A method, apparatus, electronic device, and storage medium for determining learning resources.

Abstract

This application provides a method, apparatus, electronic device, and storage medium for determining learning resources. The method includes: obtaining the target question corresponding to the current lecture from a knowledge graph resource database; obtaining the current user's answer results for the target question, and constructing a mastery matrix for each knowledge point in the current lecture based on the answer results; determining the current user's weak knowledge points in the current lecture based on the mastery matrix; obtaining a first teaching resource corresponding to the weak knowledge points from the knowledge graph resource database, and displaying the first teaching resource to the current user; if the current user's feedback information on the first teaching resource meets the path recommendation requirements, planning a target learning path using the knowledge graph resource database, and recommending a corresponding second teaching resource to the current user based on the target learning path. By adopting the above-mentioned method, apparatus, electronic device, and storage medium for determining learning resources, the problem of low learning efficiency caused by the inability to provide users with suitable learning resources is solved.

Description

Technical Field

[0001] This application relates to the field of online education technology, and more specifically, to a method, apparatus, electronic device, and storage medium for determining learning resources. Background Technology

[0002] With the popularization of computer technology and the rapid development of mobile internet, many traditional industries are gradually moving towards internetization, and the education industry is one of them. As online education becomes more and more widespread, various intelligent education products have emerged, such as automatic grading, homework recommendation, and practice question delivery.

[0003] However, existing methods for determining learning resources suffer from significant individual differences in learning ability and process among users during self-study, making it difficult to provide suitable learning resources and resulting in low learning efficiency. Summary of the Invention

[0004] In view of this, the purpose of this application is to provide a method, apparatus, electronic device and storage medium for determining learning resources, so as to solve the problem of low learning efficiency caused by the inability to provide users with suitable learning resources.

[0005] In a first aspect, embodiments of this application provide a method for determining learning resources, including:

[0006] Retrieve the target questions for this lecture from the knowledge graph resource base;

[0007] Obtain the current user's answer to the target question, and construct a mastery matrix of the current user for each knowledge point in this lecture based on the answer results;

[0008] Based on the mastery matrix, identify the user's weak knowledge points in this lecture;

[0009] Retrieve the first teaching resource corresponding to the weak knowledge point from the knowledge graph resource library and display the first teaching resource to the current user;

[0010] If the current user's feedback on the first teaching resource meets the path recommendation requirements, a target learning path is planned using the knowledge graph resource base. Based on the target learning path, a corresponding second teaching resource is recommended to the current user for learning. The second teaching resource is the teaching resource corresponding to the related knowledge points of the weak knowledge points.

[0011] Optionally, a mastery matrix for each knowledge point in this lecture can be constructed based on the answer results, including: for each knowledge point in this lecture, determining the current user's mastery level of the knowledge point based on the answer results corresponding to that knowledge point; and constructing the current user's mastery matrix based on the current user's mastery level of each knowledge point, the learning ability of the corresponding knowledge point, the difficulty of the target question, and the complexity of each knowledge point.

[0012] Optionally, based on the mastery matrix, the weak knowledge points of the current user in this lecture are determined, including: determining the overall mastery rating of the current user in this lecture based on the mastery matrix; determining the mastery score corresponding to each knowledge point; if the overall mastery rating is the first rating, sorting all knowledge points in descending order of mastery score, and selecting the knowledge point ranked last as the weak knowledge point; if the overall mastery rating is the second rating, selecting the knowledge points with mastery scores less than the score threshold as the weak knowledge points.

[0013] Optionally, the first teaching resource corresponding to the weak knowledge point is obtained from the knowledge graph resource base, including: sorting multiple question explanation videos of the target question corresponding to the weak knowledge point according to the set rules, and selecting the explanation video ranked first as the target explanation video; selecting the lowest difficulty among multiple difficulty levels of the target question corresponding to the weak knowledge point as the target difficulty, and obtaining the question corresponding to the target difficulty from the knowledge graph resource base as the new target question; and using the new target question and the target explanation video as the first teaching resource.

[0014] Optionally, the target learning path is planned using a knowledge graph resource base, including: obtaining the associated knowledge points corresponding to the weak knowledge points from the knowledge graph resource base; constructing at least one candidate learning path corresponding to the weak knowledge points according to the association relationship of knowledge points in the knowledge graph resource base; and selecting the target learning path from the at least one candidate learning path according to preset rules.

