Knowledge point mastery degree updating method and device, equipment and readable storage medium

Abstract

The invention provides a knowledge point mastery degree updating method, device and equipment and a readable storage medium, a constructed knowledge graph is utilized to represent a dependency relationship between knowledge points, knowledge points related to exercise answering of a user are determined, the mastery degree of the knowledge points is updated according to the correct or wrong answering of the user, and the user experience is improved. When the answer is correct, on the basis of the mastery degree of the current knowledge point, propagation updating is conducted on the mastery degree of the preorder knowledge point in the preorder direction, and when the answer is wrong, propagation updating is conducted on the mastery degree of the post-order knowledge point in the post-order direction on the basis of the mastery degree of the current knowledge point; according to the method, the logic relation between the knowledge points and the answering condition of the user are fully considered, and the mutual influence between the knowledge points can be reflected more accurately, so that the knowledge mastery degree of the student is accurately and dynamically tracked according to the answering behavior of the user in the learning process of the user.

Description

Technical Field

[0001] This application relates to the field of educational technology, and in particular, to a method, apparatus, device, and readable storage medium for updating the degree of knowledge point mastery. Background Art

[0002] In current methods for evaluating students' knowledge mastery, the mastery of students is usually modeled in a static or coarse-grained manner, and the mastery of each knowledge point is viewed in isolation, lacking the ability to accurately and dynamically track the learning process. As a result, it is difficult to dynamically and accurately reflect the changing trend of students' mastery of knowledge points based on users' real-time learning behaviors, thereby affecting the accuracy of personalized learning path recommendation and prediction of future learning performance. Summary of the Invention

[0003] In view of this, to solve the above technical problems, this application provides a method, apparatus, device, and readable storage medium for updating the degree of knowledge point mastery.

[0004] Specifically, this application is implemented through the following technical solutions: According to the first aspect of the embodiments of this application, a method for updating the degree of knowledge point mastery is provided. The method includes: Determine each knowledge point involved in the exercise completed by the user in the constructed knowledge graph, and update the mastery degree of the knowledge point based on the user's answer situation; each node in the knowledge graph represents a corresponding knowledge point, and the directed edge between two nodes in the knowledge graph indicates the dependency relationship between the two nodes, where the starting node of the directed edge is the pre-order node of the ending node of the directed edge, and the ending node of the directed edge is the post-order node of the starting node of the directed edge; When the user's answer situation indicates a correct answer, for each knowledge point involved in the exercise, take the knowledge point as the current knowledge point, and based on the mastery degree of the current knowledge point, update the mastery degree of the pre-order knowledge point of the current knowledge point in the knowledge graph. When the first iteration termination condition is not currently satisfied, take the pre-order knowledge point of the current knowledge point as the current knowledge point, and return to update the mastery degree of the pre-order knowledge point of the current knowledge point in the knowledge graph based on the mastery degree of the current knowledge point; When the user's answer situation indicates an incorrect answer, for each knowledge point involved in the exercise, take the knowledge point as the current knowledge point, and based on the mastery degree of the current knowledge point, update the mastery degree of the post-order knowledge point of the current knowledge point in the knowledge graph. When the second iteration termination condition is not currently satisfied, take the post-order knowledge point of the current knowledge point as the current knowledge point, and return to update the mastery degree of the pre-order knowledge point of the current knowledge point in the knowledge graph based on the mastery degree of the current knowledge point.

[0005] Optionally, updating the mastery level of the knowledge point based on the user's answer situation includes: When the user's answer situation indicates a correct answer, the adjustment amplitude of the first confidence factor corresponding to the user's answer result is corrected according to the duration for which the confidence factor of the knowledge point persists, and the confidence factor of the knowledge point is updated based on the corrected adjustment amplitude of the first confidence factor; wherein, the longer the duration, the smaller the absolute value of the corrected adjustment amplitude of the first confidence factor; When the user's answer situation indicates a wrong answer, the adjustment amplitude of the second confidence factor corresponding to the user's answer result is corrected according to the duration for which the confidence factor of the knowledge point persists, and the confidence factor of the knowledge point is updated based on the corrected adjustment amplitude of the second confidence factor; wherein, the longer the duration, the smaller the absolute value of the corrected adjustment amplitude of the second confidence factor; Adjust the mastery level of the knowledge point in the knowledge graph based on the updated confidence factor.

[0006] Optionally, adjusting the mastery level of the knowledge point in the knowledge graph based on the updated confidence factor includes: Perform weighted adjustment on the adjustment amplitude of the knowledge point based on the updated confidence factor to obtain the adjustment amplitude of the mastery level of the knowledge point, and update the mastery level of the knowledge point using the adjustment amplitude of the mastery level.

[0007] Optionally, the adjustment amplitude of the knowledge point is determined by the following method: Based on the difficulty of the exercise and the mastery level of the knowledge point, predict the user's answering situation when answering the exercise; Based on the error between the user's answer situation obtained after the user answers the exercise and the predicted answering situation, determine the adjustment amplitude of the knowledge point.

[0008] Optionally, when the user's answer situation indicates a correct answer, updating the mastery level of the previous knowledge point of the current knowledge point in the knowledge graph based on the mastery level of the current knowledge point includes: If none of the exercises associated with the previous knowledge point have been answered by the user and the mastery level of the previous knowledge point is the initially set value, then the previous knowledge point inherits the mastery level of the current knowledge point as the updated mastery level of the previous knowledge point; If the mastery level of the previous knowledge point is less than the mastery level of the current knowledge point, then perform weighted processing on the mastery level of the previous knowledge point and the mastery level of the current knowledge point to determine the updated mastery level of the previous knowledge point; If the mastery level of the previous knowledge point is greater than or equal to the mastery level of the current knowledge point, then keep the mastery level of the previous knowledge point unchanged.

[0009] Optionally, when the user's answer indicates an incorrect answer, based on the mastery level of the current knowledge point, update the mastery levels of the subsequent knowledge points of the current knowledge point in the knowledge graph, including: If none of the exercises associated with the subsequent knowledge point have been answered by the user and the mastery level of the subsequent knowledge point is the initially set value, then the subsequent knowledge point inherits the mastery level of the current knowledge point as the updated mastery level of the subsequent knowledge point; If the mastery level of the subsequent knowledge point is greater than the mastery level of the current knowledge point, then perform a weighted processing on the mastery level of the subsequent knowledge point and the mastery level of the current knowledge point to determine the updated mastery level of the subsequent knowledge point; If the mastery level of the subsequent knowledge point is less than or equal to the mastery level of the current knowledge point, then keep the mastery level of the subsequent knowledge point unchanged.

[0010] Optionally, update the mastery level of the target knowledge point of the current knowledge point based on the mastery level of the current knowledge point, where the target knowledge point includes any one of the previous knowledge point or the subsequent knowledge point, including: Detect whether the mastery level update operation has been performed on the target knowledge point after the user has completed answering the exercise; If so, skip the current mastery level update; If not, then update the mastery level of the target knowledge point of the current knowledge point based on the mastery level of the current knowledge point, and after the mastery level of the target knowledge point is updated, mark that the mastery level update operation has been performed on the target knowledge point after the user has completed answering the exercise.

[0011] Optionally, the knowledge graph is constructed in the following manner: Use a large model to determine the dependency relationship between the two knowledge points in each knowledge point pair in the knowledge point set, and use the large model to add the chapter information of the textbook tree to each knowledge point; Construct a knowledge graph based on the dependency relationship of each knowledge point pair, and for the isolated knowledge points in the knowledge graph, use the chapter information of the textbook tree of the isolated knowledge point and the chapter sequence of the textbook tree to connect the isolated knowledge point to the knowledge graph; Adjust the node connection relationship in the knowledge graph to convert the knowledge graph into a directed acyclic graph, and obtain the constructed knowledge graph.

[0012] According to the second aspect of the embodiments of the present application, there is provided a device for updating the mastery level of knowledge points, the device includes: The exercise-related knowledge point mastery update module is used to determine each knowledge point involved in the exercise completed by the user in the constructed knowledge graph, and update the mastery degree of the knowledge point based on the user's answer situation; each node in the knowledge graph represents a corresponding knowledge point, and the directed edge between two nodes in the knowledge graph indicates the dependency relationship between the two nodes, where the starting node of the directed edge is the pre-order node of the ending node of the directed edge, and the ending node of the directed edge is the post-order node of the starting node of the directed edge; The pre-order knowledge point mastery update module is used to, when the user's answer situation indicates a correct answer, for each knowledge point involved in the exercise, take this knowledge point as the current knowledge point, and based on the mastery degree of the current knowledge point, update the mastery degree of the pre-order knowledge point of the current knowledge point in the knowledge graph, and when the first iteration termination condition is not currently met, take the pre-order knowledge point of the current knowledge point as the current knowledge point, and return the update of the mastery degree of the pre-order knowledge point of the current knowledge point in the knowledge graph based on the mastery degree of the current knowledge point; The post-order knowledge point mastery update module is used to, when the user's answer situation indicates an incorrect answer, for each knowledge point involved in the exercise, take this knowledge point as the current knowledge point, and based on the mastery degree of the current knowledge point, update the mastery degree of the post-order knowledge point of the current knowledge point in the knowledge graph, and when the second iteration termination condition is not currently met, take the post-order knowledge point of the current knowledge point as the current knowledge point, and return the update of the mastery degree of the pre-order knowledge point of the current knowledge point in the knowledge graph based on the mastery degree of the current knowledge point.

[0013] Optionally, the exercise-related knowledge point mastery update module includes: The first correction module is used to, when the user's answer situation indicates a correct answer, correct the first confidence factor adjustment amplitude corresponding to the user's answer result according to the duration for which the confidence factor of the knowledge point persists, and update the confidence factor of the knowledge point based on the corrected first confidence factor adjustment amplitude; where the longer the duration, the smaller the corrected first confidence factor adjustment amplitude, and the smaller the absolute value of the corrected first confidence factor adjustment amplitude; The second correction module is used to, when the user's answer situation indicates an incorrect answer, correct the second confidence factor adjustment amplitude corresponding to the user's answer result according to the duration for which the confidence factor of the knowledge point persists, and update the confidence factor of the knowledge point based on the corrected second confidence factor adjustment amplitude; where the longer the duration, the smaller the corrected second confidence factor adjustment amplitude, and the smaller the absolute value of the corrected second confidence factor adjustment amplitude; The update module is used to adjust the mastery degree of the knowledge point in the knowledge graph based on the updated confidence factor.

