Teaching program generation method and system based on student performance change analysis

Abstract

The application discloses a kind of teaching scheme generation method and system based on student achievement change analysis, which comprises the following steps: obtaining a plurality of historical achievement data of a target student at a plurality of historical time points;According to the historical achievement data and the corresponding historical time point, a plurality of achievement change curves corresponding to the target student are constructed;Based on graph neural network, according to the plurality of achievement change curves, the achievement influencing factors corresponding to the target student are predicted;According to the achievement influencing factors and the achievement change curve, the teaching scheme corresponding to the target student is generated based on LLM algorithm model.It can be seen that the present application can realize accurate achievement influencing factor analysis and personalized teaching scheme formulation based on historical achievement trend, improve student learning effect and education resource utilization efficiency.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a method and system for generating teaching plans based on the analysis of changes in student performance. Background Technology

[0002] With the continued growth in demand for personalized education, educational institutions are increasingly emphasizing data analytics to optimize student learning outcomes and resource allocation. Existing technologies typically collect historical student performance data, employ simple statistical analysis or fixed assessment models to analyze performance trends, and develop learning plans based on templated teaching suggestions to improve teaching quality. However, existing solutions lack in-depth modeling of performance curves and dynamic prediction of influencing factors, making it difficult to accurately identify personalized performance factors. Commonly used, generic teaching plans fail to adapt to students' specific learning needs, resulting in poor learning outcomes and limiting the efficiency of educational resource utilization and the relevance of instruction. Therefore, existing technologies have shortcomings that urgently need to be addressed. Summary of the Invention

[0003] The technical problem to be solved by this invention is to provide a method and system for generating teaching plans based on the analysis of changes in student performance, which can realize accurate analysis of factors influencing performance based on historical performance trends and the formulation of personalized teaching plans, thereby improving student learning outcomes and the efficiency of educational resource utilization.

[0004] To address the aforementioned technical problems, the first aspect of this invention discloses a method for generating teaching plans based on student performance change analysis, the method comprising:

[0005] Obtain historical score data for the target student at multiple historical points in time;

[0006] Based on the historical performance data and the corresponding historical time points, multiple performance change curves for the target student are constructed.

[0007] Based on graph neural networks, the factors influencing the performance of the target student are predicted according to the multiple performance change curves.

[0008] Based on the factors influencing the performance and the performance change curve, a teaching plan corresponding to the target student is generated using the LLM algorithm model.

[0009] As an optional implementation, in the first aspect of the present invention, constructing multiple performance change curves for the target student based on the historical performance data and the corresponding historical time points includes:

[0010] For each subject, the corresponding historical score data are used as the dependent variable, and the corresponding historical time point is used as the independent variable.

[0011] Based on the mathematical fitting algorithm model, the dependent variable and the independent variable are fitted to obtain the mathematical relationship model of the performance change corresponding to the subject.

[0012] Based on the drawing algorithm, according to the mathematical relationship model of the performance change, the performance change curve corresponding to each subject is drawn on the preset canvas to obtain multiple performance change curves corresponding to the target student.

[0013] As an optional implementation, in the first aspect of the present invention, the mathematical relationship model of the performance change is a polynomial regression model with a highest order greater than two and less than five.

[0014] As an optional implementation, in the first aspect of the present invention, the step of using the multiple historical score data corresponding to the subject as the dependent variable and the corresponding historical time point as the independent variable includes:

[0015] Sort all the historical score data corresponding to the subject from earliest to latest according to the corresponding historical time points to obtain the score data sequence;

[0016] The performance data sequence is input into a trained anomaly prediction neural network to obtain the historical performance data that belongs to the anomalous data in the performance data sequence; the anomaly prediction neural network is trained on a training dataset that includes multiple training performance data sequences and corresponding data labels that belong to the anomalous performances.

[0017] All historical performance data that are considered abnormal data are removed, and the remaining historical performance data is used as the dependent variable, with the corresponding historical time point as the independent variable.

[0018] As an optional implementation, in the first aspect of the present invention, the step of predicting the performance influencing factors corresponding to the target student based on the graph neural network and according to the multiple performance change curves includes:

[0019] Each historical time point is taken as a graph node, and the historical score data of all subjects corresponding to that historical time point are determined as the node feature value corresponding to that graph node.

[0020] For any two graph nodes, the weighted average of the differences between the curve values ​​corresponding to the two graph nodes is determined in each performance change curve to obtain the adjacency feature value between the two graph nodes.

[0021] Determine whether the adjacency feature value is greater than the feature value threshold. If yes, establish a node connection between the two graph nodes and use the adjacency feature value as the connection feature value. If no, do not establish a node connection between the two graph nodes.

[0022] All the graph nodes and their corresponding node feature values, as well as the node connections between the graph nodes and their corresponding connection feature values, are determined as the performance graph feature network corresponding to the target student.