[0015] Optionally, before retrieving the target questions for this lecture from the knowledge graph resource base, the process further includes: determining the teaching content for each subject; performing cluster analysis on the teaching content based on content similarity and content relevance, grouping teaching content belonging to the same category into a single lesson, and creating a chapter tree; or, breaking down the teaching content into multiple target knowledge points at different levels according to the characteristics of the teaching topics, and creating a chapter tree according to the hierarchical structure between different target knowledge points; and generating a systematic learning framework based on the chapter tree so that the current user can learn according to the systematic learning framework.

[0016] Optionally, the method further includes: for each subject, breaking it down into multi-level knowledge points, determining the dependencies between knowledge points at the same level and different levels; using knowledge points at different levels as nodes, connecting different nodes according to the dependencies between knowledge points to construct a knowledge graph; acquiring learning resources, including teaching resources, question resources, and material resources; and for nodes in the knowledge graph, associating the node with the corresponding learning resources to generate a knowledge graph resource library.

[0017] Secondly, embodiments of this application also provide a learning resource determination apparatus, the apparatus comprising:

[0018] The question acquisition module is used to retrieve the target questions for this lecture from the knowledge graph resource library;

[0019] The matrix construction module is used to obtain the current user's answer results for the target question and construct a mastery matrix of the current user for each knowledge point in this lecture based on the answer results;

[0020] The knowledge point identification module is used to identify the user's weak knowledge points in this lecture based on the mastery matrix.

[0021] The first resource identification module is used to retrieve the first teaching resource corresponding to the weak knowledge point from the knowledge graph resource library and display the first teaching resource to the current user.

[0022] The second resource determination module is used to plan a target learning path using a knowledge graph resource base if the current user's feedback information on the first teaching resource meets the path recommendation requirements, and recommend corresponding second teaching resources to the current user based on the target learning path. The second teaching resource is the teaching resource corresponding to the related knowledge points of the weak knowledge points.

[0023] Thirdly, embodiments of this application also provide an electronic device, including: a processor, a memory, and a bus. The memory stores machine-readable instructions executable by the processor. When the electronic device is running, the processor communicates with the memory via the bus. When the machine-readable instructions are executed by the processor, the steps of the learning resource determination method described above are performed.

[0024] Fourthly, embodiments of this application also provide a computer-readable storage medium storing a computer program, which, when executed by a processor, performs the steps of the learning resource determination method described above.

[0025] The embodiments of this application bring the following beneficial effects:

[0026] This application provides a learning resource determination method, apparatus, electronic device, and storage medium that can determine a mastery matrix based on the current user's answer performance in a current lecture, and determine the current user's weak knowledge points based on the mastery matrix. It can then obtain the first teaching resource corresponding to the weak knowledge point from a knowledge graph resource library and obtain the second teaching resource based on the target learning path. Compared with existing learning resource determination methods, this solves the problem of low learning efficiency caused by the inability to provide users with suitable learning resources.

[0027] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0028] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0029] Figure 1 A flowchart of the learning resource determination method provided in an embodiment of this application is shown;

[0030] Figure 2 This illustration shows a schematic diagram of the structure of the knowledge graph provided in an embodiment of this application;

[0031] Figure 3 A schematic diagram of the learning resource determination device provided in an embodiment of this application is shown;

[0032] Figure 4 A schematic diagram of the structure of the electronic device provided in the embodiments of this application is shown. Detailed Implementation

[0033] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of this application. Based on the embodiments of this application, every other embodiment obtained by those skilled in the art without inventive effort falls within the scope of protection of this application.

[0034] It is worth noting that prior to this application, with the popularization of computer technology and the rapid development of the mobile internet, many traditional industries, including the education industry, have gradually moved towards internetization. As online education becomes increasingly prevalent, various intelligent education products have emerged, such as automatic grading, homework recommendations, and practice question delivery. However, existing methods for determining learning resources suffer from significant individual differences in learning ability and process among users, making it difficult to provide suitable learning resources and resulting in low learning efficiency.

[0035] Based on this, embodiments of this application provide a method for determining learning resources to improve the accuracy of learning resource recommendations and increase self-learning efficiency.

[0036] Please see Figure 1 , Figure 1 This is a flowchart illustrating a method for determining learning resources provided in an embodiment of this application. Figure 1 As shown in the embodiments of this application, the method for determining learning resources includes:

[0037] Step S101: Obtain the target topic corresponding to this lecture from the knowledge graph resource base.

[0038] In this step, the knowledge graph resource library can refer to a learning resource library built based on a knowledge graph, which includes learning resources corresponding to each knowledge point.

[0039] The target questions can refer to the multiple practice or test questions sent to the current user in this course.

[0040] In this embodiment of the application, the current user is a student who studies on his own through a teaching application software installed on a smart terminal device. The teaching application software sets up multiple lectures, each corresponding to multiple knowledge points. The questions corresponding to these knowledge points are the target questions for this lecture.

