Recommended knowledge learning method, system, computer device and storage medium

By constructing a direct knowledge path map and integrating knowledge graph, combining knowledge fuzzy points and mastery calculations, the problems of inaccurate knowledge diagnosis and large differences in recommendation resources in the existing technology are solved, and more accurate and comprehensive recommendations of knowledge learning resources are achieved. .

CN112000881BActive Publication Date: 2024-12-06GUANGZHOU WEIMING ZHONGZHI EDUCATION TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202010761038.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2020-07-31
Publication Date
2024-12-06
Estimated Expiration
2040-07-31

AI Technical Summary

Technical Problem

The existing learning platform system is not accurate enough in knowledge diagnosis and recommendation, and cannot accurately identify specific problems of learners in the knowledge path, resulting in huge differences between the recommended resources and the original questions' knowledge points, and cannot effectively help learners transfer their problem-solving ideas.

Method used

By obtaining the correct problem-solving steps of the problem, building a directed knowledge path map, collecting learners' problem-solving information, obtaining knowledge fuzzy points, and correlating the directed knowledge path map with the fuzzy points, pushing recommended learning questions, and using knowledge point confidence and mastery calculation, split the path map, integrate the knowledge map, calculate the examination probability and recommendation priority, and accurately recommend the study questions.

Benefits of technology

The accuracy and comprehensiveness of knowledge recommendations have been improved. Learners can focus on wrong knowledge paths, accurately reflect the mastery, identify the weaknesses of important knowledge paths, and provide more accurate learning resources.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN112000881B_ABST
    Figure CN112000881B_ABST
Patent Text Reader

Abstract

The present application relates to a learning method, system, computer device and storage medium for recommending knowledge. The method includes obtaining the correct problem-solving steps of a question, associating the knowledge points of the question through the correct problem-solving steps, connecting the knowledge points to construct a knowledge directed path diagram; collecting the learner's problem-solving information, and obtaining the learner's knowledge fuzzy points according to the problem-solving information; associating the knowledge directed path diagram with the knowledge fuzzy points to obtain recommended learning questions, and pushing the recommended learning questions to the learner. Through this method, based on the knowledge point directed path, the learner's knowledge problems can be better reflected, and the questions found by recommending based on the knowledge point directed path are more similar to the original questions, which improves the accuracy of recommended knowledge learning. After the learner has done a lot of questions, the mastery of each knowledge point will be accurately reflected, and the learner's problems can be more comprehensively grasped, which improves the comprehensiveness of recommended knowledge learning.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence, and in particular to a learning method, system, computer device and storage medium for recommending knowledge. Background Art

[0002] With the development of the Internet, artificial intelligence and big data technology, more and more companies are entering the field of online education, providing learners with platform systems for personalized learning or adaptive learning. These learning platform systems can help learners get rid of the dilemma of blindly brushing questions and make learning more targeted. The key link is the diagnosis and recommendation of knowledge. These learning platform systems currently implement recommendations for knowledge learning through the following steps: 1. Further split the knowledge points into smaller units of knowledge points. 2. Questions and other learning resources are associated with one or more knowledge points. 3. When learners are doing questions, the questions are wrong, which means that the learners have problems with the knowledge points associated with the questions. 4. Recommend learning resources associated with these knowledge points to learners.

[0003] However, the problem with the learning method of recommending knowledge in the above learning platform system is that the analysis and diagnosis of learners' problems and resource recommendations are not accurate enough. For example, a question is not a pile of knowledge points, but represents the entire path from known conditions to the connected paths between knowledge points and finally reaching the solution. Learners seem to understand a single knowledge point, but when doing the question, they cannot think of the solution idea and cannot migrate from one knowledge point to another under known conditions.

