Methods, devices, and electronic equipment for determining learning content based on a smart education platform

Abstract

This disclosure presents a method, apparatus, and electronic device for determining learning content based on a smart education platform, relating to the field of data processing technology. The method includes: extracting clusters of basic and key knowledge points from student textbooks to generate a target question bank containing questions of varying difficulty and types; generating a knowledge point association network based on the correlation and strength between knowledge points corresponding to questions in the target question bank; generating the target association strength between target knowledge points by combining the student's learning trajectory on a learning map; updating the knowledge point association network accordingly to obtain a student-specific target association network; and then updating the target question bank to determine the user's target learning content. This solution solves the problem of single question types in existing technologies by extracting knowledge points from textbooks and generating a target question bank with diverse question types. By combining learning trajectories and association networks to achieve dynamic updates of the question bank, it allows students to focus on core knowledge points and improves learning efficiency.

Description

Technical Field

[0001] This disclosure relates to the field of data processing technology, and in particular to a method, apparatus, and electronic device for determining learning content based on a smart education platform. Background Technology

[0002] Practice software typically includes a large number of practice questions, mock tests, and detailed answer explanations, which can help users improve their learning efficiency.

[0003] Currently, the main approach involves importing history exam questions, generating questions based on the corresponding knowledge points and question types, creating a question bank for students, and then determining the learning content based on the questions in the question bank to help students study the knowledge points and reminders to be tested.

[0004] However, the question bank obtained in this way contains a large number of questions for each knowledge point, which means that students can only learn the questions for each knowledge point. This results in students learning a relatively limited range of question types for each knowledge point. Furthermore, generating a large number of questions based on each knowledge point also leads to students learning knowledge points in a scattered manner during the learning process, making it difficult to focus on specific knowledge points and thus resulting in low learning efficiency. Summary of the Invention

[0005] In view of this, this disclosure provides a method, device, storage medium, and electronic device for determining learning content based on a smart education platform. The main purpose is to improve the existing question bank, where each knowledge point corresponds to a large number of questions of the current examination question type. This results in students only being exposed to a single examination question type for that knowledge point and failing to fully grasp the diverse examination formats of the knowledge point. At the same time, generating a large number of questions based on each knowledge point makes the knowledge points in the question bank scattered. Students have to face scattered knowledge point questions during the learning process, making it difficult to focus on core knowledge points for systematic learning, ultimately resulting in low learning efficiency.

[0006] Firstly, this disclosure provides a method for determining learning content based on a smart education platform, including:

[0007] The target question bank is formed by extracting basic knowledge point clusters and key knowledge point clusters from student textbooks and generating questions of different difficulty levels and question types based on these basic knowledge point clusters and key knowledge point clusters.

[0008] Based on the relationship and strength of association between the knowledge points corresponding to each question in the target question bank, a knowledge point association network corresponding to the target question bank is generated.

[0009] Based on the student's learning trajectory in the learning map, the target association strength between the target knowledge points learned by the student based on the learning map is generated. The learning map is obtained by mapping the basic knowledge point cluster and the key knowledge point cluster in multiple learning grids arranged according to a predetermined relationship.

[0010] The knowledge point association network is updated based on the target association strength to obtain the target knowledge point association network corresponding to the student, and the target question bank is updated based on the target knowledge point association network. The target question bank is used to determine the user's target learning content.

[0011] Optionally, generating a knowledge point association network corresponding to the target question bank based on the association relationships and association strengths between the knowledge points corresponding to each question in the target question bank includes:

[0012] Based on the knowledge point combination corresponding to each question in the target question bank, a set of knowledge point combinations corresponding to the target question bank is determined, and the relationship between knowledge points is determined based on the set of knowledge point combinations.

[0013] Determine the number of questions corresponding to each knowledge point in the knowledge point combination and the number of combinations with other knowledge points to form a knowledge point combination;

[0014] Based on the target decay parameter at the current moment, a weighted analysis is performed on the number of questions and the number of combinations to obtain the correlation strength between the knowledge points corresponding to the knowledge point combination set at the current moment.

[0015] Based on the relationships and strengths of association between the knowledge points corresponding to the knowledge point combination set, a knowledge point association network corresponding to the target question bank is generated.

[0016] Optionally, before generating the target association strength between the target knowledge points learned by the student based on the student's learning trajectory in the learning map, the method further includes:

[0017] The questions corresponding to the basic knowledge point clusters in the target question bank are marked with a first identifier, and the questions corresponding to the key knowledge point clusters are marked with a second identifier.

[0018] The questions marked with the first identifier and the questions marked with the second identifier are mapped to multiple learning grids in the preset map according to the chapter order in the textbook to obtain the learning map. Each learning grid in the learning map corresponds to all the questions of a knowledge point cluster. The learning grid corresponding to the questions marked with the second identifier is the key learning grid.

[0019] Optionally, generating the target association strength between target knowledge points learned by the student based on the student's learning trajectory in the learning map includes:

[0020] Extract the student's learning trajectory from the learning map, and select multiple first target knowledge point combinations corresponding to the learning trajectory from the knowledge point combination set;

[0021] Determine the knowledge point transfer relationship corresponding to the multiple first target knowledge point combinations, and determine the association strength increment corresponding to the multiple first target knowledge point combinations based on the transfer relationship;

[0022] The target correlation strength is determined based on the correlation strength increment and the correlation strength.

