target student group (class i) at the grade level on the target knowledge point k; Options can represent the set of different target student groups i.

[0049] if If the result of the horizontal comparison is significant, then the horizontal comparison result will be marked as a significant horizontal bias; where, This could be a pre-configured horizontal comparison reference threshold. For example, for a certain question, if the entire grade level mainly makes mistakes in step A, while the target class mainly makes mistakes in step B, then the difference is recorded.

[0050] In step S322, the assessment data is compared longitudinally. Based on historical assessment data and the current assessment data, the longitudinal trend value of the target learner group for the target knowledge point is determined. The comparison result of the longitudinal trend can be expressed as follows: Vertical trend comparison can be based on the score rate of knowledge point k in the target class's past N (e.g., N=3) exams. (corresponding time point) Using the least squares method to fit a linear trend The slope That is Specifically, as shown in Formula 2.

[0051] (Formula 2)

[0052] in, It can represent vertical trend values; It can represent the time sequence number of the i-th exam (e.g., i=1,2,…,N); It can represent the average of time sequence numbers; It can represent the score rate of knowledge point k in the i-th exam; This can represent the average score rate of knowledge point k across N exams. By calculating the longitudinal trend, we can analyze the curve of mastery change of the target knowledge point over the past N exams and calculate its slope (progress / regression / stagnation).

[0053] In step S330, the horizontal deviation value and the vertical trend value are fused to construct a teaching deviation feature vector. By introducing dual data verification of the target student group's horizontal data (spatial) and historical vertical data (temporal), a vector with diagnostic significance is extracted from the original assessment data. For example, a teaching deviation feature vector reflecting the learning characteristics of the target student group can be constructed, which can avoid misjudgment caused by fluctuations in a single test.

[0054] In one exemplary embodiment of this disclosure, a teaching deviation feature vector is constructed based on the horizontal deviation value and the vertical trend value, including: obtaining candidate error patterns based on the baseline data of all participants; performing similarity matching between the error distribution of the target student group for the target knowledge point and the candidate error patterns, and determining the error pattern identifier of the target student group for the target knowledge point based on the matching result; and constructing the teaching deviation feature vector according to the horizontal deviation value, the vertical trend value and the error pattern identifier.

[0055] Since the benchmark data for all employees includes a library of typical error modes, multiple pre-built candidate error modes can be obtained. These candidate error modes can be used... This indicates that candidate error patterns can be constructed based on frequently occurring incorrect problem-solving paths or misconceptions across the entire school / region. Each candidate error pattern has its own unique candidate error pattern identifier. The error pattern identifier can be the identifier corresponding to the error pattern matched by the target student group for the target knowledge point.

[0056] The error distribution (or missing scoring patterns for subjective questions) of the target learner group regarding the target knowledge point is matched with candidate error patterns in a typical error pattern library. For example, cosine similarity matching is used, and the pattern ID with the highest similarity, such as Err_Concept_Confusion_03, is selected as the error pattern identifier for the target learner group regarding the target knowledge point. After determining the error pattern identifier, it can be used as one component in constructing the teaching deviation feature vector.

[0057] For example, concatenating the lateral deviation value, the longitudinal trend value, and the error pattern identifier yields a teaching deviation feature vector. ;in, It can represent the lateral deviation value; It can represent vertical trend values. This can represent an error pattern identifier. In step S340, the teaching deviation feature vector is output. The error pattern identifier component supplements the teaching deviation feature vector with qualitative semantic information, enabling diagnosis to know not only "how large the deviation is" (as determined by...). and (Quantification) can also identify "what type of deviation".

[0058] In one exemplary embodiment of this disclosure, step S130, which involves performing attribution diagnosis processing on the teaching deviation feature vector to obtain lesson plan optimization strategy instructions, includes: inputting the teaching deviation feature vector into a pre-constructed attribution diagnosis model, which includes a rule engine and a feature mapping classifier; preprocessing the teaching deviation feature vector using the rule engine; inputting the preprocessed teaching deviation feature vector into the feature mapping classifier, which then outputs the teaching defect label and lesson plan optimization strategy instructions corresponding to the teaching deviation feature vector.

[0059] Continue to refer to Figure 2 After constructing the teaching deviation feature vector, in step S230, attribution diagnosis and strategy matching operations are performed on it. (Reference) Figure 4 , Figure 4The flowchart illustrating an exemplary embodiment of the present disclosure provides a flowchart for generating lesson plan optimization strategy instructions. In step S410, the teaching deviation feature vector is used as input to an attribution diagnostic model. The attribution diagnostic model can be a hybrid model built based on a rule engine and a feature mapping classifier. The feature mapping classifier can be a pre-built random forest classifier, or other types of classifiers, such as decision tree classifiers, gradient boosting machines (GBM) classifiers, support vector machines (SVM) classifiers, and neural network classifiers. The attribution diagnostic model can be used to analyze the teaching deviation feature vector to determine the corresponding teaching defect labels and lesson plan optimization strategy instructions.