[0004] In addition, even if the learner associates the knowledge points, since the question only associates a small number of knowledge points, when the learner makes a mistake in solving the question, we can only know that the learner may have problems with these knowledge points, but we cannot find out which link in the knowledge path the learner has a problem with. Therefore, this solution expands the scope of knowledge problems when diagnosing learners' problems; at the same time, the recommended resources may also associate one or more other knowledge points, which will lead to a huge difference between the knowledge points of the recommended resources and the knowledge points of the original question. Summary of the invention

[0005] Based on this, it is necessary to provide a learning method, system, computer device and storage medium for recommending knowledge in response to the above technical problems.

[0006] In a first aspect, an embodiment of the present invention provides a method for learning recommended knowledge, the method comprising the following steps:

[0007] Obtaining correct problem-solving steps for the question, associating the knowledge points of the question with the correct problem-solving steps, and connecting the knowledge points to construct a knowledge directed path diagram;

[0008] Collecting learners' problem-solving information, and obtaining learners' knowledge ambiguities based on the problem-solving information;

[0009] The knowledge directed path graph is associated with the knowledge fuzzy points to obtain recommended learning questions, and the recommended learning questions are pushed to the learners.

[0010] Furthermore, obtaining the learner's knowledge ambiguity points according to the problem-solving information includes:

[0011] Disassembling the problem-solving information to obtain incorrect problem-solving steps;

[0012] The confidence of a single knowledge point is obtained through the wrong problem-solving steps, and the confidence of the single knowledge point is accumulated to construct the mastery of the knowledge point;

[0013] The knowledge fuzzy points are obtained according to the mastery degree of the knowledge points.

[0014] Furthermore, the method of associating the knowledge directed path graph and the knowledge fuzzy point to obtain recommended learning questions includes: splitting the knowledge directed path graph into a number of knowledge point sub-path graphs, storing the questions mapped by the knowledge point sub-path graphs in a question list, searching for questions similar to the knowledge fuzzy point in the question list, and obtaining directly recommended learning questions.

[0015] Furthermore, the method of associating the knowledge directed path graph and the knowledge fuzzy points to obtain recommended learning questions also includes: integrating the knowledge directed path graph into a knowledge graph, obtaining question examination probability through the knowledge graph, calculating the question examination probability and the knowledge mastery to obtain question recommendation priority, and obtaining knowledge point indirect recommended learning questions based on the question recommendation priority.

[0016] On the other hand, an embodiment of the present invention further provides a learning system for recommending knowledge, including:

[0017] A knowledge path module is used to obtain the correct solution steps of the question, associate the knowledge points of the question with the correct solution steps, and connect the knowledge points to construct a knowledge directed path diagram;

[0018] A knowledge diagnosis module is used to collect the learner's problem-solving information and obtain the learner's knowledge ambiguity points based on the problem-solving information;

[0019] The recommended learning module is used to associate the knowledge directed path diagram with the knowledge fuzzy points to obtain recommended learning questions, and push the recommended learning questions to learners.

[0020] Furthermore, the knowledge diagnosis module includes a knowledge analysis unit, which is used to disassemble the problem-solving information to obtain incorrect problem-solving steps; obtain the confidence of a single knowledge point through the incorrect problem-solving steps, accumulate the confidence of the single knowledge point to construct the knowledge point mastery; and obtain the knowledge fuzzy point according to the knowledge point mastery.

[0021] Furthermore, the recommendation learning module includes a direct recommendation learning unit, which is used to split the knowledge directed path graph into several knowledge point sub-path graphs, store the questions mapped by the knowledge point sub-path graphs in a question list, search for questions similar to the knowledge ambiguity points in the question list, and obtain direct recommendation learning questions.

[0022] Furthermore, the recommendation learning module includes an indirect recommendation learning unit, which is used to integrate the knowledge directed path graph into a knowledge graph, obtain the question examination probability through the knowledge graph, calculate the question examination probability and the knowledge mastery to obtain the question recommendation priority, and obtain the knowledge point indirect recommended learning questions according to the question recommendation priority.