[0041] Since the knowledge graph resource library stores questions corresponding to each knowledge point, multiple target questions for this lecture can be retrieved from the knowledge graph resource library based on the knowledge points corresponding to this lecture.

[0042] In an optional embodiment, before obtaining the target topic corresponding to the current lecture from the knowledge graph resource base, the method further includes: determining the teaching content corresponding to each subject; performing cluster analysis on the teaching content based on content similarity and content relevance, grouping teaching content belonging to the same category as a section, and creating a chapter tree; or, splitting the teaching content into multiple target knowledge points at different levels according to the characteristics of the teaching topics, and creating a chapter tree according to the hierarchical structure between different target knowledge points; and generating a system learning framework based on the chapter tree so that the current user can learn according to the system learning framework.

[0043] Specifically, to provide users with systematic learning content, the educational application software also offers a chapter tree for users to study systematically. The chapter tree is arranged according to the structure of chapters, sections, and lectures.

[0044] When constructing a chapter tree for a specific subject, the following approach can be used: First, obtain regional textbooks from different regions, arrange the chapter tree according to the regional textbooks, and then display the chapter tree corresponding to the current user's region.

[0045] Secondly, when textbooks are identical or largely identical, cluster analysis can be performed on the content based on content similarity and relevance. For example, related teaching content from different chapters can be grouped into one category, forming a section. A chapter tree can then be created based on the chapters, sections, and lectures derived from the cluster analysis. Alternatively, multiple teaching topics can be identified, along with the characteristics of each topic. For instance, if the teaching topic is trigonometric functions, the teaching content can be broken down into multiple target knowledge points corresponding to trigonometric functions. These target knowledge points can be categorized into different levels based on their granularity. Assuming there are two levels, each first-level knowledge point can correspond to multiple second-level knowledge points. Then, each topic can be treated as a chapter, each first-level knowledge point as a section, and one or more corresponding second-level knowledge points under each section as a lecture, creating a chapter tree.

[0046] Third, construct specialized chapter trees based on teaching objectives and scenarios. Here, one or more sub-chapter trees are selected from the chapter tree according to the teaching objectives and scenarios, and the selected sub-chapter trees are reconstructed to generate specialized chapter trees. For example, for a particular error-prone question, because the question is relatively complex, a corresponding specialized chapter tree can be constructed to help users learn systematically.

[0047] In an optional embodiment, the method further includes: for each subject, breaking down the subject into multi-level knowledge points, and determining the dependencies between knowledge points at the same level and different levels; using knowledge points at different levels as nodes, connecting different nodes according to the dependencies between knowledge points to construct a knowledge graph; acquiring learning resources, including teaching resources, question resources, and material resources; and for nodes in the knowledge graph, associating the node with the corresponding learning resources to generate a knowledge graph resource library.

[0048] Here, teaching resources include, but are not limited to: textbook outlines, curriculum systems, teaching objectives, teaching materials, and lecture videos.

[0049] The resources include, but are not limited to: pre-class exercises, post-class exercises, in-class quizzes, and exam questions.

[0050] Resources include, but are not limited to: audio and video, images, text, fun animations, and interactive games.

[0051] Specifically, before obtaining the target topics for this lecture, a knowledge graph resource base needs to be constructed. Taking third-grade first-semester Chinese as an example, the subject is broken down into multiple knowledge points. When breaking down, the size and granularity of the knowledge points need to be considered. Here, the size and granularity of the knowledge points are determined according to the course duration. For example, if a class is 40 minutes long, and there are 3 knowledge points that can be taught well in these 40 minutes, then if there are 80 Chinese classes in a semester, the third-grade first-semester Chinese can be broken down into 240 knowledge points.

[0052] Then, following the principle of gradual teaching, determine the dependency relationship between different knowledge points. For example, if knowledge point A is taught before knowledge point B is taught, then knowledge point A and knowledge point B have a sequential relationship. Another example is that knowledge point A and knowledge point B can be taught in parallel, then they have a parallel relationship.

[0053] The identified knowledge points are designated as the first layer of knowledge points. For each first-level knowledge point, all possible problem-solving approaches are analyzed. Then, the knowledge point is subdivided according to the solution methods of the problems, with each sub-category becoming a problem module. These problem modules are designated as the second layer of knowledge points. Furthermore, different problem modules have dependencies such as sequential or parallel relationships, which need to be determined before constructing the knowledge graph. Taking equation solving problems as an example, the first problem module involves factoring the equation before solving it, while the second involves directly using the radical formula. Since factoring the equation is simpler than using the radical formula and is learned before the latter, the first and second problem modules are determined to have a sequential relationship.