[0023] Optionally, updating the target question bank based on the target knowledge point association network includes:

[0024] From the target knowledge point association network, determine multiple combinations of second target knowledge points whose target association strength meets the association conditions;

[0025] Select questions corresponding to the multiple combinations of second target knowledge points from the target question bank, and determine the selected questions as the updated question bank of the target question bank.

[0026] Optionally, after updating the knowledge point association network based on the target association strength to obtain the target knowledge point association network corresponding to the student, and updating the target question bank based on the target knowledge point association network, the method further includes:

[0027] A correlation evaluation is conducted based on the student's learning materials and the combination of multiple second-target knowledge points;

[0028] Based on the relevance evaluation results, target learning materials that meet the relevance criteria are determined from the students' learning materials, and these target learning materials are identified as the students' target learning content.

[0029] The system generates recommendation information for the target learning content, which is used to recommend the target learning content to the student.

[0030] Secondly, this disclosure provides a learning content determination device based on a smart education platform, comprising:

[0031] The extraction module is configured to extract basic knowledge point clusters and key knowledge point clusters from student textbooks, and generate a target question bank based on the basic knowledge point clusters and key knowledge point clusters with different difficulty levels and question types.

[0032] The generation module is configured to generate a knowledge point association network corresponding to the target question bank based on the association relationship and association strength between the knowledge points corresponding to each question in the target question bank.

[0033] The generation module is also configured to generate the target association strength between the target knowledge points learned by the student based on the learning trajectory in the learning map, wherein the learning map is obtained by mapping the basic knowledge point cluster and the key knowledge point cluster in multiple learning grids arranged according to a predetermined relationship.

[0034] The update module is configured to update the knowledge point association network based on the target association strength to obtain the target knowledge point association network corresponding to the student, and update the target question bank based on the target knowledge point association network, wherein the target question bank is used to determine the user's target learning content.

[0035] Thirdly, this disclosure provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method described in the first aspect.

[0036] Fourthly, this disclosure provides an electronic device including a storage medium, a processor, and a computer program stored on the storage medium and executable on the processor, wherein the processor executes the computer program to implement the method described in the first aspect.

[0037] Fifthly, this disclosure provides a computer program product comprising a computer program that, when executed by a processor, implements the method described in the first aspect.

[0038] By utilizing the above technical solutions, this disclosure provides a method, apparatus, storage medium, and electronic device for determining learning content based on a smart education platform. Compared with existing technologies, this disclosure generates a target question bank with multiple difficulty levels and question types by extracting basic and key knowledge points from student textbooks. It constructs a correlation network by combining the relationships and strengths of knowledge points in the question bank, generates target correlation strength based on the student's learning trajectory in the learning map, and updates the correlation network and target question bank to determine the target learning content. This allows this disclosure to extract knowledge points from textbooks and generate a target question bank with multiple question types, solving the problem of single question types in existing technologies. By combining the learning trajectory and correlation network to achieve dynamic updates of the question bank, students can focus on core knowledge points and improve learning efficiency. Attached Figure Description

[0039] The accompanying drawings, which are incorporated in and form a part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure.

[0040] To more clearly illustrate the technical solutions in the embodiments of this disclosure or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, those skilled in the art can obtain other drawings based on these drawings without creative effort.

[0041] Figure 1 A flowchart illustrating a method for determining learning content based on a smart education platform, provided in an embodiment of this disclosure, is shown.

[0042] Figure 2 A schematic diagram illustrating an example of a knowledge point network provided in an embodiment of this disclosure is shown;

[0043] Figure 3 A flowchart illustrating a method for determining learning content based on a smart education platform, provided in an embodiment of this disclosure, is shown.

[0044] Figure 4 A schematic diagram illustrating an example of a basic knowledge point cluster and an important knowledge point cluster provided in an embodiment of this disclosure is shown;

[0045] Figure 5 A schematic diagram of an example learning map provided in an embodiment of this disclosure is shown;

[0046] Figure 6 A schematic diagram of the structure of a learning content determination device based on a smart education platform provided in an embodiment of this disclosure is shown.

[0047] Figure 7 A schematic diagram of the structure of an electronic device provided in an embodiment of this disclosure is shown. Detailed Implementation

[0048] Embodiments of this disclosure will now be described in more detail with reference to the accompanying drawings. It should be noted that, unless otherwise specified, the embodiments and features described herein can be combined with each other.

[0049] To address the technical problem that existing question banks only provide a large number of questions corresponding to the current question type for each knowledge point, resulting in students only encountering a single question type for that knowledge point and failing to fully grasp the diverse testing formats for that knowledge point, and that generating a large number of questions based on each knowledge point leads to a scattered distribution of knowledge points in the question bank, students face fragmented questions during the learning process, making it difficult to focus on core knowledge points for systematic learning, ultimately resulting in low learning efficiency, this embodiment provides a method for determining learning content based on a smart education platform, such as... Figure 1 As shown, the method includes:

[0050] Step 101: Extract basic knowledge point clusters and key knowledge point clusters from student textbooks, and generate a target question bank with questions of different difficulty levels and question types based on the basic knowledge point clusters and key knowledge point clusters.