[0023] The embodiment of the present invention further discloses a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the following steps when executing the computer program:

[0024] Obtaining correct problem-solving steps for the question, associating the knowledge points of the question with the correct problem-solving steps, and connecting the knowledge points to construct a knowledge directed path diagram;

[0025] Collecting learners' problem-solving information, and obtaining learners' knowledge ambiguities based on the problem-solving information;

[0026] The knowledge directed path graph is associated with the knowledge fuzzy points to obtain recommended learning questions, and the recommended learning questions are pushed to the learners.

[0027] The embodiment of the present invention further discloses a computer-readable storage medium, on which a computer program is stored, wherein when the computer program is executed by a processor, the following steps are implemented:

[0028] Obtaining correct problem-solving steps for the question, associating the knowledge points of the question with the correct problem-solving steps, and connecting the knowledge points to construct a knowledge directed path diagram;

[0029] Collecting learners' problem-solving information, and obtaining learners' knowledge ambiguities based on the problem-solving information;

[0030] The knowledge directed path graph is associated with the knowledge fuzzy points to obtain recommended learning questions, and the recommended learning questions are pushed to the learners.

[0031] The learning method, device, computer equipment and storage medium for recommending knowledge, the method includes obtaining the correct solution steps of the question, associating the knowledge points of the question through the correct solution steps, connecting the knowledge points to construct a knowledge directed path diagram; collecting the learner's problem-solving information, obtaining the learner's knowledge fuzzy points according to the problem-solving information; associating the knowledge directed path diagram with the knowledge fuzzy points to obtain recommended learning questions, and pushing the recommended learning questions to the learner. Through this method, first, based on the knowledge point directed path, the learner's knowledge problems can be better reflected. The learner does not need to study each knowledge point contained in the right or wrong question to the same extent, but only needs to focus on the wrong knowledge path. Since the learner's problem is not just a single knowledge point problem, when there are multiple knowledge points jointly examined, the questions found by recommending based on the knowledge point directed path are more similar to the original questions, which improves the accuracy of knowledge recommendation learning. Second, after the learner has done a lot of questions, the mastery of each knowledge point will be accurately reflected. At the same time, combined with the joint examination probability diagram, which important knowledge paths are poorly mastered can also be reflected. In this way, we can more comprehensively grasp the problems that learners have and improve the comprehensiveness of recommended knowledge learning. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 A schematic diagram of a flow chart of a method for learning recommended knowledge in an embodiment of the present invention;

[0033] Figure 2 A schematic diagram of a process for obtaining knowledge ambiguity points according to problem-solving information in an embodiment of the present invention;

[0034] Figure 3 A flowchart of the steps of obtaining learning questions for direct recommendation of questions in an embodiment of the present invention;

[0035] Figure 4 A schematic diagram of a flow chart of steps for obtaining learning questions for indirect recommendation of knowledge points in an embodiment of the present invention;

[0036] Figure 5 A structural block diagram of a learning system for recommending knowledge in an embodiment provided by the present invention;

[0037] Figure 6 An internal structure diagram of a computer device in an embodiment of the present invention. DETAILED DESCRIPTION

[0038] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0039] In one embodiment, Figure 1 As shown, a method for learning recommended knowledge is provided, which can be applied to a terminal or a server. The terminal or the server can be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, and portable wearable devices. The method includes the following steps:

[0040] Step 101, obtain the correct solution steps of the question, associate the knowledge points of the question through the correct solution steps, and connect the knowledge points to construct a knowledge directed path diagram.

[0041] Step 102, collecting the learner's problem-solving information, and obtaining the learner's knowledge ambiguity points based on the problem-solving information.

[0042] Step 103, associating the knowledge directed path graph with the knowledge fuzzy points to obtain recommended learning questions, and pushing the recommended learning questions to the learners.