[0023] The performance graph feature network is input into a trained graph neural network to obtain the performance influencing factors corresponding to the target student; the graph neural network is trained using a training dataset that includes multiple trained performance graph feature networks and corresponding performance influencing factor annotations.

[0024] As an optional implementation, in the first aspect of the present invention, the performance influencing factors include at least one of weak subjects, weak time points, weak user characteristics, combinations of weak subjects, and combinations of weak time points.

[0025] As an optional implementation, in the first aspect of the present invention, generating a teaching plan corresponding to the target student based on the LLM algorithm model according to the performance influencing factors and the performance change curve includes:

[0026] Determine the historical time points corresponding to the curve points in each performance change curve where the rate of change is greater than a preset rate of change threshold, and obtain the set of abrupt change time points corresponding to each performance change curve.

[0027] Calculate the intersection of the sets of abrupt change time points corresponding to all the aforementioned performance change curves to obtain multiple abrupt change time points corresponding to the target student;

[0028] Based on the learning plan table corresponding to the target students, determine the learning plan to be improved for each of the abrupt change time points.

[0029] Based on the LLM algorithm model, a teaching plan corresponding to the target student is generated according to the aforementioned performance influencing factors and all the aforementioned learning plans to be improved.

[0030] As an optional implementation, in the first aspect of the present invention, the step of generating a teaching plan corresponding to the target student based on the LLM algorithm model, according to the performance influencing factors and all the learning plans to be improved, includes:

[0031] Based on preset prompt word text rules, and according to the performance influencing factors and all the learning plans to be improved, prompt words for the corresponding solutions for the target student are generated.

[0032] The prompt words generated by the scheme are input into the trained LLM algorithm model to obtain the teaching scheme corresponding to the target student; the LLM algorithm model is trained using a training dataset that includes multiple training teaching schemes and corresponding labels of performance influencing factors and labels of plans to be improved.

[0033] A second aspect of this invention discloses a teaching plan generation system based on student performance change analysis, the system comprising:

[0034] The acquisition module is used to acquire multiple historical score data of a target student at multiple historical time points;

[0035] The construction module is used to construct multiple performance change curves for the target student based on the historical performance data and the corresponding historical time points.

[0036] The prediction module is used to predict the performance influencing factors of the target student based on the multiple performance change curves using a graph neural network.

[0037] The generation module is used to generate a teaching plan for the target student based on the influencing factors of the performance and the performance change curve, using the LLM algorithm model.

[0038] As an optional implementation, in a second aspect of the invention, the specific method by which the construction module constructs multiple performance change curves corresponding to the target student based on the historical performance data and the corresponding historical time points includes:

[0039] For each subject, the corresponding historical score data are used as the dependent variable, and the corresponding historical time point is used as the independent variable.

[0040] Based on the mathematical fitting algorithm model, the dependent variable and the independent variable are fitted to obtain the mathematical relationship model of the performance change corresponding to the subject.

[0041] Based on the drawing algorithm, according to the mathematical relationship model of the performance change, the performance change curve corresponding to each subject is drawn on the preset canvas to obtain multiple performance change curves corresponding to the target student.

[0042] As an optional implementation, in the second aspect of the present invention, the mathematical relationship model of the performance change is a polynomial regression model with a highest order greater than two and less than five.

[0043] As an optional implementation, in the second aspect of the present invention, the specific method by which the construction module uses multiple historical score data corresponding to the subject as dependent variables and the corresponding historical time points as independent variables includes:

[0044] Sort all the historical score data corresponding to the subject from earliest to latest according to the corresponding historical time points to obtain the score data sequence;

[0045] The performance data sequence is input into a trained anomaly prediction neural network to obtain the historical performance data that belongs to the anomalous data in the performance data sequence; the anomaly prediction neural network is trained on a training dataset that includes multiple training performance data sequences and corresponding data labels that belong to the anomalous performances.

[0046] All historical performance data that are considered abnormal data are removed, and the remaining historical performance data is used as the dependent variable, with the corresponding historical time point as the independent variable.

[0047] As an optional implementation, in a second aspect of the invention, the prediction module, based on a graph neural network, predicts the specific methods by which it predicts the performance influencing factors corresponding to the target student based on the multiple performance change curves, including:

[0048] Each historical time point is taken as a graph node, and the historical score data of all subjects corresponding to that historical time point are determined as the node feature value corresponding to that graph node.

[0049] For any two graph nodes, the weighted average of the differences between the curve values ​​corresponding to the two graph nodes is determined in each performance change curve to obtain the adjacency feature value between the two graph nodes.

[0050] Determine whether the adjacency feature value is greater than the feature value threshold. If yes, establish a node connection between the two graph nodes and use the adjacency feature value as the connection feature value. If no, do not establish a node connection between the two graph nodes.

[0051] All the graph nodes and their corresponding node feature values, as well as the node connections between the graph nodes and their corresponding connection feature values, are determined as the performance graph feature network corresponding to the target student.