[0054] After determining the dependencies between knowledge points at the same level and between knowledge points at different levels, the knowledge graph is constructed. See below for reference. Figure 2 Let's introduce knowledge graphs.

[0055] Figure 2 A schematic diagram of the structure of the knowledge graph provided in an embodiment of this application is shown.

[0056] like Figure 2As shown, circles represent nodes in the knowledge graph. Black circles represent first-level knowledge points, and white circles represent second-level knowledge points. Knowledge points 211 and 212 are connected by line segments with arrows, indicating their order of instruction; knowledge point 211 is taught after knowledge point 212. Line segments also connect first-level and second-level knowledge points: knowledge point 211 corresponds to knowledge points 221, 222, and 223, and knowledge point 212 corresponds to knowledge points 224 and 225. Furthermore, dashed lines represent dependencies between different second-level knowledge points; dashed lines without arrows represent parallel relationships, while dashed lines with arrows represent sequential relationships, with knowledge point 223 taught after knowledge point 222.

[0057] After constructing the knowledge graph, for each node in the knowledge graph, the corresponding learning resources are stored as an association with that knowledge point node. For example, the resource IDs of the teaching resources, question resources, and material resources corresponding to the node are associated with that node. By associating each node with its corresponding learning resources, a knowledge graph resource library can be generated.

[0058] Step S102: Obtain the current user's answer to the target question, and construct the current user's mastery matrix of each knowledge point in this lecture based on the answer results.

[0059] In this step, the mastery matrix refers to a matrix used to reflect the current user's mastery of each knowledge point in this lecture.

[0060] For example, the mastery matrix is ​​a two-dimensional matrix, with the first dimension being the knowledge point identifier and the second dimension being multiple evaluation items used to reflect the degree of mastery.

[0061] In this embodiment of the application, after the current user answers and submits the answers to multiple target questions in this lecture, the teaching application software will obtain the answer results and construct a mastery matrix reflecting the degree of mastery of the knowledge points based on the answer results.

[0062] In one optional embodiment, constructing a mastery matrix for each knowledge point in the current lecture based on the answer results includes: for each knowledge point in the current lecture, determining the current user's mastery level of the knowledge point based on the answer results corresponding to the knowledge point; and constructing the current user's mastery matrix based on the current user's mastery level of each knowledge point, the learning ability of the corresponding knowledge point, the difficulty of the target question, and the complexity of each knowledge point.

[0063] Here, "mastery level" refers to the score indicating how well the user has mastered the knowledge point. This score is used to evaluate the user's current level of mastery of the knowledge point from the perspective of test-taking scores. The mastery level is related not only to the score of the target questions in this lecture, but also to the user's previous scores on the same knowledge point.

[0064] Learning ability is used to characterize the degree of improvement in mastery between adjacent lectures for a given knowledge point. Adjacent lectures for a given knowledge point can be either consecutive or spaced out, but they are adjacent lectures for that specific knowledge point.

[0065] The difficulty of the target question can refer to the difficulty level of the target question. In the embodiments of this application, the questions under each knowledge point can be divided into multiple difficulty levels, and a question of a certain difficulty level under the knowledge point is selected as the target question.

[0066] The complexity of a knowledge point refers to the level of difficulty of the knowledge point itself, and the complexity of a knowledge point represents the learning cost of that knowledge point.

[0067] Specifically, after a user submits their answer to a target question in this lecture, for each knowledge point in this lecture, the user's mastery score for this lecture is calculated by multiplying their answer score for that knowledge point by the question's difficulty coefficient. This mastery score is then summed with the weighted average mastery scores for that knowledge point across multiple points within a certain historical timeframe to obtain the user's overall mastery score for that knowledge point. Answers more recent than those given in the lecture have higher weights, while those further back in time have lower weights. This method of obtaining a mastery score more comprehensively and accurately reflects the user's current level of understanding of that knowledge point.

[0068] The first mastery score is calculated by weighting the overall mastery score for that knowledge point, the difficulty score corresponding to the difficulty of the target question, and the complexity score corresponding to the complexity of that knowledge point. This first mastery score is based on the user's performance on questions related to that knowledge point. Clearly, evaluating the user's mastery matrix for that knowledge point solely based on question-answering performance is inaccurate.