[0043] Specifically, in the embodiment of the present invention, by associating the problem-solving steps of each question with knowledge points, a knowledge point connected path diagram is formed; the path diagram is split into sub-paths for storage, and each sub-path is associated with related questions; when the learner views the problem-solving steps, he / she selects his / her wrong problem-solving steps; and according to the problem-solving knowledge point path diagram or the knowledge point examination probability diagram, questions or other learning resources are pushed. Through this method, first, based on the knowledge point directed path, the learner's knowledge problems can be better reflected. The learner does not need to study each knowledge point contained in the right or wrong question to the same extent, but only needs to focus on the wrong knowledge path. Since the learner's problem is not just a single knowledge point problem, when there are multiple knowledge points jointly examined, the questions found by recommending based on the knowledge point directed path are more similar to the original questions, which improves the accuracy of recommended knowledge learning. Second, after the learner has done a lot of questions, the mastery of each knowledge point will be accurately reflected. At the same time, combined with the joint examination probability diagram, which important knowledge paths are poorly mastered can also be reflected. In this way, the problems existing in the learner can be more comprehensively grasped, and the comprehensiveness of recommended knowledge learning is improved.

[0044] In one embodiment, Figure 2 As shown, the acquiring of the learner's knowledge ambiguity points according to the problem-solving information includes:

[0045] Step 201, disassembling the problem-solving information to obtain incorrect problem-solving steps;

[0046] Step 202, obtaining the confidence of a single knowledge point through the incorrect problem-solving steps, and accumulating the confidence of the single knowledge point to construct the knowledge point mastery;

[0047] Step 203, obtaining the knowledge fuzzy points according to the mastery of the knowledge points.

[0048] Among them, when the learner makes a mistake in a question, how to judge where the learner's problem lies? The traditional method can only judge all related knowledge points as wrong. Because the present invention disassembles the problem-solving steps, the learner can choose his own wrong steps. You can choose a single step or multiple steps. When a certain problem-solving step is selected incorrectly, it can be judged on the surface that there is a gap in which knowledge point. However, it is not enough to just judge that the learner has a problem in this knowledge point. Because the learner may have no problem when examining the knowledge point alone, but when examining multiple knowledge points in a mixed way, which condition to start from and how to reach the goal. It's confusing. Therefore, we still need to locate the error on the entire examination path of the question. So, how to use the mastery of knowledge points to express his knowledge mastery? Here, we define the mastery of knowledge points as M. The mastery of this knowledge point in a single question is mi, the difficulty factor corresponding to the question is di, and the confidence of a single question is ci.

[0049]

[0050] That is, the mastery of knowledge points M is equal to the single mastery multiplied by the difficulty factor of the question multiplied by the single confidence and then added together. The single mastery represents the mastery of each knowledge point of the question when doing this question. If the question is answered correctly, the single mastery of all knowledge points is 1. If the question is answered incorrectly, the knowledge point corresponding to the wrong step has the lowest single mastery, followed by the subsequent knowledge points, and finally the previous knowledge points. The principle is that the closer to the error point, the lower the mastery, and the mastery behind the error point is lower than the previous one. The specific value is set by the system. Then, the system will assign a difficulty factor to questions of different difficulty levels. The more difficult the question, the higher the difficulty factor, and the simpler the question, the lower the difficulty factor. This setting will not lead to the situation where the question is too simple and the mastery is considered high when it is answered correctly. Confidence is a parameter to measure whether the mastery is credible. Generally speaking, the more times, the greater the credibility. The closest to the current time, the greater the credibility. Therefore, the single confidence can be decayed in order from near to far according to the time of answering the question. The most recent confidence is the highest, the farther away from the current time, the lower, and the confidence after N times can be ignored. In addition, it is necessary to ensure that the sum of N confidences is 1. Here N can be set by the system. After the above steps, the mastery of each knowledge point can be calculated. Since there is a mutual dependence between knowledge points in the knowledge graph of step 1, when a knowledge point has not been examined, the mastery of another related knowledge point can be inferred through the knowledge point with known mastery. In this way, more problems of learners can be diagnosed with fewer questions.