[0052] The performance graph feature network is input into a trained graph neural network to obtain the performance influencing factors corresponding to the target student; the graph neural network is trained using a training dataset that includes multiple trained performance graph feature networks and corresponding performance influencing factor annotations.

[0053] As an optional implementation, in a second aspect of the present invention, the performance influencing factors include at least one of weak subjects, weak time points, weak user characteristics, combinations of weak subjects, and combinations of weak time points.

[0054] As an optional implementation, in a second aspect of the present invention, the specific method by which the generation module generates the teaching plan corresponding to the target student based on the performance influencing factors and the performance change curve, using an LLM algorithm model, includes:

[0055] Determine the historical time points corresponding to the curve points in each performance change curve where the rate of change is greater than a preset rate of change threshold, and obtain the set of abrupt change time points corresponding to each performance change curve.

[0056] Calculate the intersection of the sets of abrupt change time points corresponding to all the aforementioned performance change curves to obtain multiple abrupt change time points corresponding to the target student;

[0057] Based on the learning plan table corresponding to the target students, determine the learning plan to be improved for each of the abrupt change time points.

[0058] Based on the LLM algorithm model, a teaching plan corresponding to the target student is generated according to the aforementioned performance influencing factors and all the aforementioned learning plans to be improved.

[0059] As an optional implementation, in a second aspect of the invention, the generation module, based on an LLM algorithm model, generates a teaching plan corresponding to the target student according to the performance influencing factors and all the learning plans to be improved, in the following specific manner:

[0060] Based on preset prompt word text rules, and according to the performance influencing factors and all the learning plans to be improved, prompt words for the corresponding solutions for the target student are generated.

[0061] The prompt words generated by the scheme are input into the trained LLM algorithm model to obtain the teaching scheme corresponding to the target student; the LLM algorithm model is trained using a training dataset that includes multiple training teaching schemes and corresponding labels of performance influencing factors and labels of plans to be improved.

[0062] A third aspect of this invention discloses another teaching plan generation system based on student performance change analysis, the system comprising:

[0063] Memory containing executable program code;

[0064] A processor coupled to the memory;

[0065] The processor calls the executable program code stored in the memory to execute some or all of the steps in the teaching plan generation method based on student performance change analysis disclosed in the first aspect of the present invention.

[0066] The fourth aspect of the present invention discloses a computer storage medium storing computer instructions, which, when invoked, are used to execute some or all of the steps in the teaching plan generation method based on student performance change analysis disclosed in the first aspect of the present invention.

[0067] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0068] This invention acquires historical performance data of target students at multiple historical time points and constructs performance change curves. It then uses graph neural networks to predict performance influencing factors and combines these factors with the performance curves to generate teaching plans using an LLM algorithm model. This enables precise analysis of performance influencing factors based on historical performance trends and the development of personalized teaching plans, thereby improving student learning outcomes and the efficiency of educational resource utilization. Attached Figure Description

[0069] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0070] Figure 1 This is a flowchart illustrating a teaching plan generation method based on student performance change analysis disclosed in an embodiment of the present invention.

[0071] Figure 2 This is a schematic diagram of the structure of a teaching plan generation system based on student performance change analysis disclosed in an embodiment of the present invention.

[0072] Figure 3 This is a schematic diagram of another teaching plan generation system based on student performance change analysis disclosed in an embodiment of the present invention. Detailed Implementation

[0073] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0074] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this invention are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, apparatus, product, or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or devices.

[0075] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0076] This invention discloses a method and system for generating teaching plans based on student performance change analysis. By acquiring historical performance data of target students at multiple historical time points and constructing performance change curves, a graph neural network is used to predict performance influencing factors. Combining these influencing factors and the performance curves, a teaching plan is generated using an LLM algorithm model. This enables accurate analysis of performance influencing factors based on historical performance trends and personalized teaching plan development, improving student learning outcomes and the efficiency of educational resource utilization. Detailed explanations follow.

[0077] Example 1

[0078] Please see Figure 1 , Figure 1 This is a flowchart illustrating a method for generating teaching plans based on student performance change analysis, as disclosed in an embodiment of the present invention. Figure 1 The described method for generating teaching plans based on student performance change analysis can be applied to data processing systems / data processing equipment / data processing servers (wherein, the server includes local processing servers or cloud processing servers). For example... Figure 1 As shown, the method for generating teaching plans based on student performance change analysis may include the following operations:

[0079] 101. Obtain multiple historical score data of the target student at multiple historical time points.

[0080] 102. Based on historical performance data and corresponding historical time points, construct multiple performance change curves for the target student.

[0081] 103. Based on graph neural networks, predict the factors influencing the performance of target students according to multiple performance change curves.

[0082] 104. Based on the factors influencing performance and the performance change curve, generate a teaching plan corresponding to the target students using the LLM algorithm model.