[0014] Optionally, the updating module is specifically configured to: Based on the updated confidence factor, weight and adjust the adjustment amplitude of the knowledge point to obtain the adjusted amplitude of the mastery degree of the knowledge point, and use the adjusted amplitude of the mastery degree to update the mastery degree of the knowledge point.

[0015] Optionally, the adjustment amplitude of the knowledge point is determined by the following method: Based on the difficulty of the exercise and the mastery degree of the knowledge point, predict the answering situation of the user when answering the exercise; Based on the error between the user's answering situation obtained after the user has answered the exercise and the predicted answering situation, determine the adjustment amplitude of the knowledge point.

[0016] Optionally, the previous knowledge point mastery degree updating module is specifically configured to: If none of the exercises associated with the previous knowledge point have been answered by the user and the mastery degree of the previous knowledge point is the initially set value, then the previous knowledge point inherits the mastery degree of the current knowledge point as the updated mastery degree of the previous knowledge point; If the mastery degree of the previous knowledge point is less than the mastery degree of the current knowledge point, then perform a weighted process on the mastery degree of the previous knowledge point and the mastery degree of the current knowledge point to determine the updated mastery degree of the previous knowledge point; If the mastery degree of the previous knowledge point is greater than or equal to the mastery degree of the current knowledge point, then keep the mastery degree of the previous knowledge point unchanged.

[0017] Optionally, the subsequent knowledge point mastery degree updating module is specifically configured to: If none of the exercises associated with the subsequent knowledge point have been answered by the user and the mastery degree of the subsequent knowledge point is the initially set value, then the subsequent knowledge point inherits the mastery degree of the current knowledge point as the updated mastery degree of the subsequent knowledge point; If the mastery degree of the subsequent knowledge point is greater than the mastery degree of the current knowledge point, then perform a weighted process on the mastery degree of the subsequent knowledge point and the mastery degree of the current knowledge point to determine the updated mastery degree of the subsequent knowledge point; If the mastery degree of the subsequent knowledge point is less than or equal to the mastery degree of the current knowledge point, then keep the mastery degree of the subsequent knowledge point unchanged.

[0018] Optionally, when the previous knowledge point / subsequent knowledge point mastery degree updating module is used to update the mastery degree of the target knowledge point of the current knowledge point based on the mastery degree of the current knowledge point, the target knowledge point includes any one of the previous knowledge point or the subsequent knowledge point, and includes: Detect whether the target knowledge point has performed a mastery degree update operation after the exercise has been answered by the user; If so, skip the current mastery update; If not, update the mastery of the target knowledge point of the current knowledge point based on the mastery of the current knowledge point, and after the mastery of the target knowledge point is updated, mark that the mastery update operation has been performed on the target knowledge point after the exercise is answered by the user.

[0019] Optionally, the knowledge graph is constructed in the following manner: Use a large model to determine the dependency relationship of each knowledge point pair in the knowledge point set by judging the front and back order of the two knowledge points in each knowledge point pair, and use the large model to add the chapter information of the textbook tree to each knowledge point; Construct a knowledge graph based on the dependency relationship of each knowledge point pair, and for the isolated knowledge points in the knowledge graph, use the chapter information of the textbook tree of the isolated knowledge point and the chapter sequence of the textbook tree to connect the isolated knowledge point to the knowledge graph; Adjust the node connection relationship in the knowledge graph to convert the knowledge graph into a directed acyclic graph, and obtain the constructed knowledge graph.

[0020] According to the third aspect of the embodiments of the present application, an electronic device is provided, and the electronic device includes: a memory and a processor; the memory is used to store a computer program; the processor is used to execute the above-mentioned knowledge point mastery update method by calling the computer program.

[0021] According to the fourth aspect of the embodiments of the present application, a computer-readable storage medium is provided, on which a computer program is stored, and when the program is executed by a processor, the above-mentioned knowledge point mastery update method is implemented.

[0022] The technical solutions provided by the embodiments of the present application may include the following beneficial effects: In the above technical solutions provided by the present application, by determining the knowledge points involved in the exercises answered by the user in the already constructed knowledge graph and updating the mastery of these knowledge points according to the answering situation, when the answer is correct, update its pre-order knowledge points based on the current knowledge point mastery and perform the propagation update of the knowledge point mastery along the pre-order direction, and when the answer is wrong, update its post-order knowledge points based on the current knowledge point mastery and perform the propagation update of the knowledge point mastery along the post-order direction, so as to accurately and dynamically track the student's knowledge mastery according to the user's answering behavior of the exercises, make full use of the knowledge structure information and the pre-order / post-order dependency relationship between knowledge points to realize the propagation update of the knowledge point mastery in the specified dependency direction, and effectively improve the accuracy of predicting the student's future performance.

[0023] It should be understood that the above general description and the following detailed description are merely exemplary and explanatory, and do not limit this application. In addition, any embodiment in this application does not need to achieve all the above effects. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The accompanying drawings herein are incorporated into and constitute a part of this specification, showing embodiments consistent with this application, and are used together with the specification to explain the principles of this application.

[0025] Figure 1A is a flowchart showing the steps of a method for updating the mastery level of knowledge points shown in an exemplary embodiment of this application; Figure 1B is a partial schematic diagram of a knowledge graph of the mathematics subject shown in an exemplary embodiment of this application; Figure 2 is a flowchart showing the steps of updating the mastery level of knowledge points involved in an exercise based on a confidence factor shown in an exemplary embodiment of this application; Figure 3 is a flowchart showing the update of the mastery level of the preceding and subsequent knowledge points of the current knowledge point shown in an exemplary embodiment of this application; Figure 4 is a schematic diagram showing the process of constructing a knowledge graph shown in an exemplary embodiment of this application; Figure 5 is a schematic diagram showing the structure of a device for updating the mastery level of knowledge points shown in an exemplary embodiment of this application; Figure 6 is a schematic diagram of the hardware of an electronic device shown in an exemplary embodiment of this application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0026] Exemplary embodiments will be described in detail herein, and examples thereof are shown in the accompanying drawings. When the following description refers to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. On the contrary, they are merely examples of devices and methods consistent with some aspects of this application as detailed in the appended claims. It should be understood that although terms such as first, second, and third may be used in this application to describe various information, such information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other.

[0027] In the current methods for evaluating students' knowledge mastery, static or coarse-grained approaches are generally adopted to model students' learning states. Usually, each knowledge point is regarded in isolation, without fully considering the relevance and dependency relationships between knowledge points, and lacking the ability to conduct refined dynamic tracking of the learning process. Such methods are difficult to dynamically and accurately reflect the changing trend of students' mastery of knowledge points based on users' real-time learning behaviors, thereby affecting the accuracy of personalized learning path recommendations and predictions of future learning performance.

[0028] In view of this, the present application provides a method for updating the mastery degree of knowledge points, which combines the knowledge graph structure, dynamically updates the mastery degree of knowledge points, and conducts an evaluation of the propagation of the mastery degree to previous or subsequent knowledge points based on the user's answering situation, so as to improve the accuracy and rationality of mastery degree modeling and achieve refined and dynamic evaluation of students' knowledge mastery status.

[0029] See Figure 1A The flowchart of the steps of an exemplary method for updating the mastery degree of knowledge points is shown. The method for updating the mastery degree of knowledge points recorded in the present application may at least include the following steps: S101, determine each knowledge point involved in the exercise completed by the user in the constructed knowledge graph, and update the mastery degree of the knowledge point based on the user's answering situation; each node in the knowledge graph represents a corresponding knowledge point, and the directed edge between two nodes in the knowledge graph indicates the dependency relationship between the two nodes. Among them, the starting node of the directed edge is the previous knowledge point of the ending node of the directed edge, and the ending node of the directed edge is the subsequent knowledge point of the starting node of the directed edge; The knowledge graph stores and organizes knowledge in the form of a graph, consisting of nodes and edges. Among them, nodes (also called vertices) represent entities, and edges (also called arcs) represent the relationships between entities. In this embodiment, the knowledge graph is used to represent the dependency relationship between knowledge points, each node is used to represent a specific knowledge point, and the edge connecting any two nodes is a directed edge (a line segment with directionality connecting two nodes), which is used to represent the dependency relationship between two knowledge points. The knowledge point represented by the previous knowledge point is the basic knowledge that needs to be mastered before learning the knowledge point represented by the subsequent knowledge point, and the knowledge point represented by the subsequent knowledge point is the content learned further on the basis of the previous knowledge point.

[0030] The starting node of the directed edge is the previous knowledge point of the ending node of the directed edge, indicating that the knowledge point represented by the starting node of the directed edge is the previous knowledge point of the knowledge point represented by the ending node of the directed edge. For example, for nodes A - B, it means that the knowledge point represented by node A, such as "the concept of linear equations with one variable", is the previous knowledge point of the knowledge point represented by node B, such as "the solution method of linear equations with one variable". In the learning process, it is usually necessary to master the knowledge point represented by node A first in order to better understand the knowledge point represented by node B.

[0031] Conversely, the end node of a directed edge is the post-order node of the start node of the directed edge, indicating that the knowledge point represented by the end node of the directed edge is the post-order knowledge point of the knowledge point represented by the start node of the directed edge. For example, for nodes C-D, it means that the knowledge point represented by node D, such as "mixed arithmetic operations", is the post-order knowledge point of the knowledge point represented by node C, such as "mixed multiplication and division operations". In the learning process, to master the knowledge point represented by node D, it is usually necessary to first master the knowledge point represented by node C.