[0069] Since the learning ability of the same user varies for different knowledge points, and the improvement level of the same user in learning the same knowledge point also varies, it is necessary to take the user's learning ability for that knowledge point into account when determining the current user's mastery matrix. Therefore, after determining the first mastery score, a learning ability curve for the current user for that knowledge point can be constructed. Here, the improvement rate of mastery in the historical lectures for that knowledge point is calculated, and the curve corresponding to the improvement rate is used as the learning ability curve. The learning ability curve and the difficulty level of the questions corresponding to each lecture are input into the predictive neural network model to calculate the current user's predicted mastery score for that knowledge point in this lecture. The predicted mastery scores of other users are ranked with the current user's predicted mastery score to obtain the current user's predicted ranking. The mastery levels of other users for the target questions in this lecture are ranked with the current user's mastery levels of the target questions in this lecture to obtain the actual ranking.

[0070] The actual ranking is compared with the predicted ranking. Based on the comparison result, it is determined whether the current user has achieved the expected learning effect on the knowledge point. If the expected learning effect has not been achieved, the corresponding score is subtracted from the first mastery score to obtain the second mastery score. If the expected learning effect has been achieved, the corresponding score is added to the first mastery score to obtain the second mastery score. If the second mastery score is within a set score range, it is determined that the current user has mastered the knowledge point. For example, if the second mastery score is greater than 70, it is determined that the current user has mastered the knowledge point. Otherwise, it is determined that the current user has not mastered the knowledge point. Alternatively, the current user's mastery rating for the knowledge point is determined according to the score range corresponding to the second mastery score, and the value of the mastery rating is used as the value of the knowledge point in the mastery matrix.

[0071] Step S103: Based on the mastery matrix, determine the weak knowledge points of the current user in this lecture.

[0072] In this step, assuming that this lecture covers a total of 6 knowledge points, the mastery matrix is ​​constructed for these 6 knowledge points. Based on the mastery matrix, the user's weak knowledge points in this lecture can be determined.

[0073] In one optional embodiment, determining the weak knowledge points of the current user in this lecture based on the mastery matrix includes: determining the overall mastery rating of the current user in this lecture based on the mastery matrix; determining the mastery score corresponding to each knowledge point; if the overall mastery rating is the first rating, sorting all knowledge points in descending order of mastery score, and selecting the knowledge point ranked last as the weak knowledge point; if the overall mastery rating is the second rating, selecting the knowledge point with a mastery score less than the score threshold as the weak knowledge point.

[0074] Specifically, assuming that one out of six knowledge points is not mastered, the calculated mastery rate is 1 / 6 = 0.1667. Then, the values ​​for the first and second rating thresholds are set. If the mastery rate is less than the first rating threshold, the overall mastery rating is set to the first level. If the mastery rate is greater than the first rating threshold but less than the second rating threshold, the overall mastery rating is set to the second level. If the mastery rate is greater than the second rating threshold, the overall mastery rating is set to the third level. The mastery score includes both the first and second mastery scores.

[0075] If the overall mastery rating is Level 1, all knowledge points are sorted in descending order of their Level 2 mastery scores, and the knowledge point ranked last is selected as the weak knowledge point. If the overall mastery rating is Level 2, knowledge points with Level 2 mastery scores below the scoring threshold are selected as weak knowledge points. If the overall mastery rating is Level 3, all knowledge points are considered weak knowledge points.

[0076] Step S104: Obtain the first teaching resource corresponding to the weak knowledge point from the knowledge graph resource library, and display the first teaching resource to the current user.

[0077] In this step, the first teaching resource may refer to the teaching resource corresponding to the weak knowledge point, and the first teaching resource is used to improve the user's mastery of the weak knowledge point.

[0078] For example, the first teaching resource can be a teaching resource, such as a video explaining example problems or a video explaining knowledge points; it can be a question resource, such as practice questions; or it can be a material resource.

[0079] In this embodiment, since each knowledge point is associated with its own learning resource, after identifying weak knowledge points, the corresponding learning resource can be found in the knowledge graph resource library. This learning resource is the first teaching resource. When presenting the first teaching resource to the current user, the order of resource delivery can be determined based on the mastery score. If the score is below a set threshold, it indicates that the current user has not truly understood the knowledge point. In this case, explanation resources and materials are pushed first. Explanation resources can better improve the current user's understanding of the knowledge point, while materials can increase the fun and interactivity of learning. Then, corresponding practice questions are pushed. If the score is above the set threshold, it indicates that the current user's understanding of the knowledge point is acceptable, but the problem-solving approach is unclear. In this case, example explanation videos and materials can be pushed, followed by subsequent practice questions. Therefore, after determining the mastery matrix, this application does not simply push practice questions or explanation videos directly, but rather accurately positions the user's mastery level of the knowledge point based on the mastery score to recommend more suitable learning resources to the user.