[0051] In this embodiment of the disclosure, the basic knowledge point cluster can be the core basic content that constitutes the subject knowledge system. It is usually classified according to the structure of textbook chapters, units, etc., and extracted from the textbook through technologies such as text mining, natural language processing, and semantic analysis, and formed into an ordered set in the form of hierarchical identification and coding.

[0052] For example, in textbook A, if the cluster of basic knowledge points includes algebraic operations (a1) and geometric proofs (b1), the cluster can be represented as Kbase={algebraic operations (a1), geometric proofs (b1)}. For example, the algebraic operations (a1) cluster can cover basic knowledge points such as integer addition, subtraction, multiplication, and division, fraction operations, and decimal operations; the geometric proofs (b1) cluster can include basic knowledge points such as line segment equality proofs, angle equality proofs, and triangle congruence criteria.

[0053] In this embodiment of the disclosure, the key knowledge point cluster can be obtained by a knowledge point extraction module based on the distribution of knowledge point clusters. For example, in textbook A, if the key knowledge point cluster may include function knowledge points (c2), probability knowledge points (c3), etc., then the key knowledge point cluster can be represented as Kimp={function(c2), probability(c3)}. Here, function knowledge points (c2) may include knowledge points on the definition of functions, graph knowledge points, property knowledge points, and solution knowledge points, etc., and probability knowledge points (c3) may include knowledge points on the definition of probability, event classification knowledge points, and random event probability knowledge points, etc.

[0054] In this embodiment, the target question bank can be a collection of questions generated based on basic knowledge point clusters and key knowledge point clusters, containing different difficulty levels and diverse question types. It can be used to provide students with targeted practice materials to help them consolidate knowledge points and become familiar with different examination formats. The difficulty level can usually correspond to the hierarchical identification of knowledge points, and the question types can cover common examination types such as multiple choice, fill-in-the-blank, problem solving, and proof.

[0055] For example, if the basic knowledge point cluster includes {a1,b1} and the key knowledge point cluster includes {c2,c3}, the target question bank may include question type 1: examining {a1,b1} (algebraic operations and geometric proofs); question type 2: examining {a1,b1,c2} (new functions); and question type 3: examining {b1,c3} (geometric proofs and probability).

[0056] In this embodiment, the basic knowledge point cluster can be generated by uploading student textbook content to the knowledge point analysis module, extracting knowledge points using text mining and other technologies, and then classifying and encoding them; the key knowledge point cluster can be obtained by statistically analyzing the number of knowledge point identifier vectors at each level, combining the frequency of examination in historical question banks, and the accuracy of answering questions; correspondingly, based on the basic knowledge point cluster and the key knowledge point cluster, the target question bank can be generated by classifying and combining the knowledge points in the basic knowledge point cluster and the key knowledge point cluster according to difficulty, and then generating questions of different difficulties and question types.

[0057] In some examples, the process of extracting basic knowledge point clusters from student textbooks may include: uploading textbook content to the first knowledge point analysis module to obtain basic knowledge point clusters; extracting knowledge points from the text and classifying them using text mining, natural language processing, and semantic analysis techniques through text analysis of the textbook content; obtaining identifier encoding values ​​by encoding the identifiers; and constructing an identifier vector set Uori (i.e., the basic knowledge point clusters in this embodiment) based on the identifier encoding values ​​to generate a knowledge point network; the hierarchical identification can divide the knowledge point network into n levels, for example, n=3, then U=[An, Bm, Ck], with the general expression: U=(k, β), where k is the knowledge point set and β is the edge set, such as... Figure 2 As shown: the first-level category An is the master node, which can be classified by chapter, resulting in An=[A1, A2...An]; the second-level category Bm is the slave node of An, which can be classified by section, resulting in Bm=[A1b1, A1b2, A2b1..Anbm]; the third-level category Ck is the slave node of Bm, which can be classified by unit result, resulting in Ck=[A1b1c1, A1b2c2, A1b2c3..Anbmck].

[0058] It should be noted that, since knowledge points appear in single or multi-unit combinations depending on their difficulty and learning depth, when constructing the knowledge point network, the existence of the first-level category An can be represented based on the identifier vector of a single unit, while the second-level and third-level categories can be represented based on the combination of multiple units. Identifier vectors are then sequentially superimposed on these to obtain multiple knowledge point networks forming a knowledge point cluster. By constructing the knowledge point network coverage through the method of this embodiment, the knowledge point network of this embodiment can cover all combinations of knowledge points. Therefore, in actual question delivery, questions corresponding to the first-level identifier can be delivered when the difficulty level is easy, questions corresponding to the second-level identifier can be delivered when the difficulty level is medium, and questions corresponding to the third-level identifier can be delivered when the difficulty level is difficult.

[0059] Step 102: Based on the relationship and strength of the association between the knowledge points corresponding to each question in the target question bank, generate the knowledge point association network corresponding to the target question bank.

[0060] In this embodiment of the disclosure, the knowledge point association relationship can be the inherent connection between the knowledge points tested in different questions in the target question bank. One knowledge point can appear in the same question along with multiple other knowledge points. For example, in the target question bank of textbook A, if question 1 tests algebraic operations (a1) and geometric proofs (b1), question 2 tests geometric proofs (b1) and functions (c2), and question 3 tests algebraic operations (a1) and functions (c2), then by analyzing these questions, it can be determined that there is an association between algebraic operations (a1) and geometric proofs (b1) and functions (c2), and there is also an association between geometric proofs (b1) and functions (c2).