[0060] This embodiment uses a random forest classifier as an example for illustration. In step S420, the rule engine performs preprocessing. The rule engine preprocesses the teaching deviation feature vector, and the constructed teaching deviation feature vector is then... The input is fed into a pre-built attribution diagnostic model, where obvious teaching incidents (such as all students scoring 0) are first filtered out using a rule engine. In step S430, random forest classification inference is performed. The random forest classifier is used to classify the teaching deviation feature vector... The algorithm maps the results to the top-K most probable teaching deficiency labels to identify the corresponding teaching deficiency labels. Furthermore, random forest classification can output a feature vector of teaching deviations. The corresponding lesson plan optimization strategy instructions. Teaching defect labels can be labels corresponding to defects in the teaching tasks for the target learner group. Lesson plan optimization strategy instructions can be operational instructions on how to optimize and update the lesson plan. The above steps insert attribution diagnosis between data and instruction generation, solving the problem of related solutions not understanding teaching logic. It can transform the mathematical "deviation vector" into a teaching "defect cause," thereby guiding the optimization and updating of the lesson plan.

[0061] In one exemplary embodiment of this disclosure, the feature mapping classifier outputs teaching defect labels and lesson plan optimization strategy instructions corresponding to teaching deviation feature vectors, including: determining teaching defect labels corresponding to teaching deviation feature vectors through the feature mapping classifier; obtaining a pre-constructed strategy knowledge graph, which is constructed based on phenomenon-cause-strategy triples; and performing a retrieval operation on the pre-constructed strategy knowledge graph based on the teaching deviation feature vectors to generate lesson plan optimization strategy instructions.

[0062] In this embodiment, a knowledge graph containing "phenomenon-cause-strategy" triples is pre-constructed. For example, in the knowledge graph, the "phenomenon" is a missing calculation step, the "cause" is an unclear explanation of the mathematical reasoning, and the "strategy" is to add a distribution demonstration. (Continue to refer to...) Figure 4 In step S440, the strategy knowledge graph is retrieved. By retrieving the pre-constructed strategy knowledge graph, the input teaching deviation feature vector is mapped to the "cause / defect label" in the knowledge graph, that is, the teaching defect label corresponding to the teaching deviation feature vector is determined, i.e., the "cause".

[0063] For example, if the teaching deviation feature vector shows "large horizontal differences and vertical stagnation", the model may determine that "specific teaching methods are inappropriate" (the corresponding ID may be Method_Improper). If the teaching deviation feature vector shows "common errors and no historical records", the model may determine that "new knowledge points are introduced too quickly" (the corresponding ID may be Pacing_Too_Fast).

[0064] In step S450, structured optimization instructions are generated. After determining the cause, the system uses the teaching defect label to search the strategy knowledge graph and matches the corresponding optimization strategy (i.e., the "strategy" node). The final output is a complete instruction package containing both, for example, {Defect: "Unclear reasoning", Strategy: "Add graphical demonstration"}. When the determined teaching defect label is [Concept_Confusion_Type_A], the lesson plan optimization strategy instruction can be to add a graphical demonstration of the difference between concept A and concept B. In step S460, the lesson plan optimization strategy instruction is output. Through the above processing, the system can output the teaching defect label and the lesson plan optimization strategy instruction corresponding to the teaching deviation feature vector.

[0065] In one exemplary embodiment of this disclosure, step S140 involves iteratively updating the original lesson plan corresponding to the target student group based on the lesson plan optimization strategy instruction to obtain an updated lesson plan. This includes: performing a structural parsing operation on the original lesson plan to obtain an original lesson plan tree structure, which includes multiple types of nodes; determining the nodes to be modified corresponding to the original lesson plan based on the lesson plan optimization strategy instruction; and performing a local iterative update on the nodes to be modified based on the lesson plan optimization strategy instruction to obtain an updated lesson plan.

[0066] After generating the lesson plan optimization strategy instructions, the original lesson plan can be iteratively updated based on these instructions. The original lesson plan can be a general basic lesson plan prepared by the teacher for a particular lesson, which may initially be applicable to one class or multiple parallel classes. In this embodiment, once the "teaching deviation feature vector" of a specific class is incorporated, the original lesson plan will be transformed into an adaptive lesson plan specific to that class. That is, the input is a general original lesson plan (or the version from the previous iteration), and the output is an updated lesson plan customized for that class.