[0083] As can be seen, the above-described embodiments of the invention acquire historical performance data of target students at multiple historical time points and construct performance change curves, use graph neural networks to predict performance influencing factors, and combine influencing factors and performance curves to generate teaching plans through an LLM algorithm model. This enables accurate analysis of performance influencing factors based on historical performance trends and the development of personalized teaching plans, thereby improving student learning outcomes and the efficiency of educational resource utilization.

[0084] As an optional embodiment, the step above, which involves constructing multiple performance change curves for the target student based on historical performance data and corresponding historical time points, includes:

[0085] For each subject, multiple historical score data corresponding to that subject are used as the dependent variable, and the corresponding historical time point is used as the independent variable.

[0086] Based on the mathematical fitting algorithm model, the mathematical relationship model of the performance change of the subject is obtained by fitting the dependent and independent variables.

[0087] Based on the drawing algorithm and according to the mathematical relationship model of grade changes, the corresponding grade change curves for each subject are drawn on a preset canvas to obtain multiple grade change curves for the target student.

[0088] As can be seen, through the above optional embodiments, by using historical score data of each subject as the dependent variable and historical time points as independent variables, a mathematical relationship model of score change is obtained by fitting a mathematical fitting algorithm model, and a drawing algorithm is used to draw the score change curve of each subject on a preset canvas, thereby realizing the generation of accurate score change curves based on mathematical fitting, improving the data visualization accuracy and personalized education effect of the teaching plan for target students.

[0089] As an optional embodiment, the mathematical relationship model for the performance change in the above steps is a polynomial regression model with a highest order greater than two and less than five.

[0090] As can be seen, the above optional embodiments define the details of the mathematical relationship model of performance changes, so as to efficiently and cost-effectively characterize the characteristics of students' performance changes by limiting the order, assisting in the accurate analysis of performance influencing factors based on historical performance trends and the formulation of personalized teaching plans, thereby improving students' learning outcomes and the efficiency of educational resource utilization.

[0091] As an optional embodiment, the steps described above, using multiple historical score data corresponding to the subject as the dependent variable and the corresponding historical time points as independent variables, include:

[0092] Sort all historical score data for the subject from earliest to latest according to the corresponding historical time point to obtain the score data sequence;

[0093] The performance data sequence is input into a trained anomaly prediction neural network to obtain historical performance data that belongs to anomalous data in the performance data sequence; optionally, the anomaly prediction neural network is trained using a training dataset that includes multiple training performance data sequences and corresponding data labels that belong to anomalous performances.

[0094] All historical score data that are considered outliers are removed, and the remaining historical score data is used as the dependent variable, with the corresponding historical time points as the independent variables.

[0095] As can be seen, through the above optional embodiments, by sorting the historical score data of each subject from early to late according to the historical time points to generate a score data sequence, inputting it into a trained anomaly prediction neural network to identify and remove abnormal score data, and using the remaining score data as the dependent variable and the corresponding time point as the independent variable, accurate score data preprocessing based on anomaly data screening is achieved, thereby improving the accuracy of score change curve fitting and the reliability of teaching plan formulation, and reducing the risk of interference from abnormal data.

[0096] As an optional embodiment, the step above, predicting the performance influencing factors for the target student based on multiple performance change curves using a graph neural network, includes:

[0097] Each historical time point is taken as a graph node, and the historical score data of all subjects corresponding to that historical time point are determined as the node feature value corresponding to that graph node.

[0098] For any two graph nodes, determine the weighted average of the differences between the curve values ​​corresponding to the two graph nodes in each performance change curve to obtain the adjacency feature value between the two graph nodes.

[0099] Determine whether the adjacency feature value is greater than the feature value threshold. If so, establish a node connection between the two graph nodes and use the adjacency feature value as the connection feature value. If not, do not establish a node connection between the two graph nodes.

[0100] All graph nodes and their corresponding node feature values, as well as the node connections between graph nodes and their corresponding connection feature values, are determined as the graph feature network of the target student's grades.

[0101] The performance graph feature network is input into the trained graph neural network to obtain the performance influencing factors corresponding to the target student; the graph neural network is trained using a training dataset that includes multiple trained performance graph feature networks and corresponding performance influencing factor annotations.

[0102] As can be seen, through the above optional embodiments, by using historical time points as graph nodes and corresponding subject historical score data as node feature values, the weighted average of the difference between any two nodes in the score change curve is calculated as the adjacent feature value. If the feature value exceeds the feature value threshold, a node connection is established and the adjacent feature value is used as the connection feature value. This constructs a graph feature network of the target student's scores, and the trained graph neural network is input to predict the factors affecting scores, thereby achieving accurate analysis of influencing factors based on graph structure and score changes, improving the pertinence of teaching plan formulation and student learning effectiveness.

[0103] As an optional embodiment, the performance influencing factors in the above steps include at least one of weak subjects, weak time points, weak user characteristics, combinations of weak subjects, and combinations of weak time points.