[0061] In this embodiment of the disclosure, the knowledge point association strength can be a quantitative description of the closeness of the knowledge point association relationship, and can be used to distinguish the importance of different association relationships.

[0062] In this embodiment of the disclosure, the knowledge point association network can be a knowledge graph constructed by taking all knowledge points in the target question bank as nodes, the association relationship between knowledge points as edges, and the association strength as the weight of the edges, which can intuitively display the association between knowledge points.

[0063] For example, if the basic knowledge point cluster is represented as Kbase={k1,k2,…,km}, and the important knowledge point cluster is represented as Kimp={kimp1,kimp2,…,kimpn}, Python and Natural Language Processing (NLP) techniques can be used to label and classify the questions in the question bank to obtain the question set represented as Q={q1,q2,…,qp}. Then, the combination of knowledge points examined by each question qi can be denoted as Ci belonging to (Kbase∪Kimp). The node set V=Kbase∪Kimp, and each edge eij in the edge set E is associated with a weight w. ij Based on the knowledge points and their correlation strength in junior high school mathematics, a knowledge point correlation network is constructed. For example, if a compound question tests knowledge points (A, B) and a student answers incorrectly, the network can retrieve review materials for Aa1 (strongly correlated with knowledge point A) and Bb2 (strongly correlated with knowledge point B), or other strongly correlated combinations of knowledge points A and B such as (AAa1Bb2). This allows for adaptive recommendation of review materials based on the student's performance.

[0064] In this embodiment of the disclosure, the relationships between knowledge points are determined, the relationship strength is calculated, and basic knowledge points and key knowledge points are used as nodes, the relationships are used as edges, and the relationship strength is used as the weight of the edges to form a knowledge point association network, which can complete the association integration of knowledge points in the question bank.

[0065] Step 103: Based on the student's learning trajectory in the learning map, generate the target association strength between the target knowledge points learned by the student based on the learning map. The learning map is obtained by mapping the basic knowledge point clusters and key knowledge point clusters in multiple learning grids arranged according to a predetermined relationship.

[0066] In this embodiment of the disclosure, the learning map may be composed of multiple learning grids arranged according to a predetermined relationship. Each grid corresponds to questions in the target question bank, which is generated based on knowledge point clusters extracted from student textbooks. It should be noted that each learning grid corresponds to all questions of a knowledge point cluster. Grids corresponding to key knowledge point clusters can be key learning grids. Selecting adjacent key learning grids can yield key learning areas.

[0067] In some examples, mapping questions from the question bank onto a learning map allows students to clearly see their learning progress and content, as well as the knowledge points in the textbook corresponding to the current learning content and their importance, when viewing the learning map.

[0068] In this embodiment, a learning grid forms a learning map, recording the order in which students answer questions on the learning map, determining the knowledge point combinations for each question, and forming a learning trajectory. The correlation strength between knowledge points is calculated for the transition relationships between each learning trajectory.

[0069] For example, the generation of target association strength can be achieved in the following ways, but not limited to: If the textbook is divided into chapters, the first grid corresponds to the questions a1, the second grid corresponds to the questions b1, and the third grid corresponds to the questions c2, and a learning map is constructed, if student A's learning trajectory in the learning map is {a1,b1}→{a1,b1,c2}→{b1,c3}, then the transition relationships of adjacent combinations in the trajectory can be extracted as {a1,b1}→{a1,b1,c2} and {a1,b1,c2}→{b1,c3}. Among them, the common knowledge points of {a1,b1}→{a1,b1,c2} are a1 and b1. The initial association strength between the common knowledge points a1 and b2 and the newly added knowledge point c2 can be increased by an increment δ. If δ=0.05, then the initial strength of a1 and c2 is 1.2, which increases by 0.05, and becomes 1.25.

[0070] Step 104: Update the knowledge point association network based on the target association strength to obtain the target knowledge point association network corresponding to the student, and update the target question bank based on the target knowledge point association network. The target question bank is used to determine the user's target learning content.

[0071] In this embodiment, the target knowledge point association network can be a network specific to that student, formed by replacing the corresponding edge weights in the initial knowledge point association network with the student's target association strength. The updated target question bank can be a question bank adapted to the student's needs, formed by filtering questions corresponding to combinations of knowledge points whose target association strength meets a threshold, based on the target knowledge point association network. The target learning content can be the key learning content determined based on the updated target question bank and the knowledge points with high error rates among students.

[0072] In this embodiment of the disclosure, based on the example of step 103, the knowledge point association network is updated with the generated target association strength, which can ensure that the updated knowledge point association network fits the individual learning situation of students.

[0073] For example, based on the example in step 103, the calculated target association strengths a1 and c2 are 1.25, which replaces the initial weights a1 and c2 of 1.2 in the initial association network, to obtain the target knowledge point association network.

[0074] The system selects combinations of knowledge points from the target knowledge point association network, and then selects questions corresponding to the knowledge point combinations from the initial target question bank to form updated target questions.

[0075] For example, an association strength threshold of 1.0 can be set, and a1 and c2, a1 and b1 can be filtered out based on the association strength threshold of 1.0. Then, questions corresponding to these combinations can be selected from the initial target question bank, such as function comprehensive questions that examine a1+c2 and algebraic geometry questions that examine a1+b1, to form an updated target question bank.