[0067] Continue to refer to Figure 2 In step S240, the lesson plan structure parsing and location operation is performed. (Reference) Figure 5 , Figure 5 The diagram illustrates a flowchart of an exemplary embodiment of this disclosure, showing the iterative update of an original lesson plan based on a lesson plan optimization strategy instruction. In step S510, the original lesson plan is used as input. In step S520, the lesson plan is parsed in a structured manner. The system reads the teacher's original lesson plan and uses Natural Language Processing (NLP) technology to parse it into a tree structure, such as a Document Object Model Tree (DOM Tree), i.e., a structured tree (Lesson Tree) corresponding to the original lesson plan. The original lesson plan tree structure can contain multiple nodes of different types.

[0068] For example, the nodes in this tree structure can include, but are not limited to, nodes of the following types: Intro, Lecture, Example, and Exercise. Intro nodes can include course introductions and scenario creation. Lecture nodes can include the explanation of core concept definitions and theorem derivations. Example nodes can include the presentation and analysis of typical examples. Exercise nodes can include in-class exercises.

[0069] In other embodiments of this disclosure, other types of nodes can be configured according to specific needs. For example, the original lesson plan tree structure can be split into teaching objective nodes and introduction nodes. In this case, the node types included in the original lesson plan tree structure can include nodes of teaching objective, introduction, core explanation, example display, classroom exercises, etc. This disclosure does not impose any special limitations on the specific type of nodes.

[0070] In step S530, the node to be modified is located. Based on the lesson plan optimization strategy instructions output in the previous steps, the node requiring modification is identified, for example, the core explanation node is located based on the lesson plan optimization strategy instructions. The system, according to the lesson plan optimization strategy instructions, locks the node to be modified in the original lesson plan tree structure (e.g., only locking the "example node"), and performs local iterative update processing on the node to be modified. For example, by calling the LLM model, only the content of this node is rewritten, while keeping the content of other nodes (such as imports and explanations) unchanged. Finally, it reassembles into a complete document as the updated lesson plan. Based on the lesson plan tree structure technology, the node to be modified in the original lesson plan tree structure can be accurately located, thereby enabling local iterative updates of the lesson plan based on the node to be modified.

[0071] In one exemplary embodiment of this disclosure, a local iterative update process is performed on the node to be modified based on the lesson plan optimization strategy instruction to obtain an updated lesson plan. This includes: rewriting the node to be modified based on the lesson plan optimization strategy instruction and related incorrect question data to obtain an updated modified node; replacing the node to be modified in the original lesson plan tree structure with the updated modified node to obtain an updated lesson plan tree structure; and performing a consistency check operation on the updated lesson plan tree structure based on the node content of the updated modified node to obtain an updated lesson plan.

[0072] Continue to refer to Figure 2 In step S250, nodes are partially rewritten and reorganized. In this embodiment, the LLM model is called to partially rewrite the nodes to be modified. For example, the node content of the node to be modified, the lesson plan optimization strategy instructions, and related incorrect question data are input into the LLM model as prompts to generate new node content. In step S540, nodes are partially rewritten. Based on the new node content, the nodes to be modified in the original lesson plan tree structure are rewritten to obtain updated modified nodes; then, the modified updated modified nodes are replaced back in the lesson plan tree.

[0073] In step S550, consistency verification and reorganization are performed. A coherence check is conducted based on the updated and modified nodes to perform a consistency verification operation. Key parameters relied upon in the consistency verification operation may include, but are not limited to, node locking thresholds and contextual consistency constraints. Finally, the updated tree structure that has passed the consistency verification is rendered as a new lesson plan document, serving as the updated lesson plan. In step S560, the updated lesson plan is output. These steps, by updating and replacing nodes, enable refined and localized updates to the lesson plan, rather than a brute-force full rewrite, thus improving the efficiency of lesson plan updates.

[0074] In one exemplary embodiment of this disclosure, the knowledge point features and modification logic corresponding to the update / modification node are determined; based on the knowledge point features and modification logic, isomorphic variation exercises corresponding to the update / modification node are generated; and a verification exercise set is generated based on the isomorphic variation exercises, which is used to verify the effectiveness of the updated lesson plan.

[0075] Continue to refer to Figure 2 In step S260, a verification exercise generation operation is performed. Based on the node content of the updated and modified nodes, the system automatically generates a set of "targeted exercises." In step S270, the final solution is output. If the verification results obtained from the verification exercise set meet the preset effect of the updated lesson plan, the updated lesson plan can be determined as the final solution. Otherwise, the updated lesson plan can be iteratively updated until the determined updated lesson plan meets the expected effect of the teaching task.