[0104] As can be seen, the above optional embodiments define the content of factors influencing academic performance, effectively characterizing the characteristics of students' academic performance, assisting in the accurate analysis of factors influencing academic performance based on historical performance trends and the formulation of personalized teaching plans, thereby improving students' learning outcomes and the efficiency of educational resource utilization.

[0105] As an optional embodiment, the step above, generating a teaching plan for the target student based on the LLM algorithm model, according to the factors influencing performance and the performance change curve, includes:

[0106] Determine the historical time points corresponding to the curve points in each performance change curve where the rate of change is greater than a preset rate of change threshold, and obtain the set of abrupt change time points corresponding to each performance change curve.

[0107] Calculate the intersection of the sets of abrupt change time points corresponding to all performance change curves to obtain multiple abrupt change time points corresponding to the target student;

[0108] Based on the learning plan table corresponding to the target students, determine the learning plan to be improved for each mutation time point;

[0109] Based on the LLM algorithm model, a teaching plan corresponding to the target student is generated according to the factors affecting performance and all learning plans that need improvement.

[0110] As can be seen, through the above optional embodiments, by identifying the historical time points corresponding to the curve points in each performance change curve where the rate of change exceeds a preset threshold, a set of abrupt change time points is formed. The intersection of all curve sets is calculated to obtain the abrupt change time points of the target student. Combined with the learning plan table, the corresponding learning plan to be improved is determined. Based on the LLM algorithm model, a teaching scheme is generated using performance influencing factors and the plan to be improved. This achieves precise optimization of the teaching scheme based on performance abrupt changes and learning plans, thereby improving student learning outcomes and the relevance of educational resources.

[0111] As an optional embodiment, the above steps, based on the LLM algorithm model, generate a teaching plan corresponding to the target student according to the factors affecting performance and all learning plans to be improved, including:

[0112] Based on preset prompt text rules, and according to factors affecting performance and all learning plans to be improved, prompts are generated for the target student.

[0113] The prompt words generated by the solution are input into the trained LLM algorithm model to obtain the teaching solution corresponding to the target student; the LLM algorithm model is trained on a training dataset that includes multiple training teaching solutions and corresponding labels of performance influencing factors and improvement plans.

[0114] As can be seen, through the above optional embodiments, prompt words are generated based on preset prompt word text rules, combined with factors affecting performance and the learning plan generation scheme to be improved. These prompt words are then input into a trained LLM algorithm model to generate a teaching plan for the target student. This enables the development of precise and personalized teaching plans based on influencing factors and learning plans, thereby improving student learning outcomes and the pertinence of educational guidance.

[0115] Example 2

[0116] Please see Figure 2 , Figure 2 This is a schematic diagram of a teaching plan generation system based on student performance change analysis, as disclosed in an embodiment of the present invention. Figure 2 The described teaching plan generation system based on student performance change analysis can be applied to data processing systems / data processing equipment / data processing servers (wherein, the server includes local processing servers or cloud processing servers). For example... Figure 2 As shown, the teaching plan generation system based on student performance change analysis may include:

[0117] The acquisition module 201 is used to acquire multiple historical score data of the target student at multiple historical time points.

[0118] Module 202 is used to construct multiple performance change curves for the target student based on historical performance data and corresponding historical time points.

[0119] The prediction module 203 is used to predict the factors influencing the performance of a target student based on a graph neural network and multiple performance change curves.

[0120] The generation module 204 is used to generate teaching plans for target students based on the factors affecting performance and performance change curves, using the LLM algorithm model.

[0121] As can be seen, the above-described embodiments of the invention acquire historical performance data of target students at multiple historical time points and construct performance change curves, use graph neural networks to predict performance influencing factors, and combine influencing factors and performance curves to generate teaching plans through an LLM algorithm model. This enables accurate analysis of performance influencing factors based on historical performance trends and the development of personalized teaching plans, thereby improving student learning outcomes and the efficiency of educational resource utilization.

[0122] As an optional implementation, the construction module constructs multiple performance change curves for the target student based on historical performance data and corresponding historical time points, including the following specific methods:

[0123] For each subject, multiple historical score data corresponding to that subject are used as the dependent variable, and the corresponding historical time point is used as the independent variable.

[0124] Based on the mathematical fitting algorithm model, the mathematical relationship model of the performance change of the subject is obtained by fitting the dependent and independent variables.

[0125] Based on the drawing algorithm and according to the mathematical relationship model of grade changes, the corresponding grade change curves for each subject are drawn on a preset canvas to obtain multiple grade change curves for the target student.

[0126] As can be seen, through the above optional embodiments, by using historical score data of each subject as the dependent variable and historical time points as independent variables, a mathematical relationship model of score change is obtained by fitting a mathematical fitting algorithm model, and a drawing algorithm is used to draw the score change curve of each subject on a preset canvas, thereby realizing the generation of accurate score change curves based on mathematical fitting, improving the data visualization accuracy and personalized education effect of the teaching plan for target students.