[0076] In this embodiment of the disclosure, by analyzing the general situation of students after updating the target question bank, the knowledge points with high error rates are identified. Combined with the target knowledge point association network, questions and review materials with high relevance to the knowledge points with high error rates can be recommended.

[0077] For example, if we analyze students' answers in the updated question bank and identify knowledge points with high error rates as knowledge point c2 and knowledge point c3, and combine this with the target knowledge point association network, we can recommend learning material R5 for knowledge point c2 and learning material R4 for knowledge point c3. Finally, we can determine that the students' target learning content may include learning material R5 and learning material R4.

[0078] Compared with existing technologies, the embodiments of this disclosure generate a target question bank with multiple difficulty levels and question types by extracting basic knowledge points and key knowledge point clusters from student textbooks, constructing an association network by combining the relationship and strength of knowledge points in the question bank, generating target association strength based on the student's learning trajectory in the learning map, updating the association network and target question bank to determine the target learning content, enabling students to focus on core knowledge points and improve learning efficiency.

[0079] When performing the task of "generating a knowledge point association network corresponding to the target question bank based on the association relationships and association strengths between the knowledge points corresponding to each question in the target question bank," the following methods can be used, but are not limited to them, such as... Figure 3 As shown, the method includes:

[0080] Step 201: Based on the knowledge point combination corresponding to each question in the target question bank, determine the knowledge point combination set corresponding to the target question bank, and determine the relationship between knowledge points based on the knowledge point combination set.

[0081] In this embodiment of the disclosure, the knowledge point combination can be a set of one or more knowledge points tested by each question in the target question bank, and the knowledge point combination set can be a set formed by summarizing and deduplicating the knowledge point combinations corresponding to all questions in the target question bank.

[0082] In some examples, if analyzing each knowledge point combination in the knowledge point combination set, there are two knowledge points that appear in at least one combination at the same time, then it is determined that there is a relationship between the two.

[0083] For example, in the target question bank, if question type 1 corresponds to the knowledge point combination {a1,b1}, question type 2 corresponds to the knowledge point combination {a1,b1,c2}, and question type 3 corresponds to the knowledge point combination {b1,c3}, after removing duplicate combinations, a knowledge point combination set {{a1,b1},{a1,b1,c2},{b1,c3}} can be formed. Analyzing each combination in the knowledge point combination set, since knowledge point a1 and knowledge point b1 appear in both knowledge point combination {a1,b1} and knowledge point combination {a1,b1,c2}, it can be determined that there is a relationship between them; similarly, if knowledge point b1 and knowledge point c3 appear in both knowledge point combination {b1,c3}, there is also a relationship.

[0084] In some examples, the process of extracting key knowledge point clusters from student textbooks may include: obtaining important knowledge point clusters through a knowledge point extraction module based on the distribution of knowledge point clusters U; such as... Figure 4 As shown, the number of first-level branch identifier vectors Z, the number of second-level branch identifier vectors X, and the number of third-level branch identifier vectors V can be counted. These values ​​are then input into a pre-trained scoring model G to calculate the knowledge point cluster score, as shown in Formula 1 below.

[0085] (Formula 1)

[0086] In Formula 1, This represents the weight score indicating the importance of the corresponding indicator for Z. This represents the weight score indicating the importance of the indicator corresponding to X. The weight score representing the importance of the corresponding indicator V is determined by the fact that the more detailed the knowledge points are, the greater the weight of importance. Therefore, we can determine that β1 < β2 < β3. Set the filtering threshold T1. When G > T1, the filtered knowledge point identifier vector will be used as the first important knowledge point cluster Unew. The threshold T can be set using custom, average, median, or other methods.

[0087] Furthermore, new clusters of important knowledge points are generated through screening using a pre-trained key knowledge point evaluation model. This includes: acquiring a historical question bank; constructing a set of identifier vectors G corresponding to the knowledge point identifiers using the same steps as extracting basic knowledge point clusters from student textbooks; generating a knowledge point network; obtaining vector sets Wn, Rm, and Hk using a hierarchical identifier method, where Wn is the master node, Rm is the slave node of Wn, and Hk is the slave node of Rm, with Zori = [Wn, Rm, Hk]; and finally, screening using the pre-trained key knowledge point evaluation model to obtain the key knowledge point evaluation model, as shown in Formula 2 below.

[0088] (Formula 2)

[0089] In Formula 2, Gj represents the number of times the test knowledge point appears; Gq represents the number of errors in the test knowledge point; Gr represents the number of branches of the test knowledge point; α represents the weight score of the corresponding indicator; a scoring threshold T2 is set, and when yG>T2, the key knowledge point combination is selected to form the second important knowledge point cluster Znew.

[0090] Step 202: Determine the number of questions corresponding to each knowledge point in the knowledge point combination and the number of combinations with other knowledge points to form a knowledge point combination.

[0091] In this embodiment of the disclosure, the number of questions corresponding to a knowledge point can be the total number of all questions in the target question bank that test that knowledge point. For example, the number of questions containing knowledge point ki or kj up to time t.

[0092] In this embodiment of the disclosure, the number of knowledge point combinations can be the number of times that knowledge point appears in the same combination with another knowledge point in the target question bank. For example, the number of questions that simultaneously contain knowledge points ki and kj up to time t.