[0076] refer to Figure 6 , Figure 6 The flowchart illustrating an exemplary embodiment of the present disclosure for generating a verification exercise set is shown. In step S610, the modified update node is taken as input. In step S620, knowledge point features and modification logic are extracted. In step S630, isomorphic variation questions are generated or retrieved. In step S640, the verification exercise set is output, which is used to verify the effectiveness of the modification points. The verification exercise set is specifically used to test whether students have mastered the knowledge points emphasized in the revised lesson plan, forming a closed-loop verification.

[0077] refer to Figure 7 , Figure 7 The diagram schematically illustrates the system architecture of a lesson plan generation method according to an exemplary embodiment of this disclosure. A data processing module compares and analyzes student assessment data to generate a teaching deviation feature vector corresponding to the target student group. This teaching deviation feature vector is used as input to an attribution diagnosis and strategy mapping module, which determines the lesson plan optimization strategy instructions corresponding to the feature vector. Simultaneously, a structured parser in the lesson plan structuring module performs structural parsing on the original lesson plan, generating a corresponding original lesson plan tree structure. The lesson plan optimization strategy instructions and the original lesson plan tree structure are input into a local iterative generator, which iteratively updates the original lesson plan, generating and outputting the updated lesson plan. Simultaneously, information from the local iterative generator flows to a verification exercise set generation module, which uses its built-in exercise generator to generate corresponding targeted exercises as a verification exercise set.

[0078] This embodiment takes the teaching scenario of "solving the vertex of a quadratic function in eighth grade mathematics" as an example to illustrate how to transform assessment data into local iterations at the lesson plan node level and form a verification closed loop through targeted exercises.

[0079] Step 1: Obtain student assessment data and construct a teaching deviation feature vector. After completing the unit teaching of "Graphs and Properties of Quadratic Functions," a unit assessment was organized for a class. A typical question was included: "(This question is worth 20 points): Given a quadratic function..." (1) Find the vertex coordinates of the graph of the function; (2) Determine the monotonicity of the function on the interval [0,4] and explain the reason. The evaluation results show that the average score of the whole school on this question is 14 points, and the average score of this class is only 8 points. Among them, most students can write out the intermediate steps of the formula process, but the vertex coordinates are calculated incorrectly; in the monotonicity judgment question, about 70% of the students only give the conclusion of "subtract first and then increase", lacking a strict explanation based on the vertex position.

[0080] Calculate the horizontal deviation based on the differences in scores between this class and the grade level for each option / scoring point of this question. Based on the changes in the score rate of related questions on "vertex and graph of quadratic functions" in the last three unit tests, the longitudinal trend was calculated. Ultimately, a teaching deviation feature vector is generated. .

[0081] Step 2: Attribution Diagnosis and Strategy Matching. The teaching deviation feature vector is output by the multidimensional deviation vector construction module. Input is sent to the attribution diagnosis and strategy mapping module. The rule engine first identifies: 1) The same knowledge point is generally well mastered in multiple classes, but significantly lower in this class; the trigger rule condition can be set as: ,and This means that the horizontal deviation comparison result is greater than the horizontal comparison reference threshold, and the overall score rate is higher than the pre-configured comparison threshold. , This could be a pre-configured comparison threshold. The specific processing logic is as follows: the system determines that the problem is not caused by "question difficulty" or "syllabus defects," but rather by a deviation in the execution of a specific teaching segment in the class. Therefore, the fault domain is locked onto the teaching implementation part of the lesson plan (such as the explanation method and example selection), rather than the definition of the knowledge points themselves.

[0082] 2) Errors mainly occurred in two scoring points: "the final step of calculating the vertex using the completing the square method" and "explaining the reason for monotonicity in words." Based on this, the feature mapping classifier, combined with historical error patterns, provided the Top-2 teaching deficiency labels: Teaching Deficiency Label 1. [Concept_Confusion_Vertex_Formula]: for "the vertex x-coordinate is..." The origin of “is not fully understood; teaching deficiency label 2. [Reasoning_Insufficient_Monotonicity]: The function graph cannot be strictly correlated with the interval monotonicity in the language description.

[0083] The system retrieves corresponding lesson plan optimization strategy instructions from the strategy knowledge graph, such as: 1) adding "vertex coordinate derivation animation / illustration" to "core explanation" to emphasize the equivalent transformation from general form to vertex form; 2) adding step-by-step language demonstration of monotonicity conclusions to "example demonstration" and providing a standard expression template of "first find the vertex, then divide the left and right intervals".