[0127] As an optional embodiment, the mathematical model for the change in performance is a polynomial regression model with a highest order greater than two and less than five.

[0128] As can be seen, the above optional embodiments define the details of the mathematical relationship model of performance changes, so as to efficiently and cost-effectively characterize the characteristics of students' performance changes by limiting the order, assisting in the accurate analysis of performance influencing factors based on historical performance trends and the formulation of personalized teaching plans, thereby improving students' learning outcomes and the efficiency of educational resource utilization.

[0129] As an optional implementation, the construction module uses multiple historical score data corresponding to the subject as dependent variables and the corresponding historical time points as independent variables in the following ways:

[0130] Sort all historical score data for the subject from earliest to latest according to the corresponding historical time point to obtain the score data sequence;

[0131] The performance data sequence is input into a trained anomaly prediction neural network to obtain historical performance data that belongs to anomalous data in the performance data sequence; optionally, the anomaly prediction neural network is trained using a training dataset that includes multiple training performance data sequences and corresponding data labels that belong to anomalous performances.

[0132] All historical score data that are considered outliers are removed, and the remaining historical score data is used as the dependent variable, with the corresponding historical time points as the independent variables.

[0133] As can be seen, through the above optional embodiments, by sorting the historical score data of each subject from early to late according to the historical time points to generate a score data sequence, inputting it into a trained anomaly prediction neural network to identify and remove abnormal score data, and using the remaining score data as the dependent variable and the corresponding time point as the independent variable, accurate score data preprocessing based on anomaly data screening is achieved, thereby improving the accuracy of score change curve fitting and the reliability of teaching plan formulation, and reducing the risk of interference from abnormal data.

[0134] As an optional embodiment, the prediction module, based on a graph neural network, predicts the specific methods by which it identifies the factors influencing a target student's academic performance based on multiple performance change curves, including:

[0135] Each historical time point is taken as a graph node, and the historical score data of all subjects corresponding to that historical time point are determined as the node feature value corresponding to that graph node.

[0136] For any two graph nodes, determine the weighted average of the differences between the curve values ​​corresponding to the two graph nodes in each performance change curve to obtain the adjacency feature value between the two graph nodes.

[0137] Determine whether the adjacency feature value is greater than the feature value threshold. If so, establish a node connection between the two graph nodes and use the adjacency feature value as the connection feature value. If not, do not establish a node connection between the two graph nodes.

[0138] All graph nodes and their corresponding node feature values, as well as the node connections between graph nodes and their corresponding connection feature values, are determined as the graph feature network of the target student's grades.

[0139] The performance graph feature network is input into the trained graph neural network to obtain the performance influencing factors corresponding to the target student; the graph neural network is trained using a training dataset that includes multiple trained performance graph feature networks and corresponding performance influencing factor annotations.

[0140] As can be seen, through the above optional embodiments, by using historical time points as graph nodes and corresponding subject historical score data as node feature values, the weighted average of the difference between any two nodes in the score change curve is calculated as the adjacent feature value. If the feature value exceeds the feature value threshold, a node connection is established and the adjacent feature value is used as the connection feature value. This constructs a graph feature network of the target student's scores, and the trained graph neural network is input to predict the factors affecting scores, thereby achieving accurate analysis of influencing factors based on graph structure and score changes, improving the pertinence of teaching plan formulation and student learning effectiveness.

[0141] As an optional embodiment, performance influencing factors include at least one of weak subjects, weak time points, weak user characteristics, combinations of weak subjects, and combinations of weak time points.

[0142] As can be seen, the above optional embodiments define the content of factors influencing academic performance, effectively characterizing the characteristics of students' academic performance, assisting in the accurate analysis of factors influencing academic performance based on historical performance trends and the formulation of personalized teaching plans, thereby improving students' learning outcomes and the efficiency of educational resource utilization.

[0143] As an optional embodiment, the generation module generates a teaching plan for the target student based on the LLM algorithm model, according to the factors influencing performance and the performance change curve. The specific methods include:

[0144] Determine the historical time points corresponding to the curve points in each performance change curve where the rate of change is greater than a preset rate of change threshold, and obtain the set of abrupt change time points corresponding to each performance change curve.

[0145] Calculate the intersection of the sets of abrupt change time points corresponding to all performance change curves to obtain multiple abrupt change time points corresponding to the target student;

[0146] Based on the learning plan table corresponding to the target students, determine the learning plan to be improved for each mutation time point;

[0147] Based on the LLM algorithm model, a teaching plan corresponding to the target student is generated according to the factors affecting performance and all learning plans that need improvement.

[0148] As can be seen, through the above optional embodiments, by identifying the historical time points corresponding to the curve points in each performance change curve where the rate of change exceeds a preset threshold, a set of abrupt change time points is formed. The intersection of all curve sets is calculated to obtain the abrupt change time points of the target student. Combined with the learning plan table, the corresponding learning plan to be improved is determined. Based on the LLM algorithm model, a teaching scheme is generated using performance influencing factors and the plan to be improved. This achieves precise optimization of the teaching scheme based on performance abrupt changes and learning plans, thereby improving student learning outcomes and the relevance of educational resources.