[0032] See <{ Figure 1B A partial schematic diagram of a knowledge graph of the mathematics discipline is exemplarily shown. Any two knowledge points in this partial knowledge graph are connected by a directed edge, and the direction of the directed edge indicates the order of the two knowledge points. Taking the node ⑦ representing the knowledge point "rational number mixed arithmetic operations" in the figure as an example, the directed edges with this node as the start node or the end node include edges L1, L2, and L3: Edge L1 has this node as the end node, so the pre-order knowledge points of the knowledge point "rational number mixed arithmetic operations" represented by this node include: the knowledge point "rational number mixed multiplication and division operations" represented by the start node ⑥ of L1; Edges L2 and L3 have this node as the start node, so the post-order knowledge points of the knowledge point "rational number mixed arithmetic operations" represented by this node include: the knowledge point "rational number operations related to the number line" represented by the end node ⑧ of L2, and the knowledge point "rational number multiplication and division and practical problems" represented by the end node ⑨ of L3.

[0033] The mastery degree of a knowledge point is used to quantitatively describe the user's understanding and mastery of the knowledge point. This mastery degree can be represented by a numerical value. The higher the value, the better the user's mastery of the knowledge point. For example, the mastery degree of a knowledge point can be represented by a numerical value in the range of [0, 1]. When the mastery degree of the knowledge point reaches 1, it means that the user has mastered this knowledge point proficiently.

[0034] After constructing the knowledge graph in this application, an initial mastery degree can be assigned to the knowledge point represented by each node in the knowledge graph by using the personalized knowledge mastery degree model established for each user in advance, so as to represent the user's own understanding and mastery of the knowledge point. That is to say, there are differences in the mastery degrees of the same knowledge point for different users, and the setting of the mastery degree of the knowledge point depends on the user's own learning and practice situations for initialization and update.

[0035] Regarding determining each knowledge point involved in the exercises completed by the user in the constructed knowledge graph, when the knowledge points involved in each exercise are stored in advance for each exercise, each knowledge point involved in the stored exercise can be obtained based on the unique identifier or other information of the exercise completed by the user, and through the correspondence between knowledge points and nodes in the constructed knowledge graph, each node in the knowledge graph representing each knowledge point involved in the exercise can be determined. For the case where the knowledge points involved in the exercise are not stored, it is also possible to analyze the text content of the exercise completed by the user through a trained model such as a large language model, a deep neural network, or natural language processing technology, extract the key text information and match it with the knowledge points represented by each node in the knowledge graph, so as to determine the knowledge points involved in the exercise, and determine each node in the knowledge graph representing each knowledge point involved in the exercise.

[0036] Regarding updating the mastery level of this knowledge point based on the user's answering situation, it should be clear that in this embodiment, the update of the mastery level of the same knowledge point by different users depends on their respective exercise practice situations. That is, for user 1 and user 2 who both complete exercise M, as two independent individuals, the mastery level of each knowledge point involved in exercise M is updated separately with the user as the index. Specifically, based on user 1's answering situation for exercise M, what is updated is user 1's mastery level of the knowledge points involved in this exercise; based on user 2's answering situation for exercise M, what is updated is user 2's mastery level of these knowledge points.

[0037] When the user's answering situation indicates a correct answer, it is considered that the user's understanding of each knowledge point involved in the exercise has been enhanced, and the mastery level of each knowledge point involved in the exercise is enhanced; when the user's answering situation indicates a wrong answer, it is considered that there may be problems with the user's understanding of the knowledge points involved in the exercise, and the mastery level of the knowledge points involved in the exercise is reduced. Among them, the user's answering situation can be quantified numerically and used as the basis for determining the amplitude of the mastery level update when updating the mastery level of the knowledge point.

[0038] S102, when the user's answering situation indicates a correct answer, for each knowledge point involved in the exercise, take this knowledge point as the current knowledge point, and based on the mastery level of the current knowledge point, update the mastery level of the previous knowledge point of the current knowledge point in the knowledge graph, and when the first iteration termination condition is not currently met, take the previous knowledge point of the current knowledge point as the current knowledge point, and return to update the mastery level of the previous knowledge point of the current knowledge point in the knowledge graph based on the mastery level of the current knowledge point; The first iteration termination condition is used to determine whether the iterative process of updating the mastery degree of the previous knowledge points of the current knowledge point has ended. In this step, when any one of the conditions in the first iteration termination condition is satisfied, the iterative process stops, that is, the mastery degree is no longer updated forward to the previous knowledge points. The first iteration termination condition can be set according to actual needs.

[0039] Optionally, the first iteration termination condition may at least include that the previous knowledge points of the current knowledge point no longer have previous knowledge points; in order to avoid the problem that the range of forward mastery degree propagation update based on a knowledge point involved in an exercise is too large, resulting in insufficient accuracy of the mastery degree of irrelevant knowledge points, the range of mastery degree update can be controlled by setting the iteration termination condition, that is, the first iteration termination condition may also include at least one of the following: the path length between the previous knowledge point of the current knowledge point and the knowledge point involved in the exercise used at the start of this iteration in the knowledge graph exceeds a set length threshold. By restricting this length, it can prevent the update process from expanding to areas that are too far from the knowledge point at the start of the iteration, thus ensuring the effectiveness and pertinence of the update; or, the knowledge relevance degree between the previous knowledge point of the current knowledge point and the knowledge point involved in the exercise used at the start of this iteration is lower than a set relevance threshold, ensuring that only the knowledge points that are truly closely related to the starting knowledge point of the iteration are included in the range of mastery degree update, improving the accuracy and efficiency of the update process. In practical applications, other conditions can also be used as the iteration termination condition, and this application does not limit this.

[0040] Among them, regarding the knowledge point involved in the exercise used at the start of this iteration, based on this step, it is executed separately for each knowledge point involved in the exercise, that is, each knowledge point involved in the exercise is used as the current knowledge point once, and the mastery degree propagation update in the forward direction starts with this current knowledge point as the starting point of an iteration. Therefore, when the user's answer indicates a correct answer, for each knowledge point involved in the exercise, taking this knowledge point as the current knowledge point, if the knowledge point involved in the exercise that is used as the current knowledge point at this time is used as the starting point of the iteration, then the knowledge point involved in the exercise used at the start of this iteration represents this starting point of the iteration.

[0041] For example, Figure 1B For example, assume that the knowledge points involved in the exercise include: "rational number multiplication and division mixed operation" represented by node ⑥, and "rational number multiplication and division and practical problems" represented by node ⑨. Then, when the knowledge point represented by node ⑥ is used as the current knowledge point to trigger the iterative process, the knowledge point represented by node ⑥ is used as the knowledge point involved in the exercise used at the start of this iteration. During the forward mastery degree propagation update process, the path lengths between the knowledge points represented by nodes ④, ③, ②, etc. and the knowledge point represented by node ⑥ are detected.

[0042] The path length in the knowledge graph between the predecessor knowledge point of the current knowledge point and the knowledge points involved in the exercises used at the beginning of this iteration can be represented by counting the number of directed edges passed through in the knowledge graph, starting from the predecessor knowledge point of the current knowledge point and following the directed edges to reach the knowledge points involved in the exercises used at the beginning of this iteration. The larger the number of directed edges, the longer the path length, and correspondingly, the smaller the impact of the user's answers to the exercises on the mastery of the current knowledge point.

[0043] The degree of knowledge relevance between the previous knowledge points of the current knowledge point and the knowledge points involved in the exercises used at the beginning of this iteration can be evaluated by using a large model or relying on the chapter information in the pre-constructed textbook tree to which the two knowledge points belong, or by using other specific algorithms to give a correlation value. The higher the correlation value, the closer the connection between the two knowledge points, and the greater the impact of the user's answer to the exercise on the mastery of the previous knowledge points of the current knowledge point. Conversely, the lower the correlation value, the sparser the connection between the two knowledge points, and the smaller the impact of the user's answer to the exercise on the mastery of the previous knowledge points of the current knowledge point. Among them, the chapter information of the textbook tree refers to the information formed after the textbook content is decomposed into chapters of different levels and orders according to the knowledge system structure of the textbook. Usually, textbooks arrange chapters according to a certain logic and order, such as from basic concepts to advanced applications, from simple to complex, etc. Each chapter contains specific knowledge points. By analyzing the position of two knowledge points in the textbook tree chapters, the hierarchical relationship, and the correlation of the chapter content, the degree of knowledge relevance between them can be determined. For example, knowledge points in the same chapter or adjacent chapters have higher correlation values, while knowledge points in chapters far apart have lower correlation values.

[0044] This implementation step is executed when the user answers the exercises correctly. It not only updates the mastery of each knowledge point involved in the exercises and its predecessor knowledge points, but also continues to use the predecessor knowledge point of each knowledge point involved in the exercises as the current knowledge point, and continues to update the mastery of multiple knowledge points along the predecessor association path in the direction of the predecessor node of the node in the knowledge graph.

[0045] by Figure 1B For example, when the user answers Exercise 1 correctly, if the knowledge point involved in the exercise is the knowledge point "Mixed Operations of Rational Number Multiplication and Division" represented by node ⑥, the first iteration termination condition is that the previous knowledge point of the current knowledge point no longer has a previous knowledge point. Then, after updating the mastery of the knowledge point represented by node ⑥: (1) Take the knowledge point represented by node ⑥ as the current knowledge point and update the mastery of its previous knowledge point (the knowledge point represented by node ④); (2) Use the prerequisite knowledge point of the current knowledge point (the knowledge point represented by node ④) as the current knowledge point, and update the mastery level of its prerequisite knowledge points (the knowledge points represented by nodes ③ and ②). Since the knowledge point represented by node ③ has no prerequisite knowledge point, terminate the propagation and update of the mastery level on the prerequisite association path for the knowledge point represented by node ③. (3) Continue to use the prerequisite knowledge point of the current knowledge point (the knowledge point represented by node ②) as the current knowledge point, and update the mastery level of its prerequisite knowledge point (the knowledge point represented by node ①). Since the knowledge point represented by node ① has no prerequisite knowledge point, terminate the propagation and update of the mastery level on the prerequisite association path for the knowledge point represented by node ①.

[0046] For each knowledge point involved in Exercise 1, use the same processing logic as the knowledge point "rational number division operation" represented by node ⑥ involved in it to complete a forward propagation and update of the mastery level starting from this knowledge point as the iteration start.