[0051] In one embodiment, Figure 3 As shown, the knowledge directed path graph and the knowledge fuzzy points are associated to obtain recommended learning questions including:

[0052] Step 3001, splitting the knowledge directed path graph into a plurality of knowledge point sub-path graphs;

[0053] Step 3002, storing the topics mapped by the knowledge point sub-path graph in a topic list;

[0054] Step 3003, searching the question list for questions similar to the ambiguity in the knowledge, and obtaining directly recommended learning questions.

[0055] Among them, in the traditional method, the question directly associates all the knowledge points examined. In comparison, each question in the present invention is not directly associated with a knowledge point, but an associated knowledge point path. First, the problem-solving steps of each question are split. Since a problem-solving step represents a knowledge point, the entire problem-solving step can be represented as a directed path diagram connected by these knowledge points. Why should the direction be included? When doing questions, although the questions examine the same few knowledge points, the paths between the knowledge points are different, and the constructed question logic is different. For example, many learners understand a certain logical path of the knowledge point in the textbook examples, and can answer the same logical path questions correctly. However, once the logical path is turned over, many students can't turn their heads around and can't draw inferences from one example. Therefore, questions involving multiple knowledge points cannot only examine a single path, but also require multi-directional examinations. So, how can this question be represented by a knowledge point path? For example, the knowledge path of a question is (A, B, C), which can be split into two sub-paths (A, B) and (B, C). (A,B) represents a directed subpath from knowledge point A to knowledge point B. The same rules still apply to forked paths. For example Figure 2 , there is a subpath (D, B) in front of knowledge point B, which means that knowledge point B needs the output conditions of two knowledge points A and D to perform subsequent calculations. Here, (A, B) and (D, B) constitute two forked paths. Other forked path graphs also use the same splitting representation method. The split subpaths can be restored to a unique directed path graph in most cases. Only when the knowledge point appears multiple times at different path nodes of a question, it may not be possible to restore it to a unique directed path graph. However, such a situation does not affect the recommendation, after all, the similarity is still very high. In order to recommend similar test questions to learners, the traditional method is to directly recommend questions corresponding to the knowledge points related to the question. Such recommendations cannot be accurate and comprehensive. The present invention recommends questions based on the knowledge point path graph. So, how to store and search the directed graph of the question? We can represent the subpath graph split from each question as multiple 2-tuples, such as (A, B) and (C, D) above. These 2-tuples will point to a list of questions, and the question is inserted into the list. Finally, these 2-tuples are inserted into a large list for storage. When the system stores a large number of subpath graphs of questions in this way, each 2-tuple in the question list will contain multiple questions. This means that these questions all contain the same subpath of knowledge points. These questions can be sorted according to difficulty, question type, number of knowledge points, etc. for recommendation. See the storage method for details. Figure 3As shown. After storing the knowledge path of the question, you can search for similar questions. For example, a question contains two sub-path graphs (A, B) and (B, C). By finding these two sub-path graphs in the large list storing 2-tuples, you can get two lists of questions. By taking the union of these two lists, you can find all the questions that contain both sub-path graphs. Of course, you can also calculate the frequency of the questions in the two lists of questions, sort the questions according to the frequency, and the questions at the front are the ones with the highest similarity. Doing this can avoid the situation where recommendations cannot be made because it is impossible to find questions that completely match the directed path graph.

[0056] In one embodiment, Figure 4 As shown, the step of associating the knowledge directed path graph with the knowledge fuzzy points to obtain recommended learning questions further includes:

[0057] Step 3011, integrating the knowledge directed path graph into a knowledge graph, and obtaining the question examination probability through the knowledge graph;

[0058] Step 3012, calculating the question examination probability and the knowledge mastery to obtain the question recommendation priority;

[0059] Step 3013, obtaining indirectly recommended learning questions for knowledge points based on the question recommendation priority.