[0080] In one optional embodiment, obtaining the first teaching resource corresponding to the weak knowledge point from the knowledge graph resource library includes: sorting multiple question explanation videos of the target question corresponding to the weak knowledge point according to a set rule, and selecting the explanation video ranked first as the target explanation video; selecting the lowest difficulty among multiple difficulty levels of the target question corresponding to the weak knowledge point as the target difficulty, obtaining the question corresponding to the target difficulty from the knowledge graph resource library as the new target question; and using the new target question and the target explanation video as the first teaching resource.

[0081] Specifically, first, determine the details of the rules. These rules can be based on the number of likes or the number of views. Taking the number of likes as an example, sort the multiple videos explaining the questions related to the weak knowledge point in descending order of the number of likes, and select the video with the most likes as the target video.

[0082] Since these are weak knowledge points, in order to improve users' learning motivation and confidence, the lowest difficulty level is directly set as the target difficulty level, and new target questions are selected from multiple questions corresponding to the target difficulty level to reduce users' frustration with answering questions incorrectly.

[0083] Step S105: If the current user's feedback information on the first teaching resource meets the path recommendation requirements, a target learning path is planned using the knowledge graph resource base, and a corresponding second teaching resource is recommended to the current user for learning based on the target learning path. The second teaching resource is the teaching resource corresponding to the related knowledge points of the weak knowledge points.

[0084] In this step, the path recommendation requirement can refer to mastering the scoring requirements.

[0085] A target learning path can refer to a recommended path for learning resources. A target learning path is used to indicate other knowledge points associated with weak knowledge points.

[0086] In this embodiment, a user's lack of mastery of a weak knowledge point may be due to a lack of understanding of that specific knowledge point or a lack of understanding of related prior knowledge points. Therefore, a first teaching resource is recommended. After the user watches the instructional video and completes the lowest-difficulty practice questions, the second mastery score is recalculated. If the second mastery score is lower than a set value (e.g., the second mastery score for the target question is used as the set value), it indicates that the user may lack understanding of prior knowledge points. Therefore, other knowledge points related to the weak knowledge point are retrieved from the knowledge graph resource library, and these knowledge points are organized into a learning path based on their dependencies. The second teaching resource is then determined based on this learning path.

[0087] In one optional embodiment, the target learning path is planned using a knowledge graph resource library, including: obtaining the associated knowledge points corresponding to the weak knowledge points from the knowledge graph resource library; constructing at least one candidate learning path corresponding to the weak knowledge points according to the association relationship of the knowledge points in the knowledge graph resource library; and selecting the target learning path from the at least one candidate learning path according to preset rules.

[0088] Specifically, the learning path formed based on the dependencies between knowledge points may be multiple, i.e., multiple candidate learning paths. Taking weak knowledge point C as an example, assuming two candidate learning paths are identified, namely ACD and BCD, we can use a random walk strategy to randomly select one candidate learning path as the target learning path, or set a business strategy based on policy rules and select a candidate learning path as the target learning path according to the set rules, or select the candidate learning path with the highest benefit as the target learning path based on reinforcement learning methods.

[0089] After determining the target learning path, the teaching resources corresponding to the knowledge points before the weak knowledge points in the target path are used as secondary teaching resources. For example, when the target learning path is ACD, the teaching resources corresponding to knowledge point A are secondary teaching resources.

[0090] Compared with existing methods for determining learning resources, this application can determine the mastery matrix based on the current user's answers in this lecture, and determine the current user's weak knowledge points based on the mastery matrix. This allows for the acquisition of the first teaching resource corresponding to the weak knowledge point from the knowledge graph resource library, and the acquisition of the second teaching resource based on the target learning path. This solves the problem of low learning efficiency caused by the inability to provide users with suitable learning resources.

[0091] Based on the same inventive concept, this application also provides a learning resource determination device corresponding to the learning resource determination method. Since the principle of the device in this application is similar to the learning resource determination method described above in this application, the implementation of the device can refer to the implementation of the method, and the repeated parts will not be described again.

[0092] Please see Figure 3 , Figure 3 This is a schematic diagram of a learning resource determination device provided in an embodiment of this application. Figure 3 As shown, the learning resource determination device 300 includes:

[0093] The question acquisition module 301 is used to retrieve the target question corresponding to this lecture from the knowledge graph resource library;

[0094] The matrix construction module 302 is used to obtain the current user's answer results for the target question and construct the current user's mastery matrix for each knowledge point in this lecture based on the answer results;

[0095] The knowledge point identification module 303 is used to identify the weak knowledge points of the current user in this lecture based on the mastery matrix.