[0047] Among them, in the process of iteratively updating the mastery level of each knowledge point involved in the exercise along the forward direction, each time the mastery level of the prerequisite knowledge point of the current knowledge point is updated, the update of the mastery level of this prerequisite knowledge point depends on the mastery level of the current knowledge point, the current mastery level of the prerequisite knowledge point (that is, the mastery level that this prerequisite knowledge point already had before this mastery level update), and the relative magnitude relationship between the two.

[0048] S103, when the user's answer indicates an error, for each knowledge point involved in the exercise, use this knowledge point as the current knowledge point, and based on the mastery level of the current knowledge point, update the mastery level of the subsequent knowledge points of this current knowledge point in the knowledge graph. And when the second iteration termination condition is not currently satisfied, use the subsequent knowledge point of the current knowledge point as the current knowledge point, and return to update the mastery level of the prerequisite knowledge points of this current knowledge point in the knowledge graph based on the mastery level of the current knowledge point.

[0049] The second iteration termination condition is used to determine whether the iterative process of updating the mastery level of the subsequent knowledge points of the current knowledge point ends. In this step, when this second iteration termination condition is satisfied, the iterative process stops, that is, no longer continue to update the mastery level to the subsequent knowledge points. This second iteration termination condition can be set according to actual needs.

[0050] Optionally, the second iteration termination condition may at least include that the subsequent knowledge points of the current knowledge point no longer have subsequent knowledge points; in order to avoid inaccurate mastery evaluation of irrelevant knowledge points due to excessively large subsequent direction mastery propagation update range, the second iteration termination condition may also include at least one of the following: the path length in the knowledge graph between the subsequent knowledge points of the current knowledge point and the knowledge points involved in the exercises used at the beginning of this iteration exceeds a set length threshold; or, the knowledge relevance between the subsequent knowledge points of the current knowledge point and the knowledge points involved in the exercises used at the beginning of this iteration is lower than a set relevance threshold. In actual application, other conditions may also be used as iteration termination conditions, which are not limited in this application. Among them, the path length in the knowledge graph between the subsequent knowledge points of the current knowledge point and the knowledge points involved in the exercises used at the beginning of this iteration, as well as the knowledge relevance, refer to the aforementioned embodiment, and the principles of the two are the same, so this application will not repeat them again.

[0051] This implementation step is executed when the user answers the exercises incorrectly. It not only updates the mastery of each knowledge point involved in the exercises and its subsequent knowledge points, but also continues to use the subsequent knowledge points of each knowledge point involved in the exercises as the current knowledge points, and continues to update the mastery of multiple knowledge points in a linked manner along the subsequent association path in the direction of the subsequent nodes of the nodes in the knowledge graph.

[0052] by Figure 1B For example, when the user answers Exercise 2 incorrectly, if the knowledge point involved in Exercise 2 is the knowledge point "rational number division operation" represented by node ④, and the second iteration termination condition is "the subsequent knowledge point of the current knowledge point no longer has a subsequent knowledge point", then after updating the mastery of the knowledge point represented by node ④: (1) Take the knowledge point represented by node ④ as the current knowledge point and update the mastery of its subsequent knowledge point (the knowledge point represented by node ⑥); (2) Take the subsequent knowledge point of the current knowledge point (the knowledge point represented by node ⑥) as the current knowledge point, and update the mastery of its subsequent knowledge point (the knowledge point represented by node ⑦); (3) Take the subsequent knowledge point of the current knowledge point (the knowledge point represented by node ⑦) as the current knowledge point, and update the mastery of its subsequent knowledge points (the knowledge points represented by nodes ⑧ and ⑨); Since the knowledge points represented by nodes ⑧ and ⑨ do not have subsequent knowledge points, the mastery degree propagation update on the subsequent association path is terminated.

[0053] For each knowledge point involved in Exercise 2, the same processing logic as the knowledge point "rational number division operation" represented by the node ④ involved is used to complete a subsequent direction mastery propagation update starting with this knowledge point as the iteration starting point.

[0054] Among them, in the process of iteratively updating the mastery degree of each knowledge point involved in the exercise along the post-order direction, each time the mastery degree of the post-order knowledge point of the current knowledge point is updated, the update of the mastery degree of this post-order knowledge point depends on the mastery degree of the current knowledge point, the current mastery degree of this post-order knowledge point (that is, the mastery degree that this post-order knowledge point already had before this mastery degree update), and the relative magnitude relationship between the two.

[0055] Regarding the mastery degree of each knowledge point in the knowledge graph updated by using the knowledge point mastery degree update method provided in this embodiment, it can not only be used to accurately predict the performance of students in subsequent learning, but also can be used to construct a personalized knowledge state portrait for each user. Based on this portrait, the teaching system can perform targeted interventions, such as recommending exercise questions of appropriate difficulty or corresponding tutoring resources, forming a closed-loop personalized learning support to improve learning efficiency. In addition, the structural information of the knowledge graph can be used to plan a more scientific learning path for students by combining the prior and subsequent dependencies and the propagated and updated mastery degrees of knowledge points. When a student masters a certain prerequisite knowledge, the system can more reasonably recommend subsequent relevant knowledge; if a certain knowledge point is found to be weak, the system can guide the student to make up the prior knowledge first to ensure that the learning route conforms to the cognitive law.

[0056] In the embodiment of the present disclosure, the constructed knowledge graph is used to represent the dependency relationship between knowledge points. By determining the knowledge points involved in the user's answered exercises, the mastery degrees of these knowledge points are updated according to whether the user's answer is correct or incorrect. When the answer is correct, based on the current mastery degree of the knowledge point, the mastery degree of the prior knowledge point is propagated and updated along the prior direction. When the answer is incorrect, also based on the current mastery degree of the knowledge point, the mastery degree of the post-order knowledge point is propagated and updated along the post-order direction, realizing the dynamic adjustment of the mastery degrees of knowledge points in the knowledge graph in the specified dependency direction under different answering situations, which can timely reflect the change in the student's mastery of knowledge points, providing accurate and dynamic feedback on the student's knowledge mastery degree, thereby realizing the accurate and dynamic tracking of the student's knowledge mastery degree according to the user's answering behavior during the user's learning process. In addition, when updating the mastery degree of knowledge points in the knowledge graph, the logical relationship between knowledge points and the user's answering situation are fully considered. When the answer is correct, the mastery degree is propagated and updated to the prior knowledge point, and when the answer is incorrect, the mastery degree is propagated and updated to the post-order knowledge point, which can more accurately reflect the mutual influence between knowledge points, thereby improving the prediction accuracy of the student's future performance and providing a more personalized learning path recommendation for the student.

[0057] In some embodiments, for updating the mastery degree of the knowledge point based on the user's answer situation in step S101 of the foregoing embodiments, in order to improve the accuracy and stability of the mastery degree evaluation, avoid misjudging the user's knowledge mastery state due to accidental answering behaviors, and at the same time enhance the adaptability of the mastery degree update to long-term learning behaviors, in this embodiment, a confidence factor is set for each knowledge point represented by each node in the knowledge graph. The confidence factor is used to represent the reliability of the mastery degree evaluation. For each exercise associated with the same knowledge point, if multiple exercises are correctly answered by the same user, the confidence factor of the user on this knowledge point will be increased. That is to say, the confidence factor is a dynamically changing parameter. For each exercise completed by the user, before updating the mastery degree of each knowledge point involved in this exercise, the confidence factor of each knowledge point involved in this exercise will be updated based on the user's answer situation of the exercise. The confidence factor of any knowledge point is used to represent the confidence or certainty of the estimation of the user's mastery degree on this knowledge point.

[0058] Each knowledge point represented by each node in the knowledge graph corresponds to a confidence factor. The confidence factors of each knowledge point can be initialized based on the user's personal learning situation when constructing the knowledge graph. For example, for any user, according to the user's personal learning situation, a mastery degree is initially set for each knowledge point represented by each node in the knowledge graph through a neural network, and at the same time, a confidence factor representing the reliability of this mastery degree evaluation is obtained through evaluation.

[0059] This embodiment proposes to correct the influence of the user's answer result of the current exercise on the confidence factor update based on the duration for which the confidence factor persists. The longer the duration for which the confidence factor persists, the smaller the influence of the current user's answer result on the confidence factor, thereby avoiding excessive reflection or fluctuation of the confidence factor, ensuring that the update of the confidence factor can not only reflect the user's recent learning performance, but also not produce violent fluctuations due to individual accidental factors, so that the mastery degree evaluation result based on the confidence factor is more reliable and accurate.

[0060] Different user answer situations correspond to different confidence factor adjustment amplitudes. Specifically, when the user answer situation indicates a correct answer, it corresponds to a first confidence factor adjustment amplitude, and when the user answer situation indicates a wrong answer, it corresponds to a second confidence factor adjustment amplitude. Since the confidence factor based on this application is used to represent the reliability of the mastery degree evaluation, this embodiment proposes a processing method of appropriately reducing the confidence factor when the user answers wrong and appropriately increasing the confidence factor when the user answers correctly. That is, the first confidence factor adjustment amplitude can be set as a positive number, and the second confidence factor adjustment amplitude can be set as a negative number.

[0061] Based on this, see Figure 2Exemplarily shown is a flowchart of the steps for updating the mastery level of the knowledge points involved in an exercise based on confidence factors. Updating the mastery level of a knowledge point based on the user's answer situation can be achieved through the following methods: S201. When the user's answer situation indicates a correct answer, according to the duration for which the confidence factor of the knowledge point persists, correct the adjustment amplitude of the first confidence factor corresponding to the user's answer result, and update the confidence factor of the knowledge point based on the corrected adjustment amplitude of the first confidence factor. Among them, the longer the duration, the smaller the absolute value of the corrected adjustment amplitude of the first confidence factor; Among them, the duration for which the confidence factor of a knowledge point persists is used to reflect the length of time since the confidence factor was last updated and has remained unchanged. The longer this duration, the longer it indicates that the assessment of the mastery level of this knowledge point has continued for a period of time and there has been no new user answer situation for update during this period. This duration can be expressed as: the time interval between the current trigger for updating the confidence factor of this knowledge point and the most recent completion of the update operation for this confidence factor.