[0060] Among them, knowledge points are the basic units for learning and examining most subject knowledge. Using knowledge point analysis technology, we can obtain the degree of learners' mastery of knowledge. Knowledge points are the basic components of adaptive learning systems. The finer the knowledge point splitting, the more detailed the knowledge portrait of the learner, and the more accurate the corresponding recommended learning. The splitting of knowledge points needs to follow the principles of atomicity, non-overlapping and completeness. For example, mathematical knowledge points need to be split into formula definitions. The knowledge point map is a knowledge relationship map formed on the basis of the split knowledge points. Because any knowledge point is not isolated, there is a certain relationship between knowledge points. It can be a dependency relationship, such as the mastery of one knowledge point depends on the mastery of another knowledge point. For example, all theorems of Euclidean geometry are derived from several axioms. It can also be a correlation relationship, such as when one knowledge point appears, another knowledge point is likely to appear. Using the relationship between knowledge points to form a knowledge map, we can plan the learner's learning path and the path of examining knowledge. Then form a personalized test and learning plan. The knowledge map is not limited to a certain subject in a certain stage of study. In fact, there are also mutual relationships between knowledge points in different subjects in different stages of primary, middle and high school. Together, they can form a more comprehensive knowledge map. This allows learners to learn more comprehensively and analyze the relationship between weak points more thoroughly. Knowledge extraction mainly adopts manual methods, which requires professional discipline personnel to split and sort out the knowledge and sort out the smallest atomic knowledge points. According to different disciplines and knowledge relationship types, the knowledge representation model can be divided into relationship representation based on rule reasoning, such as the derivation relationship between mathematical formulas; relationship representation based on semantics, such as the relationship between objects and attributes; and relationship representation based on examination relevance, such as the relationship between multiple knowledge points examined in the question.

[0061] It should be understood that, although the various steps in the above-mentioned flow chart are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least a part of the steps in the above-mentioned flow chart may include a plurality of sub-steps or a plurality of stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these sub-steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the sub-steps or stages of other steps.

[0062] In one embodiment, Figure 5 As shown, this embodiment provides a learning system 400 for recommending knowledge, including: a knowledge path module 401, a knowledge diagnosis module 402, and a recommendation learning module 403, wherein:

[0063] The knowledge path module 401 is used to obtain the correct solution steps of the question, associate the knowledge points of the question with the correct solution steps, and connect the knowledge points to construct a knowledge directed path diagram;

[0064] The knowledge diagnosis module 402 is used to collect the learner's problem-solving information and obtain the learner's knowledge ambiguity points according to the problem-solving information;

[0065] The recommended learning module 403 is used to associate the knowledge directed path diagram with the knowledge fuzzy points to obtain recommended learning questions, and push the recommended learning questions to the learners.

[0066] In one embodiment, Figure 5 As shown, the knowledge diagnosis module 402 includes a knowledge analysis unit 4021, and the knowledge analysis unit 4021 is used to disassemble the problem-solving information to obtain incorrect problem-solving steps; obtain the confidence of a single knowledge point through the incorrect problem-solving steps, accumulate the confidence of the single knowledge point to construct the knowledge point mastery; and obtain the knowledge fuzzy point according to the knowledge point mastery.

[0067] In one embodiment, Figure 5 As shown, the recommendation learning module 403 includes a direct recommendation learning unit 4031 and an indirect recommendation learning unit 4032.

[0068] Among them, the direct recommendation learning unit 4031 is used to split the knowledge directed path graph into several knowledge point sub-path graphs, store the questions mapped by the knowledge point sub-path graphs in a question list, search for questions similar to the knowledge ambiguity points in the question list, and obtain direct recommendation learning questions.

[0069] The indirect recommendation learning unit 4032 is used to integrate the knowledge directed path graph into a knowledge graph, obtain the question examination probability through the knowledge graph, calculate the question examination probability and the knowledge mastery to obtain the question recommendation priority, and obtain the knowledge point indirect recommendation learning question according to the question recommendation priority.