[0093] Step 203: Perform a weighted analysis on the number of questions and the number of combinations based on the target decay parameter at the current moment to obtain the correlation strength between the knowledge points corresponding to the knowledge point combination set at the current moment.

[0094] In this embodiment, the target attenuation parameter can be a coefficient used to reduce the influence of historical data. It can be substituted into the association strength calculation formula to calculate the association strength, so that the association strength can adapt to the dynamic updates of the question bank. For example, the target attenuation parameter can be the historical attenuation function α, where 0 < α < 1.

[0095] In this embodiment of the disclosure, the weighted analysis can be a process of combining the number of questions, the number of combinations, and the target attenuation parameter, and then substituting them into the association strength update formula for calculation.

[0096] For example, when performing weighted analysis, the target attenuation parameter α=0.6, the number of combinations Nij=2, the number of questions Ni+Nj=3, and the number of knowledge graph branches Bi+Bj=5 are first substituted into Formula 1 for calculation. wij≈2.0, and the correlation strength between a1 and b1 at the current time is approximately 2.0, thus realizing the dynamic calculation of correlation strength.

[0097] In some examples, a knowledge point association network containing the strength of knowledge point associations can be obtained using Formula 3, as shown below:

[0098] (Formula 3)

[0099] In Formula 3, w ij (t) α can represent the association strength at time t (initial value is 0); α can represent the historical weight decay coefficient (0 < α < 1), for example, 0.6, which represents the decay of historical association; β, γ, and η can represent adjustment coefficients; N ij (t) This can represent the deadline t, and the number of questions in the question bank that simultaneously contain knowledge points ki and kj; N i (t) N j (t) It can represent the deadline t, and the number of questions containing knowledge points ki or kj; R i (t) R j (t) It can represent the deadline t, which includes the number of branches of knowledge point ki or kj in the knowledge graph.

[0100] As an alternative approach, in a practice problem-solving scenario, the stronger the correlation between multiple knowledge points, the more important the combination of knowledge points is and the more they belong to the same link. Therefore, based on the combination of knowledge points with strong correlation, review materials that are suitable for the user's problem-solving situation can be recommended.

[0101] For example, taking algebraic operations (a1) and basic functions (c2) as examples, as of the current time t, if the number of questions in the question bank that simultaneously contain both knowledge points (Nij) is 12, the number of questions that contain either algebraic operations (a1) or basic functions (c2) (Ni+Nj) is 35, the number of branches of algebraic operations (a1) and basic functions (c2) in the knowledge graph (Bi+Bj) is 6, the historical weight decay coefficient α=0.6, and the adjustment coefficient λ=0.8, substituting into the formula for calculating the correlation strength, we can calculate that the correlation strength between the two is wij=0.6×0.8×(12 / 35)×6≈0.99. The higher this value, the stronger the correlation between algebraic operations (a1) and functions (c2).

[0102] Step 204: Based on the relationships and strengths of associations between the knowledge points in the knowledge point combination set, generate the knowledge point association network corresponding to the target question bank.

[0103] In this embodiment of the disclosure, the relationships and strengths between knowledge points corresponding to the knowledge point combination set are integrated to construct a knowledge point association network specific to the target question bank. Specifically, this can be done by: using basic and key knowledge point clusters as network nodes; using relationships as edges of the knowledge point association network; and using the strength of relationships as the weights of the edges to form a knowledge point association network corresponding to the target question bank.

[0104] For example, if the basic knowledge point cluster is Kbase={a1,b1}, and the key knowledge point clusters are Kimp={c2,c3}, then the basic knowledge point cluster and the key knowledge point clusters Kbase={a1,b1} and Kimp={c2,c3} can be used as network nodes V; the related knowledge points a1 and b1, and b1 and c3 can be used as network edges E; and the association strength w can be used as the network edge V. a1-b1 =2.0、w b1-c3 =1.5 is used as the weight of the edge to form a knowledge point association network corresponding to the target question bank.

[0105] Before executing the step of "generating the target association strength between target knowledge points learned by students based on their learning trajectories in the learning map", the following method can be used, but is not limited to: marking the questions corresponding to basic knowledge point clusters in the target question bank with a first identifier and marking the questions corresponding to key knowledge point clusters with a second identifier; mapping the questions marked with the first identifier and the questions marked with the second identifier to multiple learning grids in the preset map according to the order of chapters in the textbook to obtain a learning map. Each learning grid in the learning map corresponds to all the questions of a knowledge point cluster, and the learning grid corresponding to the questions marked with the second identifier is the key learning grid.

[0106] In this embodiment of the disclosure, the first identifier may be a mark (such as a specific color, symbol or code) used to distinguish the questions corresponding to the basic knowledge point clusters, which can be used to quickly identify the basic knowledge point category to which the question belongs, and facilitate subsequent mapping to the learning map; the second identifier may be a mark used to distinguish the questions corresponding to the key knowledge point clusters. Unlike the first identifier, the second identifier can be used to clarify the key knowledge point category to which the question belongs, highlighting the core examination content.

[0107] For example, in the junior high school mathematics target question bank, the questions corresponding to the basic knowledge point cluster Kbase={algebraic operations(a1), geometric proof(b1)} are marked in white (first identifier); the questions corresponding to the key knowledge point cluster Kimp={function(c2), probability(c3)} are marked in black (second identifier). The knowledge point type corresponding to the question can be intuitively distinguished by color marking.