[0096] The first resource determination module 304 is used to obtain the first teaching resource corresponding to the weak knowledge point from the knowledge graph resource library and display the first teaching resource to the current user.

[0097] The second resource determination module 305 is used to plan a target learning path using a knowledge graph resource base if the current user's feedback information on the first teaching resource meets the path recommendation requirements, and recommend the corresponding second teaching resource to the current user for learning based on the target learning path. The second teaching resource is the teaching resource corresponding to the related knowledge point of the weak knowledge point.

[0098] Please see Figure 4 , Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Figure 4 As shown, the electronic device 400 includes a processor 410, a memory 420, and a bus 430.

[0099] The memory 420 stores machine-readable instructions executable by the processor 410. When the electronic device 400 is running, the processor 410 communicates with the memory 420 via the bus 430. When the machine-readable instructions are executed by the processor 410, they can perform the operations described above. Figure 1 The steps of the learning resource determination method in the illustrated method embodiment can be found in the method embodiment for specific implementation, and will not be repeated here.

[0100] This application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, can perform the above-described actions. Figure 1 The steps of the learning resource determination method in the illustrated method embodiment can be found in the method embodiment for specific implementation, and will not be repeated here.

[0101] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0102] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the shown or discussed mutual couplings, direct couplings, or communication connections may be through some communication interfaces; indirect couplings or communication connections between devices or units may be electrical, mechanical, or other forms.

[0103] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0104] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0105] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a processor-executable, non-volatile, computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0106] Finally, it should be noted that the above-described embodiments are merely specific implementations of this application, used to illustrate the technical solutions of this application, and not to limit them. The scope of protection of this application is not limited thereto. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments, or make equivalent substitutions for some of the technical features, within the scope of the technology disclosed in this application. Such modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for determining learning resources, characterized in that, include: Retrieve the target questions for this lecture from the knowledge graph resource base; Obtain the current user's answer to the target question, and construct a mastery matrix of the current user for each knowledge point in this lecture based on the answer. Based on the mastery matrix, identify the user's weak knowledge points in this lecture. Obtain the first teaching resource corresponding to the weak knowledge point from the knowledge graph resource library, and display the first teaching resource to the current user; If the current user's feedback on the first teaching resource meets the path recommendation requirements, a target learning path is planned using the knowledge graph resource library, and a corresponding second teaching resource is recommended to the current user for learning based on the target learning path; Wherein, the first teaching resource is the teaching resource corresponding to the weak knowledge point, and the second teaching resource is the teaching resource corresponding to the related knowledge point of the weak knowledge point; The process of determining the current user's weak knowledge points in this lecture based on the mastery matrix includes: Based on the non-mastery rate of multiple knowledge points in the mastery matrix, the overall mastery rating of the current user in this lecture is determined. For each knowledge point, a mastery score is determined. The mastery score includes a first mastery score and a second mastery score. The first mastery score is determined based on the answer score, the difficulty and complexity of the target question, and the second mastery score is determined based on the first mastery score and the current user's learning ability. If the overall mastery rating is the highest, all knowledge points are sorted in descending order of mastery score, and the knowledge point ranked last is selected as the weak knowledge point. If the overall mastery rating is the second rating, the knowledge points with a mastery score less than the rating threshold are selected as weak knowledge points. If the overall mastery rating is level three, all knowledge points are considered weak knowledge points. The second mastery score is determined using the following methods: The curve corresponding to the improvement of mastery in the historical lectures of this knowledge point is used as the learning ability curve. The learning ability curve and the difficulty level of the questions corresponding to each lecture are input into the prediction neural network model to calculate the predicted mastery score of the current user in this lecture. The predicted ranking of the current user is obtained by ranking the predicted mastery scores of other users with those of the current user. The actual ranking is obtained by ranking the mastery of the target questions in this lecture by other users with those of the current user. Based on the comparison between the actual ranking and the predicted ranking, determine whether the current user has achieved the expected learning effect on this knowledge point; Based on the first mastery score, the second mastery score is obtained by adding or subtracting points according to whether the expected learning effect has been achieved.

2. The method according to claim 1, characterized in that, The process of constructing a mastery matrix for each knowledge point in this lecture based on the answer results includes: For each knowledge point in this lecture, the user's level of mastery of that knowledge point is determined based on the corresponding quiz results. Based on the user's current mastery of each knowledge point, their learning ability for the corresponding knowledge point, the difficulty of the target question, and the complexity of each knowledge point, a mastery matrix is ​​constructed for the current user.