[0084] Step 3: Structured Analysis and Node-Level Local Rewriting of Lesson Plans. Read the teacher's original lesson plan, where the original content of the node "Example 1: Solving for the Vertex of a Quadratic Function" is "Example 1: Finding the Vertex of a Function..." The vertex coordinates are given, and their monotonicity on [0,4] is discussed. Solution: Omitted. After structured parsing, the "Example 1" node is marked as an example demonstration (i.e., "example_demonstration" type) and set as "target" according to the optimization instruction, i.e., the node to be modified. The relevant incorrect question data of Example 1 are used as typical incorrect question samples. The local iterative generator inputs "original node content (content of the node to be modified) + optimization instruction + typical incorrect solution sample" as prompt words into the LLM model to generate new node content (update the content of the modified node). The specific example is as follows: Example 1 (rewritten): (1) The formula is Therefore, the vertex is (2, -1).

[0085] (2) Because the parabola opens upwards and the x-coordinate of its vertex is 2, therefore in Monotonically decreasing above, in The interval is monotonically increasing. The interval [0,4] is contained in the two intervals mentioned above, therefore it is monotonically decreasing on [0,2] and monotonically increasing on [2,4].

[0086] Example of corresponding language expression: First write "determine based on vertex position", then write the conclusion "left decrease, right increase". After the system completes the replacement of old and new nodes in the lesson plan and the context consistency check, it automatically generates the updated lesson plan after iteration.

[0087] Step 4: Targeted Verification Exercises and Automatic Grading. Based on the knowledge point characteristics of the modified nodes, the targeted exercise generation module generates a set of isomorphic variation questions, as follows: Verification Question (20 points): Given a quadratic function... .

[0088] (1) Find the coordinates of the vertex of the function; (2) Determine the monotonicity of the function on the interval [-3,1] and explain your reasoning.

[0089] The system structures students' answers into a "step-score" sequence and automatically grades them step by step. For example, consider a student's answer: Step 1 (the method of filling in the blanks is written as follows) The process is correct, so you get 6 points (out of 6). Step 2 (Write down the vertex (-1,-4)): The result is correct, and you get 4 points (out of 4). Step 3 (Write "Monotonically decreasing on [-3,-1], monotonically increasing on [-1,1]"): The interval division is correct, and you get 6 points (out of 6). Step 4 (Explanation: "Because the parabola opens upwards and its vertex is (-1, -4)"): The reasoning is complete, earning 4 points (out of 4).

[0090] The system automatically assigns a total score of 20 points and provides tagged feedback on whether the teaching objectives have been achieved at each step, which is used for subsequent model evaluation.

[0091] Step 5: Effect Comparison and Closed-Loop Iteration. After conducting one round of teaching using the iterated lesson plan and completing the above targeted verification assessment, the system again statistically analyzed the scores of the sub-questions related to "solving the vertex of a quadratic function and explaining its monotonicity": the average score for this type of question in this class increased from 8 points to 15 points; the error rate on the "vertex coordinate calculation" scoring point decreased from 45% to 10%; and the missing rate on the "monotonicity language explanation" scoring point decreased from 70% to 15%.

[0092] The system will input the verification data again into the multidimensional deviation vector construction module. and If the preset "achievement threshold" has been reached, the status of the corresponding lesson plan node will be updated to "stable" and no new round of major adjustments will be triggered. If there are still local weaknesses, the next round of small local iterations can be triggered, thereby realizing the structured adaptive closed-loop optimization of lesson plans driven by teaching deviation feature vectors.

[0093] In summary, the lesson plan generation method disclosed herein acquires student assessment data, including baseline data for all students and target group assessment data corresponding to the target student group; it compares and analyzes the baseline data for all students and the target group assessment data to construct a teaching deviation feature vector corresponding to the target student group; it performs attribution diagnosis processing on the teaching deviation feature vector to obtain lesson plan optimization strategy instructions; and iteratively updates the original lesson plan corresponding to the target student group based on the lesson plan optimization strategy instructions to obtain an updated lesson plan. On the one hand, by integrating baseline data for all students and target group assessment data, a teaching deviation feature vector reflecting the learning characteristics of the target student group is constructed, thereby generating targeted lesson plan optimization strategy instructions based on this vector and automatically iteratively updating the lesson plan. On the other hand, iteratively updating the lesson plan based on the lesson plan optimization strategy instructions solves the problems of the lack of targeting in related lesson plan generation schemes and the disconnect between data analysis and teaching implementation, enabling precise iterative updates of lesson plans and significantly improving the personalization and iteration efficiency of teaching content. On the other hand, the generated targeted exercises are strongly correlated with the points for modifying the lesson plans, thus constructing a quantifiable effect feedback mechanism to achieve the goal of targeted closed-loop verification.

[0094] It should be noted that although the steps of the method in this invention are described in a specific order in the accompanying drawings, this does not require or imply that the steps must be performed in that specific order, or that all the steps shown must be performed to achieve the desired result. Additional or alternative steps may be omitted, multiple steps may be combined into one step, and / or one step may be broken down into multiple steps.