[0070] For the specific definition of the learning system for recommended knowledge, please refer to the definition of the learning method for recommended knowledge in the above text, which will not be repeated here. Each module in the above-mentioned learning system for recommended knowledge can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0071] Figure 6FIG. 1 shows an internal structure diagram of a computer device in an embodiment. The computer device may specifically be a terminal or a server. Figure 6 As shown, the computer device includes a processor, a memory, a network interface, an input device and a display screen connected through a system bus. Among them, the memory includes a non-volatile storage medium and an internal memory. The non-volatile storage medium of the computer device stores an operating system and may also store a computer program. When the computer program is executed by the processor, the processor can implement the permission anomaly detection method. The internal memory may also store a computer program. When the computer program is executed by the processor, the processor can execute the permission anomaly detection method. The display screen of the computer device may be a liquid crystal display or an electronic ink display screen, and the input device of the computer device may be a touch layer covered on the display screen, or a button, trackball or touchpad provided on the computer device housing, or an external keyboard, touchpad or mouse, etc.

[0072] Those skilled in the art will understand that Figure 6 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0073] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the following steps are implemented: obtaining correct problem-solving steps for a question, associating the knowledge points of the question through the correct problem-solving steps, and connecting the knowledge points to construct a knowledge directed path graph; collecting learners' problem-solving information, and obtaining learners' knowledge fuzzy points based on the problem-solving information; associating the knowledge directed path graph with the knowledge fuzzy points to obtain recommended learning questions, and pushing the recommended learning questions to the learners.

[0074] In one embodiment, when the processor executes the computer program, the following steps are also implemented: disassembling the problem-solving information to obtain incorrect problem-solving steps; obtaining the confidence of a single knowledge point through the incorrect problem-solving steps, and accumulating the confidence of the single knowledge point to construct the knowledge point mastery; and obtaining the knowledge fuzzy point based on the knowledge point mastery.

[0075] In one embodiment, when the processor executes the computer program, the following steps are also implemented: splitting the knowledge directed path graph into several knowledge point sub-path graphs, storing the questions mapped by the knowledge point sub-path graphs in a question list, searching the question list for questions similar to the knowledge fuzzy points, and obtaining directly recommended learning questions.

[0076] In one embodiment, when the processor executes the computer program, the following steps are also implemented: integrating the knowledge directed path graph into a knowledge graph, obtaining the question examination probability through the knowledge graph, calculating the question examination probability and the knowledge mastery to obtain the question recommendation priority, and obtaining the knowledge point indirectly recommended learning questions based on the question recommendation priority.

[0077] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented: obtaining correct problem-solving steps for a question, associating the knowledge points of the question through the correct problem-solving steps, and connecting the knowledge points to construct a knowledge directed path diagram; collecting learners' problem-solving information, and obtaining learners' knowledge fuzzy points based on the problem-solving information; associating the knowledge directed path diagram with the knowledge fuzzy points to obtain recommended learning questions, and pushing the recommended learning questions to learners.

[0078] In one embodiment, when the processor executes the computer program, the following steps are also implemented: disassembling the problem-solving information to obtain incorrect problem-solving steps; obtaining the confidence of a single knowledge point through the incorrect problem-solving steps, and accumulating the confidence of the single knowledge point to construct the knowledge point mastery; and obtaining the knowledge fuzzy point based on the knowledge point mastery.

[0079] In one embodiment, when the processor executes the computer program, the following steps are also implemented: splitting the knowledge directed path graph into several knowledge point sub-path graphs, storing the questions mapped by the knowledge point sub-path graphs in a question list, searching the question list for questions similar to the knowledge fuzzy points, and obtaining directly recommended learning questions.