[0108] In the embodiments disclosed herein, such as Figure 5 As shown, the preset map can be a pre-designed basic framework containing multiple blank learning grids, with the grid arrangement order consistent with the chapter order of the textbook; the learning grids in the preset map can be independent units in the preset map, and each grid can correspond to all the questions of a knowledge point cluster, which can achieve a precise correspondence between questions and knowledge points; the key learning grids can be the learning grids corresponding to the second identifier (key knowledge point questions), which can be used to prompt students to focus on key learning areas and core test points.

[0109] For example, taking the order of Chapter A in the textbook—Chapter 1 Algebraic Operations, Chapter 2 Geometric Proofs, Chapter 3 Functions, and Chapter 4 Probability—as an example, four learning grids can be divided in the preset map:

[0110] First learning grid (corresponding to Chapter 1): Maps all algebraic operation (a1) problems (such as integer operation problems and fraction operation problems) marked in white (first identifier). This grid is a normal learning grid.

[0111] The second learning grid (corresponding to Chapter 2): all geometric proofs (b1) problems marked with white (first identifier) ​​(such as line segment equality proofs, triangle congruence proofs), this grid is the ordinary learning grid;

[0112] The third learning grid (corresponding to Chapter 3): Maps all problems of functions (c2) marked in black (second identifier) ​​(such as problems of linear function graphs and quadratic function properties). This grid is the key learning grid.

[0113] The fourth learning grid (corresponding to Chapter 4): all problems (such as random event probability calculation problems and probability application problems) of the black map marker (second identifier) ​​in the C3 grid. This grid is the key learning grid.

[0114] In this embodiment, a learning grid refers to a small cell or node on a learning map, with each learning grid corresponding to a question in the target question bank. The learning map is constructed according to the number and difficulty level of questions in the target question bank, and each question has a corresponding position on the learning map, which is the learning grid. The learning grid not only represents the position of the question on the learning map but also contains relevant information about the question, such as its difficulty level. When the learning grids corresponding to the questions that a student has practiced are marked and connected, a student's learning trajectory can be formed, helping to understand the student's learning progress and mastery.

[0115] In this embodiment, the learning map is a visual learning path planning tool that presents learning content, the relationships between knowledge points, and the learning sequence in map form, helping students better understand and plan their learning process. Target learning trajectory: A specific learning path formed on the learning map by the student during the learning process, based on their learning progress, mastery, and learning goals, reflecting the student's learning direction and focus. Target review materials: Recommended review materials tailored to the student's target learning trajectory, aiming to help students consolidate knowledge and improve learning outcomes.

[0116] In this embodiment, each question in the target question bank is mapped to a learning grid. The grid attributes include question ID, knowledge point, difficulty level, and a list of associated grids. These grids can be arranged according to the chapter order of the knowledge point or in ascending order of difficulty for the same knowledge point. Based on the student's answer record, the grid corresponding to the completed question is marked as learned. The grids are connected according to the practice time order or knowledge point dependency relationship to form a learning trajectory.

[0117] To facilitate understanding, an example is provided: the student's practice order is: G1 (basic Pythagorean theorem) → G5 (basic trigonometric functions) → G8 (Pythagorean theorem application problems); the generated trajectory is: G1—G5—G8 (the dotted line connecting G5→G8 indicates skipping learning; the system detected that G5→G8 lacks the intermediate knowledge point G6, prompting for supplementary practice).

[0118] To facilitate a better understanding of learning maps, such as Figure 5 As shown, Figure 5 This is a schematic diagram of the structure of a learning map provided in an embodiment of this application. Figure 5 Each rectangle in the diagram represents a test question. Unfilled rectangles represent test questions that students have not practiced, while black-filled rectangles represent test questions that students have practiced.

[0119] Optionally, a learning map can be used to visually display the learning path and the connections between knowledge points, helping students better plan their learning process and improve learning efficiency.

[0120] Using the above methods, textbook texts and historical question banks (i.e., the target question bank of this application) are obtained. Textbook knowledge points are extracted and combined to form basic knowledge point clusters. Important knowledge point clusters are then selected using a pre-set model, generating target question banks of different difficulty levels. These are then mapped onto a learning map to generate students' learning trajectories and recommend suitable review materials. This system can recommend review materials suitable for students' current learning stage based on their learning progress and knowledge mastery, effectively solving the problem of lacking continuity in review materials and ensuring that review materials match students' learning status.

[0121] Based on the difficulty level of the questions in the target question bank, the questions are divided into different groups, such as easy, medium, and difficult questions. Then, the different groups are arranged according to their difficulty level or the degree of correlation between the questions, and the relationships between the different groups are analyzed. A corresponding learning grid is created for each question, and each learning grid contains detailed information about the question, such as the question content, difficulty level, and knowledge point tags. The learning grid can be in the form of a table or a graph. The rows and columns of a table can represent different knowledge points and difficulty levels, respectively, while a graph can more intuitively show the relationships between questions. Learning grids corresponding to questions within the same group are arranged together, and learning grids corresponding to questions in different groups with correlations are arranged together, thus integrating them into a complete learning map. Each learning grid has a clear location and label on the learning map so that students can clearly view and navigate it.