[0095] Furthermore, in this example embodiment, a lesson plan generation device is also provided. (See reference) Figure 8 The lesson plan generation device 800 may include: an assessment data acquisition module 810, a deviation vector construction module 820, an optimization instruction generation module 830, and a lesson plan update module 840.

[0096] Specifically, the assessment data acquisition module 810 is used to acquire student assessment data, which includes baseline data for all students and target group assessment data corresponding to the target student group; the deviation vector construction module 820 is used to construct a teaching deviation feature vector corresponding to the target student group based on the baseline data for all students and the target group assessment data; the optimization instruction generation module 830 is used to perform attribution diagnosis processing on the teaching deviation feature vector to obtain lesson plan optimization strategy instructions; and the lesson plan update module 840 is used to iteratively update the original lesson plan corresponding to the target student group based on the lesson plan optimization strategy instructions to obtain an updated lesson plan.

[0097] In one exemplary embodiment of this disclosure, the target group assessment data includes historical assessment data and current assessment data; the deviation vector construction module 820 includes a deviation vector construction unit, used for: performing a horizontal analysis and comparison of the baseline data of all participants and the target group assessment data, and taking the divergence value between the error distribution of the target knowledge point in the target student group and the global error distribution as the horizontal deviation value; performing a vertical analysis and comparison of the historical assessment data and the current assessment data to determine the vertical trend value of the target student group for the target knowledge point; and constructing a teaching deviation feature vector based on the horizontal deviation value and the vertical trend value.

[0098] In one exemplary embodiment of this disclosure, the deviation vector construction unit includes a deviation vector construction subunit, which is used to: construct candidate error patterns based on the benchmark data of all participants; perform similarity matching between the error distribution of the target student group for the target knowledge point and the candidate error patterns, and determine the error pattern identifier of the target student group for the target knowledge point based on the matching result; and construct a teaching deviation feature vector based on the horizontal deviation value, the vertical trend value and the error pattern identifier.

[0099] In one exemplary embodiment of this disclosure, the optimization instruction generation module 830 includes an optimization instruction generation unit, configured to: input a teaching deviation feature vector into a pre-built attribution diagnosis model, the attribution diagnosis model including a rule engine and a feature mapping classifier; preprocess the teaching deviation feature vector using the rule engine; input the preprocessed teaching deviation feature vector into the feature mapping classifier, and have the feature mapping classifier output the teaching defect label and lesson plan optimization strategy instruction corresponding to the teaching deviation feature vector.

[0100] In one exemplary embodiment of this disclosure, the optimization instruction generation unit includes an optimization instruction generation subunit, configured to: determine the teaching defect label corresponding to the teaching deviation feature vector through a feature mapping classifier; obtain a pre-constructed strategy knowledge graph, which is constructed based on the phenomenon-cause-strategy triplet; and generate lesson plan optimization strategy instructions by performing a retrieval operation on the pre-constructed strategy knowledge graph based on the teaching deviation feature vector by the feature mapping classifier.

[0101] In one exemplary embodiment of this disclosure, the lesson plan update module 840 includes a local update unit, configured to: perform a structural parsing operation on the original lesson plan to obtain an original lesson plan tree structure, the original lesson plan tree structure including multiple types of nodes; determine the node to be modified corresponding to the original lesson plan based on the lesson plan optimization strategy instruction; and perform local iterative update processing on the node to be modified based on the lesson plan optimization strategy instruction to obtain an updated lesson plan.

[0102] In one exemplary embodiment of this disclosure, the local update unit includes a local update subunit, configured to: rewrite the node to be modified based on the lesson plan optimization strategy instruction and related incorrect question data to obtain an updated modified node; replace the node to be modified in the original lesson plan tree structure with the updated modified node to obtain an updated lesson plan tree structure; and perform a consistency check operation on the updated lesson plan tree structure based on the node content of the updated modified node to obtain an updated lesson plan.

[0103] In one exemplary embodiment of this disclosure, the lesson plan generation device 800 further includes an exercise set generation module, used to: determine the knowledge point features and modification logic corresponding to the update / modification node; generate isomorphic variation exercises corresponding to the update / modification node based on the knowledge point features and modification logic; and generate a verification exercise set based on the isomorphic variation exercises, the verification exercise set being used to verify the effectiveness of the updated lesson plan.

[0104] The specific details of the virtual modules of each lesson plan generation device mentioned above have been described in detail in the corresponding lesson plan generation methods. For any undisclosed details, please refer to the implementation methods in the method section, and therefore will not be repeated here.

[0105] It should be noted that although several modules or units of the lesson plan generation device have been mentioned in the detailed description above, this division is not mandatory. In fact, according to embodiments of this disclosure, the features and functions of two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided and embodied by multiple modules or units.