[0149] As an optional implementation, the generation module, based on the LLM algorithm model, generates a specific method for producing a teaching plan corresponding to the target student based on factors influencing performance and all learning plans to be improved, including:

[0150] Based on preset prompt text rules, and according to factors affecting performance and all learning plans to be improved, prompts are generated for the target student.

[0151] The prompt words generated by the solution are input into the trained LLM algorithm model to obtain the teaching solution corresponding to the target student; the LLM algorithm model is trained on a training dataset that includes multiple training teaching solutions and corresponding labels of performance influencing factors and improvement plans.

[0152] As can be seen, through the above optional embodiments, prompt words are generated based on preset prompt word text rules, combined with factors affecting performance and the learning plan generation scheme to be improved. These prompt words are then input into a trained LLM algorithm model to generate a teaching plan for the target student. This enables the development of precise and personalized teaching plans based on influencing factors and learning plans, thereby improving student learning outcomes and the pertinence of educational guidance.

[0153] Example 3

[0154] Please see Figure 3 , Figure 3 This is another teaching plan generation system based on the analysis of changes in student performance disclosed in the embodiments of the present invention. Figure 3 The described teaching plan generation system based on student performance change analysis is applied in a data processing system / data processing equipment / data processing server (wherein, the server includes a local processing server or a cloud processing server). Figure 3 As shown, the teaching plan generation system based on student performance change analysis may include:

[0155] Memory 301 storing executable program code;

[0156] Processor 302 coupled to memory 301;

[0157] The processor 302 calls the executable program code stored in the memory 301 to execute the steps of the teaching plan generation method based on student performance change analysis described in Embodiment 1.

[0158] Example 4

[0159] This invention discloses a computer read storage medium that stores a computer program for electronic data interchange, wherein the computer program causes a computer to execute the steps of the teaching plan generation method based on student performance change analysis described in Embodiment 1.

[0160] Example 5

[0161] This invention discloses a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to cause a computer to perform the steps of the teaching plan generation method based on student performance change analysis described in Embodiment 1.

[0162] The foregoing has described specific embodiments of this specification; other embodiments are within the scope of the appended claims. In some cases, the actions or steps described in the claims may be performed in a different order than those shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily have to follow the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0163] The systems, devices, modules, or units described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, a computer can be, for example, a personal computer, laptop computer, cellular phone, camera phone, smartphone, personal digital assistant, media player, navigation device, email device, game console, tablet computer, wearable device, or any combination of these devices.

[0164] For ease of description, the above devices are described in terms of function, divided into various units. Of course, in implementing this specification, the functions of each unit can be implemented in one or more software and / or hardware components.

[0165] Those skilled in the art will understand that the embodiments of this specification can be provided as methods, systems, or computer program products. Therefore, the embodiments of this specification can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the embodiments of this specification can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0166] This specification is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this specification. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create a machine for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0167] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0168] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0169] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0170] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0171] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0172] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0173] This specification can be described in the general context of computer-executable instructions that are executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform a specific task or implement a specific abstract data type. This specification can also be practiced in distributed computing environments, where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0174] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to interchangeably. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments.

[0175] Finally, it should be noted that the teaching plan generation method and system based on student performance change analysis disclosed in the embodiments of the present invention are merely preferred embodiments of the present invention, and are only used to illustrate the technical solutions of the present invention, not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for generating teaching plans based on student performance change analysis, characterized in that, The method includes: Obtain historical score data for the target student at multiple historical points in time; Based on the historical performance data and the corresponding historical time points, multiple performance change curves for the target student are constructed, including: For each subject, the corresponding historical score data are used as the dependent variable, and the corresponding historical time point is used as the independent variable. Based on the mathematical fitting algorithm model, the dependent variable and the independent variable are fitted to obtain the mathematical relationship model of the performance change corresponding to the subject. Based on the drawing algorithm, according to the mathematical relationship model of the performance change, the performance change curve corresponding to each subject is drawn on the preset canvas to obtain multiple performance change curves corresponding to the target student. Based on graph neural networks, and according to the multiple performance change curves, the performance influencing factors for the target student are predicted, including: Each historical time point is taken as a graph node, and the historical score data of all subjects corresponding to that historical time point are determined as the node feature value corresponding to that graph node. For any two graph nodes, the weighted average of the differences between the curve values ​​corresponding to the two graph nodes is determined in each performance change curve to obtain the adjacency feature value between the two graph nodes. Determine whether the adjacency feature value is greater than the feature value threshold. If yes, establish a node connection between the two graph nodes and use the adjacency feature value as the connection feature value. If no, do not establish a node connection between the two graph nodes. All the graph nodes and their corresponding node feature values, as well as the node connections between the graph nodes and their corresponding connection feature values, are determined as the performance graph feature network corresponding to the target student. The performance graph feature network is input into a trained graph neural network to obtain the performance influencing factors corresponding to the target student; the graph neural network is trained using a training dataset that includes multiple trained performance graph feature networks and corresponding performance influencing factor annotations. Based on the factors influencing the performance and the performance change curve, a teaching plan corresponding to the target student is generated using the LLM algorithm model.