[0062] The duration for which the confidence factor of a knowledge point persists is negatively correlated with the adjustment amplitude of the first confidence factor. The longer the duration, the smaller the absolute value of the adjusted first confidence factor based on the confidence factor correction. For example, if t represents the time interval (i.e., the duration) between the current confidence factor update and the previous update, and α(t) represents a decay function that changes with time t, such as exponential decay / linear decay, then the corrected adjustment amplitude of the first confidence factor can be expressed as: Corrected adjustment amplitude of the first confidence factor = Adjustment amplitude of the first confidence factor when the user's answer situation indicates a correct answer * α(t).

[0063] S202. When the user's answer situation indicates an incorrect answer, according to the duration for which the confidence factor of the knowledge point persists, correct the adjustment amplitude of the second confidence factor corresponding to the user's answer result, and update the confidence factor of the knowledge point based on the corrected adjustment amplitude of the second confidence factor. Among them, the longer the duration, the smaller the absolute value of the corrected adjustment amplitude of the second confidence factor; This implementation step is applicable to updating the confidence factor of a knowledge point when the user answers incorrectly. The longer the duration for which the confidence factor of the knowledge point persists, the smaller the impact of the user's incorrect answer behavior result on the confidence factor, and the smaller the absolute value of the corrected adjustment amplitude of the second confidence factor.

[0064] Regarding adjusting the amplitude based on the corrected first / second confidence factor and updating the confidence factor of this knowledge point. When updating the confidence factor of this knowledge point by adjusting the amplitude with the first confidence factor, the confidence factor is increased through the amplitude adjusted by the first confidence factor. When updating the confidence factor of this knowledge point by adjusting the amplitude with the second confidence factor, the confidence factor is decreased through the amplitude adjusted by the second confidence factor.

[0065] For example, assume that the amplitude adjusted by the first confidence factor is , and the amplitude adjusted by the second confidence factor is , denotes the absolute value, denotes the existing confidence factor of the k-th knowledge point involved in the exercise, then the updated confidence factor can be expressed as:

[0066] Based on the fact that the confidence factor is used to represent the reliability of the mastery degree assessment, so this confidence factor can take values in the interval [0, 1]. During the process of updating the confidence factor, the maximum value of the updated confidence factor is restricted to 1 to prevent the confidence factor from exceeding the reasonable range, ensure that it always reflects the effective confidence interval, and maintain the numerical stability and logical consistency of the mastery degree assessment.

[0067] S203, adjust the mastery degree of this knowledge point in the knowledge graph based on the updated confidence factor.

[0068] In this embodiment, the determined confidence factor serves as the adjustment weight of the adjustment amplitude of the mastery degree, and is used to control the size of the adjustment amplitude of the mastery degree of the knowledge point. The larger the confidence factor, the higher the reliability of the assessment of the mastery degree of the knowledge point, and a more significant adjustment can be made when adjusting the mastery degree of this knowledge point; conversely, if the confidence factor is lower, it indicates that the judgment of the mastery of this knowledge point is not yet clear. At this time, a smaller adjustment amplitude should be adopted to avoid drastic fluctuations in the mastery degree estimation due to excessive uncertainty, thereby enhancing the stability and rationality of the mastery degree assessment process.

[0069] Based on this, the adjustment amplitude of this knowledge point can be weighted and adjusted based on the updated confidence factor to obtain the adjustment amplitude of the mastery degree of this knowledge point, and the mastery degree of this knowledge point is updated using this adjustment amplitude of the mastery degree.

[0070] Among them, the adjustment amplitude of this knowledge point can be a set fixed value. In order to flexibly adapt to the learning status of different users, the difficulty level of knowledge points, and the actual difficulty level of exercises, it is also possible to determine the adjustment amplitude of each knowledge point involved in the exercise based on the existing mastery level and exercise difficulty of each knowledge point before updating the mastery level of each knowledge point involved in the exercise after the user has completed answering the exercise, so as to achieve a more refined and personalized mastery level evaluation and update mechanism. Based on this, the adjustment amplitude of this knowledge point can also be determined in the following way: Based on the question difficulty of this exercise and the mastery level of this knowledge point, predict the user's answering situation when answering this exercise; based on the error between the user's answering situation obtained after the user has completed answering this exercise and the predicted answering situation, determine the adjustment amplitude of this knowledge point.

[0071] Among them, predicting the user's answering situation when answering this exercise can represent the probability of predicting that the user answers correctly or predicts that the user answers incorrectly. Taking the question difficulty d of this exercise and the existing mastery level of the user for this knowledge point (the k-th knowledge point involved in the exercise) as the prediction basis, the user's answering situation of this exercise can be predicted through a trained prediction model or a custom prediction algorithm. For example, the user's answering situation when answering this exercise can be predicted based on the normalization of the product of the question difficulty of this exercise and the existing mastery level of the user for this knowledge point.

[0072] The error between the user's answering situation obtained after the user has completed answering this exercise and the predicted answering situation represents the difference between the actual answering result and the predicted answering result of the user under this knowledge point. The sign of the error indicates the adjustment direction of the mastery level. For example, assuming that the user's actual correct answer is quantified as the value 1 and the user's actual wrong answer is quantified as the value 0, and the predicted answering situation for the k-th knowledge point involved in the exercise is expressed as f(d, Mk), then the error corresponding to this k-th knowledge point can be expressed as:

[0073] When determining the adjustment amplitude of this knowledge point based on the error between the user's answering situation obtained after the user has completed answering this exercise and the predicted answering situation, the error can be directly used as the adjustment amplitude, or a learning rate or other control parameters can also be introduced to the error to control the jump amplitude of the mastery level during update.

[0074] After determining the adjustment amplitude of the knowledge point, the updated confidence factor can be used as the weighted weight of the adjustment amplitude of the knowledge point, and the product of the confidence factor and the adjustment amplitude is used as the adjustment amplitude of the mastery degree of the knowledge point; alternatively, a stability coefficient can be introduced to control the influence intensity of the confidence factor on the adjustment amplitude, and based on the stability coefficient, the confidence factor, and the adjustment amplitude, the adjustment amplitude of the mastery degree of the knowledge point is determined; alternatively, based on the duration for which the confidence factor of the knowledge point lasts, which can reflect the interval duration since the mastery degree of the knowledge point was last updated, the duration for which the confidence factor of the knowledge point lasts can be introduced into the determination process of the adjustment amplitude of the mastery degree of the knowledge point. The longer the duration, the smaller the absolute value of the adjustment amplitude of the mastery degree of the knowledge point obtained, avoiding excessive fluctuations in the mastery degree of the knowledge point caused by a single accidental answering factor and enhancing the stability of the evaluation of the mastery degree of the knowledge point. Regarding the process of determining the adjustment amplitude of the mastery degree of the knowledge point using the confidence factor and adjustment amplitude of the knowledge point, any other applicable calculation method can also be adopted, and this application does not limit it.

[0075] In the embodiments of the present disclosure, by comprehensively considering the user's answering result and the duration for which the confidence factor lasts, and dynamically correcting the first confidence factor adjustment amplitude and the second confidence factor adjustment amplitude according to the user's answering situation (correct or incorrect), differential processing of the confidence factor for different answering results is realized. Furthermore, based on the updated confidence factor, the mastery degree of the knowledge point is adjusted. Considering the influence of the natural attenuation of the confidence factor on the answering behavior over time, short-term fluctuations are prevented from causing excessive interference to the mastery degree, and drastic oscillations of the mastery degree due to accidental answering errors or coincidental correctness are avoided, thereby enhancing the stability and accuracy of the evaluation of the mastery degree.

[0076] In some embodiments, for the mastery degree of the previous knowledge point of the current knowledge point in the knowledge graph based on the mastery degree of the current knowledge point described in the foregoing embodiments, refer to Figure 3 An exemplary flowchart showing the update of the mastery degrees of the previous and subsequent knowledge points of a current knowledge point. When the user's answering situation indicates a correct answer, the update of the mastery degree of the previous knowledge point of the current knowledge point can be achieved through the following process: S301a, if none of the exercises associated with the previous knowledge point have been answered by the user and the mastery degree of the previous knowledge point is the initially set value, then the previous knowledge point inherits the mastery degree of the current knowledge point as the updated mastery degree of the previous knowledge point; The initially set value represents an initially set mastery degree value preset for each knowledge point. For example, the initially set value is 0.7 (assuming the mastery degree range is between 0 and 1, where 0 represents complete non-mastery and 1 represents complete mastery).

[0077] When the mastery level of the previous knowledge point remains at the initial set value and the user has not answered the exercises associated with the previous knowledge point, it indicates that the mastery level of the previous knowledge point has not been updated by actual learning behavior and lacks an effective evaluation basis. Based on the logical dependency relationship between the current knowledge point and the previous knowledge point, in general, to master the current knowledge point, one needs to first master the previous knowledge point. The user's mastery of the current knowledge point can, to a certain extent, reflect their mastery of the previous knowledge point. Therefore, through the mastery level propagation mechanism from front to back, the mastery level information of the current knowledge point can be fed back to the previous knowledge point and used to update its mastery level, thereby enhancing the integrity and rationality of knowledge mastery evaluation.

[0078] In this implementation step, it is detected whether all the exercises associated with the previous knowledge point have not been answered by the user. In other words, it is detected whether there is an exercise among the user's historical answered exercises whose involved knowledge point is the previous knowledge point, and it is confirmed whether the mastery level of the previous knowledge point is the initial set value. If the previous knowledge point meets both of these conditions, the mastery level of the current knowledge point is directly assigned to the previous knowledge point as the updated mastery level of the previous knowledge point.

[0079] S302a, if the mastery level of the previous knowledge point is less than the mastery level of the current knowledge point, then a weighted process is performed on the mastery level of the previous knowledge point and the mastery level of the current knowledge point to determine the updated mastery level of the previous knowledge point; Based on the fact that when the user answers an exercise correctly, the mastery level of each knowledge point involved in the exercise is increased. When the mastery level of the previous knowledge point is less than the mastery level of the current knowledge point, since the user's mastery of the current knowledge point can, to a certain extent, reflect their mastery of the previous knowledge point, therefore, in this embodiment, the mastery levels of the previous knowledge point and the current knowledge point are comprehensively considered, and by performing a weighted process on the mastery level of the previous knowledge point and the mastery level of the current knowledge point, the mastery level of the previous knowledge point is increased.