[0106] Exemplary embodiments of this disclosure also provide a computer program product. The computer program product includes a computer program that, when executed by a processor, implements the above-described lesson plan generation method.

[0107] In one implementation, the computer program product may be a tangible product containing a computer program, such as a computer-readable storage medium storing the computer program. (See reference...) Figure 9 , Figure 9 The schematic diagram illustrates a computer-readable storage medium 900 according to an exemplary embodiment of the present disclosure. The computer-readable storage medium 900 can be a storage medium based on electrical, magnetic, optical, electromagnetic, infrared, or other signals, including but not limited to: random access memory (RAM), read-only memory (ROM), magnetic tape, floppy disk, flash memory, hard disk drive (HDD), solid-state drive (SSD), etc. Exemplarily, a computer program product can be implemented as a non-volatile storage medium storing a computer program, such as read-only memory, NAND flash memory, etc.

[0108] In one implementation, the computer program product can be an intangible product containing a computer program. For example, the computer program product can be implemented as a virtual digital product, such as an executable file, installation package, or other digital file storing the computer program.

[0109] Computer program code can be written in one or more programming languages. Examples of programming languages ​​include C, Java, and C++. Program code can execute entirely on the user's computing device, partially on the user's computing device, or as a standalone software package. It can also execute partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, such as a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via an internet connection provided by a mobile network operator).

[0110] Computer programs can be carried or transmitted via signals such as electricity, magnetism, light, electromagnetic radiation, and infrared radiation. Electronic devices can convert the signals carrying computer programs into digital signals, thereby running the computer programs. When a computer program runs on an electronic device, its code is used to cause the electronic device to execute (more specifically, to execute by the processor of the electronic device) the method steps of various exemplary embodiments of this disclosure, such as the above-described lesson plan generation method.

[0111] Exemplary embodiments of this disclosure also provide an electronic device, which may include a processor and a memory. The memory stores executable instructions of the processor, such as a computer program. The processor executes the executable instructions to perform the method steps of various exemplary embodiments of this disclosure. Furthermore, the electronic device may also include a display for displaying a graphical user interface.

[0112] The following is for reference. Figure 10 The electronic device is illustrated by way of a general-purpose computing device. It should be understood that... Figure 10 The electronic device 1000 shown is merely an example and should not be construed as limiting the functionality and scope of use of the embodiments disclosed herein.

[0113] like Figure 10 As shown, the electronic device 1000 may include: a processor 1010, a memory 1020, a bus 1030, an I / O (input / output) interface 1040, a network adapter 1050, and a display 1060.

[0114] The memory 1020 may include volatile memory, such as RAM 1021 and cache unit 1022, and may also include non-volatile memory, such as ROM 1023. The memory 1020 may also include one or more program modules 1024, such program modules 1024 including, but not limited to, an operating system, one or more application programs, other program modules, and program data. Each or some combination of these examples may include an implementation of a network environment. For example, program module 1024 may include the modules described above.

[0115] The processor 1010 may include one or more processing units, such as an AP (Application Processor), a modem processor, a GPU (Graphics Processing Unit), an ISP (Image Signal Processor), a controller, an encoder, a decoder, a DSP (Digital Signal Processor), a baseband processor, and / or an NPU (Neural-Network Processing Unit).

[0116] The processor 1010 can be used to execute executable instructions stored in the memory 1020, such as the lesson plan generation method described above.

[0117] Bus 1030 is used to connect different components of electronic device 1000, and may include data bus, address bus and control bus.

[0118] Electronic device 1000 can communicate with one or more external devices 1100 (such as keyboard, mouse, external controller, etc.) through I / O interface 1040.

[0119] Electronic device 1000 can communicate with one or more networks via network adapter 1050. For example, network adapter 1050 can provide mobile communication solutions such as 3G / 4G / 5G, or wireless communication solutions such as wireless LAN, Bluetooth, and near-field communication. Network adapter 1050 can communicate with other modules of electronic device 1000 via bus 1030.

[0120] The electronic device 1000 can display a graphical user interface through the display 1060, such as a display interface showing an updated lesson plan that has been iteratively updated.

[0121] although Figure 10As not shown in the diagram, other hardware and / or software modules may also be configured in the electronic device 1000, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0122] As can be seen from the above, the technical solutions disclosed herein can be implemented as methods, apparatus, systems, computer program products, storage media, electronic devices, etc. Those skilled in the art will understand that various aspects of this disclosure can be specifically implemented in the following forms: a completely hardware implementation, a completely software implementation (including firmware, microcode, etc.), or an implementation combining hardware and software aspects, which may be referred to as "circuit," "module," or "system," respectively.