2. The teaching plan generation method based on student performance change analysis according to claim 1, characterized in that, The mathematical model for the changes in performance is a polynomial regression model with a highest order greater than two and less than five.

3. The teaching plan generation method based on student performance change analysis according to claim 2, characterized in that, The step of using multiple historical score data corresponding to the subject as dependent variables and the corresponding historical time points as independent variables includes: Sort all the historical score data corresponding to the subject from earliest to latest according to the corresponding historical time points to obtain the score data sequence; The performance data sequence is input into a trained anomaly prediction neural network to obtain the historical performance data that belongs to the anomalous data in the performance data sequence; the anomaly prediction neural network is trained on a training dataset that includes multiple training performance data sequences and corresponding data labels that belong to the anomalous performances. All historical performance data that are considered abnormal data are removed, and the remaining historical performance data is used as the dependent variable, with the corresponding historical time point as the independent variable.

4. The teaching plan generation method based on student performance change analysis according to claim 1, characterized in that, The factors affecting performance include at least one of the following: weak subjects, weak time points, weak user characteristics, combinations of weak subjects, and combinations of weak time points.

5. The teaching plan generation method based on student performance change analysis according to claim 1, characterized in that, The step of generating a teaching plan for the target student based on the performance influencing factors and the performance change curve, using an LLM algorithm model, includes: Determine the historical time points corresponding to the curve points in each performance change curve where the rate of change is greater than a preset rate of change threshold, and obtain the set of abrupt change time points corresponding to each performance change curve. Calculate the intersection of the sets of abrupt change time points corresponding to all the aforementioned performance change curves to obtain multiple abrupt change time points corresponding to the target student; Based on the learning plan table corresponding to the target students, determine the learning plan to be improved for each of the abrupt change time points. Based on the LLM algorithm model, a teaching plan corresponding to the target student is generated according to the aforementioned performance influencing factors and all the aforementioned learning plans to be improved.

6. The teaching plan generation method based on student performance change analysis according to claim 5, characterized in that, The LLM algorithm model generates a teaching plan for the target student based on the performance influencing factors and all the learning plans to be improved, including: Based on preset prompt word text rules, and according to the performance influencing factors and all the learning plans to be improved, prompt words for the corresponding solutions for the target student are generated. The prompt words generated by the scheme are input into the trained LLM algorithm model to obtain the teaching scheme corresponding to the target student; the LLM algorithm model is trained using a training dataset that includes multiple training teaching schemes and corresponding labels of performance influencing factors and labels of plans to be improved.

7. A teaching plan generation system based on student performance change analysis, characterized in that, The system includes: The acquisition module is used to acquire multiple historical score data of a target student at multiple historical time points; The construction module is used to construct multiple performance change curves for the target student based on the historical performance data and the corresponding historical time points, including: For each subject, the corresponding historical score data are used as the dependent variable, and the corresponding historical time point is used as the independent variable. Based on the mathematical fitting algorithm model, the dependent variable and the independent variable are fitted to obtain the mathematical relationship model of the performance change corresponding to the subject. Based on the drawing algorithm, according to the mathematical relationship model of the performance change, the performance change curve corresponding to each subject is drawn on the preset canvas to obtain multiple performance change curves corresponding to the target student. The prediction module, based on a graph neural network, predicts the performance influencing factors for the target student according to the multiple performance change curves, including: Each historical time point is taken as a graph node, and the historical score data of all subjects corresponding to that historical time point are determined as the node feature value corresponding to that graph node. For any two graph nodes, the weighted average of the differences between the curve values ​​corresponding to the two graph nodes is determined in each performance change curve to obtain the adjacency feature value between the two graph nodes. Determine whether the adjacency feature value is greater than the feature value threshold. If yes, establish a node connection between the two graph nodes and use the adjacency feature value as the connection feature value. If no, do not establish a node connection between the two graph nodes. All the graph nodes and their corresponding node feature values, as well as the node connections between the graph nodes and their corresponding connection feature values, are determined as the performance graph feature network corresponding to the target student. The performance graph feature network is input into a trained graph neural network to obtain the performance influencing factors corresponding to the target student; the graph neural network is trained using a training dataset that includes multiple trained performance graph feature networks and corresponding performance influencing factor annotations. The generation module is used to generate a teaching plan for the target student based on the influencing factors of the performance and the performance change curve, using the LLM algorithm model.

8. A teaching plan generation system based on student performance change analysis, characterized in that, The system includes: Memory containing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the teaching plan generation method based on student performance change analysis as described in any one of claims 1-6.

Images