[0080] The specific rules of the weighted process can be designed according to actual needs. For example, a weight parameter α can be set to control the influence of the mastery level of the current knowledge point and the mastery level of the previous knowledge point on the update of the mastery level of the previous knowledge point. For instance, based on this weight parameter α, the updated mastery level of the previous knowledge point can be expressed as: the updated mastery level of the previous knowledge point = (1 - α) * the mastery level of the previous knowledge point + α * the mastery level of the current knowledge point.

[0081] S303a, if the mastery level of the previous knowledge point is greater than or equal to the mastery level of the current knowledge point, then the mastery level of the previous knowledge point remains unchanged.

[0082] If the mastery level of the previous knowledge point is greater than or equal to that of the current knowledge point, it indicates that the user has a good grasp of the previous knowledge point. In this case, there is no need to adjust the mastery level of the previous knowledge point, and its original value can be maintained, so as to prevent unreasonably "dragging down" the basic knowledge points that have been well mastered, effectively preventing the problem of incorrect adjustment of the mastery level of the previous knowledge point due to the low mastery level of the current knowledge point, thus ensuring that the mastery level evaluation process conforms to the logical dependency relationship between knowledge points.

[0083] For example, assume that the current knowledge point is B, and the previous knowledge point of B is A. Given that the mastery level of the current knowledge point B is Mb and the mastery level of A is Ma, then the updated mastery level Manew of A can be expressed as:

[0084] Based on a similar principle, as Figure 3 shown, when the answer to the user's response indicates an error in the foregoing embodiment, based on the mastery level of the current knowledge point, the mastery level of the subsequent knowledge point of the current knowledge point in the knowledge graph can be updated. The update of the mastery level of the subsequent knowledge point of the current knowledge point can be achieved through the following process: S301b, if none of the exercises associated with the subsequent knowledge point have been answered by the user and the mastery level of the subsequent knowledge point is the initial set value, then the subsequent knowledge point inherits the mastery level of the current knowledge point as the updated mastery level of the subsequent knowledge point; When the mastery level of the subsequent knowledge point is still the initial set value and the user has not answered the exercises associated with the subsequent knowledge point, it indicates that the mastery level of the subsequent knowledge point has not been updated by actual learning behavior and lacks an effective evaluation basis. In this embodiment, based on the logical dependency relationship between the current knowledge point and the subsequent knowledge point, and based on the fact that usually, to master the subsequent knowledge point, one needs to first master the current knowledge point, the user's mastery of the current knowledge point can to a certain extent reflect the user's mastery of the subsequent knowledge point. Therefore, through the mastery level propagation mechanism from the previous knowledge point to the subsequent knowledge point, the mastery level information of the current knowledge point can be propagated to the subsequent knowledge point and used to update its mastery level.

[0085] In this implementation step, it is detected whether all the exercises associated with the subsequent knowledge point have not been answered by the user. In other words, it is detected whether there is an exercise in the user's historical answers whose involved knowledge point is the subsequent knowledge point, and it is confirmed whether the mastery level of the subsequent knowledge point is the initial set value. If the subsequent knowledge point meets both of these conditions, then the mastery level of the current knowledge point is directly assigned to the subsequent knowledge point as the updated mastery level of the subsequent knowledge point.

[0086] S302b, if the mastery level of the subsequent knowledge point is greater than that of the current knowledge point, then a weighted process is performed on the mastery level of the subsequent knowledge point and the mastery level of the current knowledge point to determine the updated mastery level of the subsequent knowledge point; There is a logical sequence relationship between knowledge points in the knowledge graph. Generally, mastering subsequent knowledge points is based on mastering the current knowledge point. When the mastery level of a subsequent knowledge point is higher than that of the current knowledge point, it can to a certain extent reflect that there is an abnormality in the assessment of the mastery level of the subsequent knowledge point (for example, due to the simplicity of the exercises related to the subsequent knowledge point or the user answering correctly by chance, resulting in a high assessment), or the assessment of the mastery level of the current knowledge point is on the low side. Through the weighted process, the mastery situations of the current knowledge point and the subsequent knowledge point can be comprehensively considered, making the mastery level of the subsequent knowledge point more in line with the actual situation, and avoiding affecting the accuracy of the overall mastery level due to the abnormal assessment of a single knowledge point. Therefore, in this embodiment, the mastery levels of the subsequent knowledge point and the current knowledge point are comprehensively considered, and by performing a weighted process on the mastery level of the subsequent knowledge point and the mastery level of the current knowledge point, the mastery level of the previous knowledge point is appropriately reduced.

[0087] The specific rules of the weighted process can be designed according to actual needs. For example, a weight parameter β can be set to control the influence of the mastery level of the current knowledge point and the mastery level of the subsequent knowledge point on the update of the mastery level of the subsequent knowledge point. For example, based on this weight parameter β, the updated mastery level of the subsequent knowledge point can be expressed as: the updated mastery level of the subsequent knowledge point = (1 - β) * the mastery level of the subsequent knowledge point + β * the mastery level of the current knowledge point.

[0088] S303b, if the mastery level of the subsequent knowledge point is less than or equal to that of the current knowledge point, then the mastery level of the subsequent knowledge point remains unchanged.

[0089] When the mastery level of the subsequent knowledge point is less than or equal to that of the current knowledge point, it indicates that the user's mastery of the subsequent knowledge point does not exceed the mastery of the current knowledge point, which conforms to the logical order of knowledge learning, that is, first master the basic knowledge point (the current knowledge point), and then master the more in-depth knowledge point (the subsequent knowledge point). In this case, there is no need to adjust the mastery level of the subsequent knowledge point, and its original value can be maintained, thereby preventing the unreasonable "increase" of the mastery level of the subsequent knowledge point and effectively preventing the problem of incorrect adjustment of the mastery level of the subsequent knowledge point.

[0090] For example, assume that the current knowledge point is B, and the subsequent knowledge point of B is C. Given that the mastery level of the current knowledge point B is Mb and the mastery level of C is Mc, then the updated mastery level Mcnew of C can be expressed as:

[0091] In the embodiments of the present disclosure, when updating the mastery degree of knowledge points in the knowledge graph, the logical relationships between knowledge points and the user's answering situation are fully considered. When the user's answering situation of an exercise indicates a correct answer, the mastery degree of each knowledge point involved in the exercise is propagated and updated along the pre-order direction, while when the user's answering situation of the exercise indicates a wrong answer, the mastery degree of each knowledge point involved in the exercise is propagated and updated along the post-order direction, making the mastery degree evaluation more in line with the user's actual learning state. During the process of propagating and updating the mastery degree, the logical sequence of knowledge points is fully considered, and different processing methods are adopted based on the relative mastery degree relationship between the current knowledge point and the pre-order / post-order knowledge point to be updated, so that the update of the pre-order knowledge point and the post-order knowledge point of the current knowledge point conforms to the logical association between knowledge points, making the mastery degree evaluation more well-founded and improving the accuracy of the mastery degree evaluation.

[0092] In some embodiments, considering that multiple knowledge points involved in an exercise may have the same pre-order knowledge point, for example, multiple current knowledge points share a pre-order knowledge point / post-order knowledge point, in order to avoid repeated updating of the mastery degree of the target knowledge point, which may affect the accuracy and stability of the mastery degree evaluation, this embodiment proposes an anti-duplicate update mechanism based on an update mark to prevent repeated updating of the same target knowledge point (such as a pre-order or post-order knowledge point) during one exercise answering event.

[0093] Based on this, when updating the mastery degree of the target knowledge point of the current knowledge point based on the mastery degree of the current knowledge point, where the target knowledge point includes either a pre-order knowledge point or a post-order knowledge point, it is possible to first detect whether the target knowledge point has undergone a mastery degree update operation after the exercise has been answered by the user: If so, skip the current mastery degree update, that is, terminate the update of the mastery degree of the target knowledge point. When the current knowledge point includes multiple target knowledge points, update the mastery degree of other target knowledge points except the target knowledge point that has been updated.

[0094] If not, update the mastery degree of the target knowledge point of the current knowledge point based on the mastery degree of the current knowledge point, and after the mastery degree update of the target knowledge point is completed, mark that the target knowledge point has undergone a mastery degree update operation after the exercise has been answered by the user.

[0095] Regarding marking that the mastery degree update operation has been performed on the target knowledge point after the user has answered the exercise, data storage forms such as a temporary set, queue, or list can be used to record the updated target knowledge points. Thus, when updating the mastery degree of the target knowledge point of the current knowledge point based on the mastery degree of the current knowledge point, first check whether the target knowledge point of the current knowledge point exists in the data storage such as the temporary set, queue, or list. If it exists, skip the update. If it does not exist, perform the mastery degree update of the target knowledge point and add the updated target knowledge point to the data storage such as the temporary set, queue, or list. Or, other marking means can also be used, such as adding relevant attribute fields to the nodes in the knowledge graph, etc. This application will not repeat it here.

[0096] For example, assume that exercise 3 involves knowledge points M, N, and P, where: The prerequisite knowledge points of knowledge point M are G and L; The prerequisite knowledge points of knowledge point N are J and K; The prerequisite knowledge point of knowledge point K is L; According to the method provided in this embodiment, after updating the mastery degrees of knowledge point M involved in the exercise and its prerequisite knowledge points G and L, mark M, G, and L to indicate that the mastery degree update has been performed after the user has answered exercise 3; Subsequently, when updating the mastery degrees of knowledge point N and its prerequisite knowledge points J and K, continue to propagate forward to the prerequisite knowledge point L of J and K. At this time, since L has been marked as "the mastery degree update has been performed after the user has answered exercise 3" during the propagation of the mastery degree of knowledge point M before, the repeated mastery degree update operation on L will be skipped, thus avoiding the problem of multiple updates caused by multiple current knowledge points sharing the same prerequisite knowledge point.