[0080] In one embodiment, when the processor executes the computer program, the following steps are also implemented: integrating the knowledge directed path graph into a knowledge graph, obtaining the question examination probability through the knowledge graph, calculating the question examination probability and the knowledge mastery to obtain the question recommendation priority, and obtaining the knowledge point indirectly recommended learning questions based on the question recommendation priority.

[0081] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing related hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods.

[0082] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0083] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A method for learning recommended knowledge, characterized in that: The method comprises the following steps: Obtaining correct steps for solving the problem, associating the knowledge points of the problem through the correct steps, and connecting the knowledge points to construct a knowledge directed path diagram; Collecting learners' problem-solving information, and obtaining learners' knowledge ambiguities based on the problem-solving information; The knowledge directed path graph and the knowledge fuzzy point are associated to obtain recommended learning questions, and the recommended learning questions are pushed to the learners; specifically, the method includes: integrating the knowledge directed path graph into a knowledge graph, obtaining the question examination probability through the knowledge graph, calculating the question examination probability and the knowledge mastery to obtain the question recommendation priority, and obtaining the knowledge point indirect recommended learning questions according to the question recommendation priority; The acquiring of the learner's knowledge ambiguity points according to the problem-solving information comprises: Disassembling the problem-solving information to obtain incorrect problem-solving steps; The confidence of a single knowledge point is obtained through the wrong problem-solving steps, and the confidence of the single knowledge point is accumulated to construct the knowledge point mastery; the knowledge point mastery is expressed by the following formula: Among them, M is the mastery of the knowledge point, mi is the mastery of the knowledge point in a single test, di is the difficulty factor corresponding to the question, and ci is the confidence level of a single test; the knowledge fuzzy point is obtained according to the mastery of the knowledge point.

2. The method for learning recommended knowledge according to claim 1, characterized in that: The method of associating the knowledge directed path graph and the knowledge fuzzy point to obtain recommended learning questions includes: splitting the knowledge directed path graph into a plurality of knowledge point sub-path graphs, storing the questions mapped by the knowledge point sub-path graphs in a question list, searching the question list for questions similar to the knowledge fuzzy point, and obtaining directly recommended learning questions.

3. A learning system for recommending knowledge, characterized in that: include: A knowledge path module is used to obtain the correct solution steps of the question, associate the knowledge points of the question with the correct solution steps, and connect the knowledge points to construct a knowledge directed path diagram; A knowledge diagnosis module is used to collect the learner's problem-solving information and obtain the learner's knowledge ambiguity based on the problem-solving information; the knowledge diagnosis module includes a knowledge analysis unit, which is used to disassemble the problem-solving information to obtain incorrect problem-solving steps; The confidence of a single knowledge point is obtained through the wrong problem-solving steps, and the confidence of the single knowledge point is accumulated to construct the knowledge point mastery; the knowledge fuzzy point is obtained according to the knowledge point mastery; the knowledge point mastery is expressed by the following formula: Among them, M is the mastery of the knowledge point, mi is the mastery of the knowledge point in a single test, di is the difficulty factor corresponding to the question, and ci is the confidence level of a single test; A recommended learning module is used to associate the knowledge directed path diagram with the knowledge fuzzy points to obtain recommended learning questions, and push the recommended learning questions to learners; the recommended learning module includes an indirect recommendation learning unit, which is used to integrate the knowledge directed path diagram into a knowledge graph, obtain the question examination probability through the knowledge graph, calculate the question examination probability and the knowledge mastery to obtain the question recommendation priority, and obtain the knowledge point indirect recommended learning questions according to the question recommendation priority.

4. The learning system for recommending knowledge according to claim 3, characterized in that: The recommendation learning module includes a direct recommendation learning unit, which is used to split the knowledge directed path graph into several knowledge point sub-path graphs, store the questions mapped by the knowledge point sub-path graphs in a question list, search the question list for questions similar to the knowledge ambiguity points, and obtain direct recommendation learning questions.

5. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 2 are implemented.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 2 are implemented.