3. The method according to claim 1, characterized in that, The step of obtaining the first teaching resource corresponding to the weak knowledge point from the knowledge graph resource base includes: According to the set rules, the video explanations of multiple questions corresponding to the weak knowledge points are sorted, and the video explanation with the first ranking is selected as the target explanation video. Select the lowest difficulty level among multiple difficulty levels of the target questions corresponding to the weak knowledge points as the target difficulty, and obtain the questions corresponding to the target difficulty from the knowledge graph resource library as new target questions; The new target questions and target explanation videos will be used as the primary teaching resources.

4. The method according to claim 1, characterized in that, The process of planning the target learning path using a knowledge graph resource base includes: Obtain the associated knowledge points corresponding to the weak knowledge points from the knowledge graph resource library; Based on the relationships between knowledge points in the knowledge graph resource base, construct at least one candidate learning path corresponding to the weak knowledge points; According to preset rules, a target learning path is selected from the at least one candidate learning path.

5. The method according to claim 1, characterized in that, Before retrieving the target topic for this lecture from the knowledge graph resource base, the following steps are also included: For each subject, determine the corresponding teaching content; The teaching content is clustered based on content similarity and content relevance. Teaching content belonging to the same category is grouped into a section, and a chapter tree is created. Alternatively, the teaching content is divided into multiple target knowledge points at different levels according to the characteristics of the teaching topics, and a chapter tree is created according to the hierarchical structure between different target knowledge points. A systematic learning framework is generated based on the chapter tree, so that the current user can learn according to the systematic learning framework.

6. The method according to claim 1, characterized in that, The method further includes: For each subject, the subject is broken down into multi-level knowledge points, and the dependencies between knowledge points at the same level and different levels are determined. By using knowledge points at different levels as nodes, and connecting these nodes according to the dependencies between them, a knowledge graph can be constructed. Acquire learning resources, including teaching resources, question resources, and material resources; For each node in the knowledge graph, the node is associated with the corresponding learning resource to generate a knowledge graph resource library.

7. A learning resource determination device, characterized in that, include: The question acquisition module is used to retrieve the target questions for this lecture from the knowledge graph resource library; The matrix construction module is used to obtain the current user's answer results for the target question, and construct the current user's mastery matrix for each knowledge point in this lecture based on the answer results; The knowledge point identification module is used to identify the weak knowledge points of the current user in this lecture based on the mastery matrix. The first resource determination module is used to obtain the first teaching resource corresponding to the weak knowledge point from the knowledge graph resource library and display the first teaching resource to the current user. The second resource determination module is used to plan a target learning path using a knowledge graph resource library if the current user's feedback information on the first teaching resource meets the path recommendation requirements, and recommend the corresponding second teaching resource to the current user for learning based on the target learning path. Wherein, the first teaching resource is the teaching resource corresponding to the weak knowledge point, and the second teaching resource is the teaching resource corresponding to the related knowledge point of the weak knowledge point; The knowledge point determination module is specifically used for: Based on the non-mastery rate of multiple knowledge points in the mastery matrix, the overall mastery rating of the current user in this lecture is determined. For each knowledge point, a mastery score is determined. The mastery score includes a first mastery score and a second mastery score. The first mastery score is determined based on the answer score, the difficulty and complexity of the target question, and the second mastery score is determined based on the first mastery score and the current user's learning ability. If the overall mastery rating is the highest, all knowledge points are sorted in descending order of mastery score, and the knowledge point ranked last is selected as the weak knowledge point. If the overall mastery rating is the second rating, the knowledge points with a mastery score less than the rating threshold are selected as weak knowledge points. If the overall mastery rating is level three, all knowledge points are considered weak knowledge points. The knowledge point determination module determines the second mastery score using the following methods: The curve corresponding to the improvement of mastery in the historical lectures of this knowledge point is used as the learning ability curve. The learning ability curve and the difficulty level of the questions corresponding to each lecture are input into the prediction neural network model to calculate the predicted mastery score of the current user in this lecture. The predicted ranking of the current user is obtained by ranking the predicted mastery scores of other users with those of the current user. The actual ranking is obtained by ranking the mastery of the target questions in this lecture by other users with those of the current user. Based on the comparison between the actual ranking and the predicted ranking, determine whether the current user has achieved the expected learning effect on this knowledge point; Based on the first mastery score, the second mastery score is obtained by adding or subtracting points according to whether the expected learning effect has been achieved.

8. An electronic device, characterized in that, include: The device includes a processor, a storage medium, and a bus, wherein the storage medium stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor communicates with the storage medium via the bus, and the processor executes the machine-readable instructions to perform the steps of the learning resource determination method as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, performs the steps of the learning resource determination method as described in any one of claims 1 to 6.

Images