[0097] In some embodiments, for the knowledge graph described in the foregoing step S101, refer to Figure 4 The schematic diagram of the knowledge graph construction process shown exemplarily, which can be constructed in the following manner: S401, use a large model to judge the anterior and posterior order of the two knowledge points in each knowledge point pair in the knowledge point set, determine the dependency relationship of each knowledge point pair, and use the large model to add the chapter information of the textbook tree to each knowledge point; The large model refers to an artificial intelligence model with powerful learning ability and complex structure, such as BERT, GPT, etc. In this embodiment, the large model can handle complex natural language understanding and generation tasks, and is used to judge the anterior and posterior order relationship between knowledge points and add chapter information.

[0098] The knowledge point set contains all the knowledge points that need to be organized and processed. Each knowledge point represents a specific learning unit or concept. The knowledge point set can be obtained by collecting a large number of questions containing knowledge points, or it can be extracted from textbooks, course outlines, or other educational materials through extracting knowledge points. A knowledge point pair consists of two randomly selected knowledge points from the knowledge point set. This knowledge point pair is used to analyze the dependency relationship between knowledge points to determine their connection method in the knowledge graph.

[0099] Predecessor and successor judgment is used to determine the learning order relationship between two knowledge points, that is, whether one knowledge point is a prerequisite or a successor knowledge point of another knowledge point. Predecessor and successor judgment is a key step in constructing the knowledge graph, which helps to determine the dependency relationship between knowledge points. This dependency relationship can represent the relationship that one knowledge point needs to be mastered before learning another knowledge point, and can be used to construct the directed edges in the knowledge graph to represent the learning order between knowledge points.

[0100] When using a large model to perform predecessor and successor judgment on the two knowledge points in each knowledge point pair in the knowledge point set, the knowledge point pair can be input so that the large model performs predecessor and successor analysis based on the concept descriptions of the two knowledge points in the knowledge point pair, and then outputs the dependency relationship between the two knowledge points. For example, for the knowledge point pair (A, B), the model output result indicates that knowledge point A is a prerequisite knowledge point of knowledge point B.

[0101] At the same time, the large model can be used to add the chapter information of the textbook tree to each knowledge point in the knowledge point pair. For example, the model can obtain the pre-stored chapter information of the textbook tree and determine the chapter information of the textbook tree that matches the knowledge point based on the concept description of the knowledge point.

[0102] S402, construct a knowledge graph based on the dependency relationship of each knowledge point pair, and for the isolated knowledge points in the knowledge graph, use the chapter information of the textbook tree of the isolated knowledge point and the chapter sequence of the textbook tree to connect the isolated knowledge point to the knowledge graph; In the process of constructing a knowledge graph based on the dependency relationship of each knowledge point pair, each knowledge point is used as a node, and the dependency relationship is used as a directed edge. The directed edge is added based on the rule that "the starting knowledge point of the directed edge is the prerequisite knowledge point of the ending knowledge point of the directed edge".

[0103] An isolated knowledge point refers to a knowledge point that has no connection with other knowledge points in the knowledge graph. There may be such isolated knowledge points in the knowledge graph constructed based on the dependency relationship of knowledge point pairs. Then, the chapter information of the isolated knowledge point can be used to connect it to the adjacent knowledge points in the chapter sequence. For example, if the isolated knowledge point belongs to Chapter 3, and there are already knowledge points in Chapter 2 and Chapter 4 in the knowledge graph, the isolated knowledge point can be connected to the knowledge points in Chapter 2 or Chapter 4.

[0104] S403. Adjust the node connection relationships in the knowledge graph to convert the knowledge graph into a directed acyclic graph, obtaining the completed knowledge graph.

[0105] A directed acyclic graph is a directed graph without directed cycles, used to represent nodes and edges in a knowledge graph. Converting the knowledge graph into a directed acyclic graph can avoid circular dependencies and ensure the logical correctness of the knowledge graph.

[0106] After the construction of the knowledge graph connection is completed, circular dependencies in the knowledge graph, that is, self - dependencies and unreasonable connection relationships, can be detected through expert review or graph algorithms, and processed into a directed acyclic graph to ensure that knowledge points have clear pre - order and post - order relationships. Additionally, the graph can be pruned. By counting all paths between every two nodes in the graph, if there are multiple paths between two nodes, only the longest path can be retained.

[0107] Corresponding to the embodiments of the foregoing method for updating the knowledge point mastery degree, refer to Figure 5 As shown, the present application also provides embodiments of an apparatus for updating the knowledge point mastery degree. The apparatus includes: A knowledge point mastery degree update module 501 for exercises, which is used to determine each knowledge point involved in the exercises completed by the user in the constructed knowledge graph, and update the mastery degree of the knowledge point based on the user's answering situation. Each node in the knowledge graph represents a corresponding knowledge point, and the directed edge between two nodes in the knowledge graph indicates the dependency relationship between the two nodes. Among them, the starting node of the directed edge is the pre - order node of the ending node of the directed edge, and the ending node of the directed edge is the post - order node of the starting node of the directed edge. A pre - order knowledge point mastery degree update module 502, which is used to, when the user's answering situation indicates a correct answer, for each knowledge point involved in the exercise, take the knowledge point as the current knowledge point, and based on the mastery degree of the current knowledge point, update the mastery degree of the pre - order knowledge point of the current knowledge point in the knowledge graph. When the first iteration termination condition is not currently met, take the pre - order knowledge point of the current knowledge point as the current knowledge point, and return to update the mastery degree of the pre - order knowledge point of the current knowledge point in the knowledge graph based on the mastery degree of the current knowledge point. The subsequent knowledge point mastery update module 503 is used to, when the user's answer indicates an incorrect answer, for each knowledge point involved in the exercise, take this knowledge point as the current knowledge point, and based on the mastery of the current knowledge point, update the mastery of the subsequent knowledge points of this current knowledge point in the knowledge graph. And when the second iteration termination condition is not currently met, take the subsequent knowledge point of the current knowledge point as the current knowledge point, and return to update the mastery of the previous knowledge points of this current knowledge point in the knowledge graph based on the mastery of the current knowledge point.

[0108] In some embodiments, the exercise-related knowledge point mastery update module includes: The first correction module is used to, when the user's answer indicates a correct answer, correct the adjustment amplitude of the first confidence factor corresponding to the user's answer result according to the duration for which the confidence factor of the knowledge point persists, and update the confidence factor of this knowledge point based on the corrected adjustment amplitude of the first confidence factor; wherein, the longer the duration, the smaller the corrected adjustment amplitude of the first confidence factor, and the smaller the absolute value of the corrected adjustment amplitude of the first confidence factor; The second correction module is used to, when the user's answer indicates an incorrect answer, correct the adjustment amplitude of the second confidence factor corresponding to the user's answer result according to the duration for which the confidence factor of the knowledge point persists, and update the confidence factor of this knowledge point based on the corrected adjustment amplitude of the second confidence factor; wherein, the longer the duration, the smaller the corrected adjustment amplitude of the second confidence factor, and the smaller the absolute value of the corrected adjustment amplitude of the second confidence factor; The update module is used to adjust the mastery of this knowledge point in the knowledge graph based on the updated confidence factor.

[0109] In some embodiments, the update module is specifically used for: Perform weighted adjustment on the adjustment amplitude of this knowledge point based on the updated confidence factor to obtain the mastery adjustment amplitude of this knowledge point, and use this mastery adjustment amplitude to update the mastery of this knowledge point.

[0110] In some embodiments, the adjustment amplitude of the knowledge point is determined by the following method: Based on the difficulty of the exercise and the mastery of this knowledge point, predict the user's answering situation when answering this exercise; Based on the error between the user's answer situation obtained after the user answers this exercise and the predicted answering situation, determine the adjustment amplitude of this knowledge point.

[0111] In some embodiments, the previous knowledge point mastery update module is specifically used for: If none of the exercises associated with the previous knowledge point has been answered by the user and the mastery level of the previous knowledge point is the initially set value, then the previous knowledge point inherits the mastery level of the current knowledge point as the updated mastery level of the previous knowledge point; If the mastery level of the previous knowledge point is less than the mastery level of the current knowledge point, then a weighted processing is performed on the mastery level of the previous knowledge point and the mastery level of the current knowledge point to determine the updated mastery level of the previous knowledge point; If the mastery level of the previous knowledge point is greater than or equal to the mastery level of the current knowledge point, then the mastery level of the previous knowledge point remains unchanged.

[0112] In some embodiments, the subsequent knowledge point mastery level update module is specifically configured to: If none of the exercises associated with the subsequent knowledge point has been answered by the user and the mastery level of the subsequent knowledge point is the initially set value, then the subsequent knowledge point inherits the mastery level of the current knowledge point as the updated mastery level of the subsequent knowledge point; If the mastery level of the subsequent knowledge point is greater than the mastery level of the current knowledge point, then a weighted processing is performed on the mastery level of the subsequent knowledge point and the mastery level of the current knowledge point to determine the updated mastery level of the subsequent knowledge point; If the mastery level of the subsequent knowledge point is less than or equal to the mastery level of the current knowledge point, then the mastery level of the subsequent knowledge point remains unchanged.

[0113] In some embodiments, when the previous knowledge point / subsequent knowledge point mastery level update module is used to update the mastery level of the target knowledge point of the current knowledge point based on the mastery level of the current knowledge point, the target knowledge point includes any one of the previous knowledge point or the subsequent knowledge point, and includes: Detect whether the target knowledge point has performed a mastery level update operation after the exercises have been answered by the user; If so, skip the current mastery level update; If not, then update the mastery level of the target knowledge point of the current knowledge point based on the mastery level of the current knowledge point, and after the mastery level of the target knowledge point is updated, mark that the target knowledge point has performed a mastery level update operation after the exercises have been answered by the user.

[0114] In some embodiments, the knowledge graph is constructed in the following manner: Use a large model to determine the dependency relationship of each pair of knowledge points in the knowledge point set by judging the front and back order of the two knowledge points in each pair of knowledge points, and use the large model to add the chapter information of the textbook tree to each knowledge point; Construct a knowledge graph based on the dependency relationship of each pair of knowledge points, and for the isolated knowledge points in the knowledge graph, use the chapter information of the textbook tree of the isolated knowledge points and the order of the chapters in the textbook tree to connect the isolated knowledge points to the knowledge graph; Adjust the node connection relationship in the knowledge graph to convert the knowledge graph into a directed acyclic graph, and obtain the constructed knowledge graph.