[0123] It should be understood that this disclosure is not limited to the specific methods, steps, or structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. Those skilled in the art will readily conceive of other embodiments based on the specific implementations provided in this disclosure. Therefore, the specific implementations provided in this disclosure are merely exemplary, and the scope and spirit of this disclosure are indicated by the claims, and should cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary technical means in the art not disclosed in this disclosure.

Claims

1. A method for generating lesson plans, characterized in that, include: Acquire student assessment data, which includes baseline data for all students and target group assessment data corresponding to the target student group; Based on the comparative analysis results of the baseline data of all participants and the assessment data of the target group, a teaching deviation feature vector corresponding to the target student group is constructed. The teaching deviation feature vector is subjected to attribution diagnosis processing to obtain lesson plan optimization strategy instructions; Based on the lesson plan optimization strategy instructions, the original lesson plans corresponding to the target student group are iteratively updated to obtain updated lesson plans.

2. The method according to claim 1, characterized in that, The target group assessment data includes historical assessment data and current assessment data; the step of constructing a teaching deviation feature vector corresponding to the target student group based on the comparative analysis results of the overall baseline data and the target group assessment data includes: A horizontal analysis and comparison is performed between the baseline data of all participants and the assessment data of the target group. The divergence value between the error distribution of the target knowledge point in the target learner group and the global error distribution is used as the horizontal deviation value. By performing a longitudinal analysis and comparison between the historical assessment data and the current assessment data, the longitudinal trend value of the target student group for the target knowledge point can be determined. Based on the horizontal deviation value and the vertical trend value, the teaching deviation feature vector is constructed.

3. The method according to claim 2, characterized in that, The construction of the teaching deviation feature vector based on the horizontal deviation value and the vertical trend value includes: Candidate error patterns are obtained based on the aforementioned baseline data for all personnel. The error distribution of the target student group for the target knowledge point is matched with the candidate error patterns, and the error pattern identifier of the target student group for the target knowledge point is determined based on the matching results. The teaching deviation feature vector is constructed based on the horizontal deviation value, the vertical trend value, and the error pattern identifier.

4. The method according to claim 1, characterized in that, The attribution diagnosis processing of the teaching deviation feature vector to obtain lesson plan optimization strategy instructions includes: The teaching deviation feature vector is input into a pre-built attribution diagnostic model, which includes a rule engine and a feature mapping classifier. The rule engine is used to preprocess the teaching deviation feature vector; The preprocessed teaching deviation feature vector is input into the feature mapping classifier, which then outputs the teaching defect label and lesson plan optimization strategy instruction corresponding to the teaching deviation feature vector.

5. The method according to claim 4, characterized in that, The step of outputting the teaching defect label and lesson plan optimization strategy instruction corresponding to the teaching deviation feature vector by the feature mapping classifier includes: The teaching defect label corresponding to the teaching deviation feature vector is determined by the feature mapping classifier. Obtain a pre-constructed policy knowledge graph, which is constructed based on the phenomenon-cause-policy triplet. The feature mapping classifier performs a retrieval operation on the pre-constructed strategy knowledge graph based on the teaching deviation feature vector to generate the lesson plan optimization strategy instruction.

6. The method according to claim 1, characterized in that, The iterative update process of the original lesson plans corresponding to the target student group based on the lesson plan optimization strategy instructions to obtain updated lesson plans includes: The original lesson plan is parsed to obtain an original lesson plan tree structure, which includes multiple types of nodes. Based on the lesson plan optimization strategy instructions, determine the nodes to be modified corresponding to the original lesson plan; Based on the lesson plan optimization strategy instructions, the node to be modified is locally iteratively updated to obtain the updated lesson plan.

7. The method according to claim 6, characterized in that, The step of performing local iterative update processing on the node to be modified based on the lesson plan optimization strategy instruction to obtain the updated lesson plan includes: Based on the lesson plan optimization strategy instructions and related incorrect question data, the node to be modified is rewritten to obtain the updated and modified node. The updated lesson plan tree structure is obtained by replacing the node to be modified in the original lesson plan tree structure with the updated node. Based on the node content of the updated and modified nodes, a consistency check operation is performed on the updated lesson plan tree structure to obtain the updated lesson plan.

8. The method according to claim 7, characterized in that, The method further includes: Determine the knowledge point characteristics and modification logic corresponding to the updated / modified nodes; Based on the aforementioned knowledge point features and modification logic, isomorphic variation exercises corresponding to the updated and modified nodes are generated. Based on the isomorphic variation exercises, a verification exercise set is generated, which is used to verify the effectiveness of the updated lesson plan.

9. An electronic device, characterized in that, include: processor; Memory used to store the processor's executable instructions; The processor is configured to execute the instructions to implement the lesson plan generation method as described in any one of claims 1 to 8.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the lesson plan generation method as described in any one of claims 1 to 8.

Images