[0115] The implementation processes of the functions and roles of each unit in the above device are specifically described in the implementation processes of the corresponding steps in the above method, and will not be elaborated here.

[0116] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this application. Those of ordinary skill in the art can understand and implement it without creative efforts.

[0117] The embodiment of the present application also provides an electronic device, and the structural schematic diagram of the electronic device is as Figure 6 shown. The electronic device 600 includes at least one processor 601, a memory 602, and a bus 603. At least one processor 601 is electrically connected to the memory 602; the memory 602 is configured to store at least one computer-executable instruction, and the processor 601 is configured to execute the at least one computer-executable instruction, so as to execute the steps of any one of the knowledge point mastery degree update methods provided in any one of the embodiments or any optional implementation manners of the present application.

[0118] Further, the processor 601 can be an FPGA (Field-Programmable Gate Array), or other devices with logical processing capabilities, such as an MCU (Microcontroller Unit) or a CPU (Central Process Unit).

[0119] The embodiment of the present application also provides another readable storage medium, storing a computer program, which is used to implement the steps of any one of the knowledge point mastery degree update methods provided in any one of the embodiments or any optional implementation manners of the present application when being executed by a processor.

[0120] The readable storage medium provided by the embodiments of the present application includes, but is not limited to, any type of disk (including floppy disks, hard disks, optical disks, CD-ROMs, and magneto-optical disks), ROM (Read-Only Memory), RAM (Random Access Memory), EPROM (Erasable Programmable Read-Only Memory), EEPROM (Electrically Erasable Programmable Read-Only Memory), flash memory, magnetic cards, or optical cards. That is, the readable storage medium includes any medium that stores or transmits information in a form readable by a device (e.g., a computer).

[0121] Thus, specific embodiments of the subject matter have been described. Other embodiments are within the scope of the appended claims. In some cases, the acts recited in the claims can be performed in a different order and still achieve the desired result. In addition, the processes depicted in the figures are not necessarily in the particular order or sequential order shown to achieve the desired result. In some implementations, multitasking and parallel processing may be advantageous.

[0122] The foregoing is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present application shall be included within the scope of protection of the present application.

Claims

1. A method for updating the mastery degree of knowledge points, characterized in that The method comprises: Determine each knowledge point involved in the exercises completed by the user in the constructed knowledge graph, and update the mastery of the knowledge point based on the user's answer situation; each node in the knowledge graph represents a corresponding knowledge point, and a directed edge between two nodes in the knowledge graph indicates a dependency relationship between the two nodes, wherein the starting node of a directed edge is the preceding node of the ending node of the directed edge, and the ending node of the directed edge is the succeeding node of the starting node of the directed edge; When the user's answer indicates that the answer is correct, for each knowledge point involved in the exercise, the knowledge point is used as the current knowledge point, and based on the mastery of the current knowledge point, the mastery of the previous knowledge point of the current knowledge point in the knowledge graph is updated; and when the first iteration termination condition is not currently met, the previous knowledge point of the current knowledge point is used as the current knowledge point, the mastery based on the current knowledge point is returned, and the mastery of the previous knowledge point of the current knowledge point in the knowledge graph is updated; When the user's answer indicates an incorrect answer, for each knowledge point involved in the exercise, the knowledge point is used as the current knowledge point, and based on the mastery of the current knowledge point, the mastery of the subsequent knowledge points of the current knowledge point in the knowledge graph is updated. When the second iteration termination condition is not met, the subsequent knowledge point of the current knowledge point is used as the current knowledge point, and the mastery based on the current knowledge point is returned to update the mastery of the preceding knowledge point of the current knowledge point in the knowledge graph.

2. The method according to claim 1, characterized in that The updating of the mastery of the knowledge point based on the user's answer includes: When the user's answer indicates that the answer is correct, the first confidence factor adjustment amplitude corresponding to the user's answer result is corrected according to the duration of the confidence factor of the knowledge point, and the confidence factor of the knowledge point is updated based on the corrected first confidence factor adjustment amplitude; wherein, the longer the duration, the smaller the absolute value of the corrected first confidence factor adjustment amplitude; When the user's answer indicates an incorrect answer, the second confidence factor adjustment amplitude corresponding to the user's answer result is corrected according to the duration of the confidence factor of the knowledge point, and the confidence factor of the knowledge point is updated based on the corrected second confidence factor adjustment amplitude; wherein, the longer the duration, the smaller the absolute value of the corrected second confidence factor adjustment amplitude; Adjust the mastery of the knowledge point in the knowledge graph based on the updated confidence factor.

3. The method according to claim 2, wherein Adjusting the mastery of the knowledge point in the knowledge graph based on the updated confidence factor includes: The adjustment amplitude of the knowledge point is weightedly adjusted based on the updated confidence factor to obtain the mastery adjustment amplitude of the knowledge point, and the mastery of the knowledge point is updated using the mastery adjustment amplitude.

4. The method according to claim 3, characterized in that The adjustment amplitude of the knowledge point is determined by the following method: Based on the difficulty of the exercise and the mastery of the knowledge point, predict the user's answer to the exercise; Determine the adjustment amplitude of the knowledge point based on the error between the user's answer situation obtained after the exercise is answered by the user and the predicted answer situation.

5. The method according to claim 1, wherein When the user's answer situation indicates a correct answer, update the mastery level of the previous knowledge points of the current knowledge point in the knowledge graph based on the mastery level of the current knowledge point, including: If none of the exercises associated with the previous knowledge point have been answered by the user and the mastery level of the previous knowledge point is the initially set value, then the previous knowledge point inherits the mastery level of the current knowledge point as the updated mastery level of the previous knowledge point; If the mastery level of the previous knowledge point is less than the mastery level of the current knowledge point, perform a weighted processing on the mastery level of the previous knowledge point and the mastery level of the current knowledge point to determine the updated mastery level of the previous knowledge point; If the mastery level of the previous knowledge point is greater than or equal to the mastery level of the current knowledge point, keep the mastery level of the previous knowledge point unchanged.

6. The method according to claim 1, wherein When the user's answer situation indicates a wrong answer, update the mastery level of the subsequent knowledge points of the current knowledge point in the knowledge graph based on the mastery level of the current knowledge point, including: If none of the exercises associated with the subsequent knowledge point have been answered by the user and the mastery level of the subsequent knowledge point is the initially set value, then the subsequent knowledge point inherits the mastery level of the current knowledge point as the updated mastery level of the subsequent knowledge point; If the mastery level of the subsequent knowledge point is greater than the mastery level of the current knowledge point, perform a weighted processing on the mastery level of the subsequent knowledge point and the mastery level of the current knowledge point to determine the updated mastery level of the subsequent knowledge point; If the mastery level of the subsequent knowledge point is less than or equal to the mastery level of the current knowledge point, keep the mastery level of the subsequent knowledge point unchanged.

7. The method according to claim 1, characterized in that, Update the mastery level of the target knowledge point of the current knowledge point based on the mastery level of the current knowledge point. The target knowledge point includes any one of the previous knowledge point or the subsequent knowledge point, including: Detect whether the mastery level update operation has been performed on the target knowledge point after the exercise is answered by the user; If so, skip the current mastery level update; If not, update the mastery level of the target knowledge point of the current knowledge point based on the mastery level of the current knowledge point, and after the mastery level update of the target knowledge point is completed, mark that the mastery level update operation has been performed on the target knowledge point after the exercise is answered by the user.

8. The method according to claim 1, characterized in that, The knowledge graph is constructed in the following manner: Use a large model to judge the front and back order of knowledge points for each pair of knowledge points in the knowledge point set, determine the dependency relationship of each pair of knowledge points, and use the large model to add chapter information of the textbook tree to each knowledge point; Construct a knowledge graph based on the dependency relationship of each pair of knowledge points, and for the isolated knowledge points in the knowledge graph, use the chapter information of the textbook tree of the isolated knowledge point and the chapter sequence of the textbook tree to connect the isolated knowledge point to the knowledge graph; Adjust the node connection relationship in the knowledge graph to convert the knowledge graph into a directed acyclic graph to obtain the constructed knowledge graph.

9. A knowledge point mastery degree updating device, characterized in that, The device includes: The exercise-related knowledge point mastery degree update module is used to determine each knowledge point involved in the exercise completed by the user in the constructed knowledge graph, and update the mastery degree of the knowledge point based on the user's answer situation; each node in the knowledge graph represents a corresponding knowledge point, and the directed edge between two nodes in the knowledge graph indicates the dependency relationship between the two nodes, where the starting node of the directed edge is the pre-order node of the ending node of the directed edge, and the ending node of the directed edge is the post-order node of the starting node of the directed edge; The pre-order knowledge point mastery degree update module is used to, when the user's answer situation indicates a correct answer, for each knowledge point involved in the exercise, take this knowledge point as the current knowledge point, and based on the mastery degree of the current knowledge point, update the mastery degree of the pre-order knowledge point of the current knowledge point in the knowledge graph, and when the first iteration termination condition is not currently met, take the pre-order knowledge point of the current knowledge point as the current knowledge point, and return to update the mastery degree of the pre-order knowledge point of the current knowledge point in the knowledge graph based on the mastery degree of the current knowledge point; The post-order knowledge point mastery degree update module is used to, when the user's answer situation indicates a wrong answer, for each knowledge point involved in the exercise, take this knowledge point as the current knowledge point, and based on the mastery degree of the current knowledge point, update the mastery degree of the post-order knowledge point of the current knowledge point in the knowledge graph, and when the second iteration termination condition is not currently met, take the post-order knowledge point of the current knowledge point as the current knowledge point, and return to update the mastery degree of the pre-order knowledge point of the current knowledge point in the knowledge graph based on the mastery degree of the current knowledge point.

10. An electronic device, characterized in that, It includes: A memory and a processor; The memory is used to store a computer program; The processor is used to call the computer program to implement the method according to any one of claims 1-8.

11. A readable storage medium, on which a computer program is stored, characterized in that, When the program is executed by the processor, it implements the method according to any one of claims 1-8.

Images