[0122] The learning management system collects records of practice questions uploaded by students from the target question bank, including question number, practice time, and answer status. Based on these records, the corresponding learning grid for each question is located on the learning map and marked. Marking methods can include color coding, icon labels, or text descriptions; for example, green marks indicate questions answered correctly, and red marks indicate questions answered incorrectly. As students practice new questions, the marking information on the learning map is updated to accurately reflect their latest learning progress. The learning grids corresponding to practiced questions are connected to obtain the target learning trajectory. Based on the order of practiced questions and the layout of the learning grids on the learning map, a learning path is planned from the starting learning grid to the last practiced learning grid. This path can be a straight line or a curved path generated based on the shortest path or the closeness of knowledge point connections. Using graphics drawing tools or programming algorithms, the planned learning path is plotted on the learning map, forming the student's learning trajectory. Learning trajectories can be presented using different colors, line styles, or animation effects to help students understand their learning progress more intuitively. Analyzing these trajectories reveals information such as students' learning habits, knowledge mastery, and learning pace. For example, if the trajectories show a student spending a significant amount of time on a particular knowledge point or repeatedly practicing certain questions, it indicates that the student may be experiencing difficulties with that knowledge point and requires further review and guidance. Furthermore, learning trajectories can serve as a basis for recommending targeted review materials. Based on the student's current learning status and trajectory, suitable review materials can be recommended to help students better consolidate and improve their knowledge.

[0123] When performing the task of "generating the target association strength between target knowledge points learned by students based on their learning trajectories in the learning map", the following methods can be used, but are not limited to: extracting students' learning trajectories from the learning map and selecting multiple first target knowledge point combinations corresponding to the learning trajectories from the knowledge point combination set; determining the knowledge point transfer relationship corresponding to the multiple first target knowledge point combinations and determining the association strength increment corresponding to the multiple first target knowledge point combinations based on the transfer relationship; and determining the target association strength based on the association strength increment and the association strength.

[0124] In this embodiment, the learning trajectory extraction can be a sequence formed by recording the order in which students answer questions on the learning map, which includes questions and corresponding knowledge point combinations. This sequence can be used to reflect the student's learning path and the order in which they come into contact with knowledge points. Color blocks that students have already learned can be filled as the first color block. Connecting the first color blocks in sequence will yield the user's learning trajectory. The first target knowledge point combination can be a combination of knowledge points that corresponds to each question in the learning trajectory and is selected from the set of knowledge point combinations. This combination can be used to analyze the transfer relationship between knowledge points in subsequent analyses.

[0125] For example, Student B's problem-solving order in the junior high school math learning map is: problems combining functions and algebraic operations in the third learning grid → problems combining geometric proofs and functions in the second learning grid → problems combining geometric proofs and probability in the fourth learning grid. The knowledge point combinations corresponding to this learning trajectory are extracted as follows: {a1,c2} (algebraic operations + functions), {b1,c2} (geometric proofs + functions), and {b1,c3} (geometric proofs + probability). Then, from the knowledge point combination set {{a1,b1},{a1,c2},{b1,c2},{b1,c3}}, the above three combinations are selected as the first target knowledge point combinations, i.e., the first target knowledge point combination set is {{a1,c2},{b1,c2},{b1,c3}}.

[0126] In this embodiment, the knowledge point transfer relationship can be the change relationship between two adjacent combinations in the first target knowledge point combination; the association strength increment can be a fixed value that increases the association strength between the common knowledge point and the newly added knowledge point for transfer relationships with common knowledge points, and can be used to reflect the strengthening effect of the learning trajectory on the association of knowledge points. The target association strength can be the sum of the initial association strength and the association strength increment, and can be used to reflect the personalized degree of association between knowledge points under the student's individual learning trajectory.

[0127] For example, a trajectory is formed based on the student's question-answering sequence, T=[C1,C2,…,Ci], where Ci is the combination of knowledge points tested in the i-th question. The knowledge point combinations transferred between adjacent steps in the trajectory are extracted, i.e., Ci→Ci+1. For any ka∈Ci, kb∈Cj, the target association strength expression is updated as shown in Formula 4, where δ is the increment of 0.05, and the target association strength wab(t) = initial strength 1.0 + increment 0.05 = 1.05.

[0128] (Formula 4)

[0129] When performing the "updating the target question bank based on the target knowledge point association network", the following methods can be used, but are not limited to these: the method includes: determining multiple combinations of second target knowledge points from the target knowledge point association network that meet the association conditions; selecting questions corresponding to multiple combinations of second target knowledge points from the target question bank, and determining the selected questions as the updated question bank of the target question bank.

[0130] In this embodiment of the disclosure, the association condition can be a preset target association strength threshold (such as 1.0), which can be used to filter out the combination of knowledge points that are important to students; correspondingly, the second target knowledge point combination can be a combination of knowledge points whose target association strength meets the association condition from the target knowledge point association network, which can be used to determine the scope of questions to update the question bank.

[0131] For example, the second target combination can be determined by association conditions, specifically: setting the association condition as target association strength > 1.0, in Student B's target knowledge point association network, the target association strengths of each knowledge point combination are: a1 and c2 = 1.2, c2 and b1 = 1.05, b1 and c3 = 0.85, a1 and b1 = 1.1. Combinations that meet the conditions are selected: {a1, c2} (1.2 > 1.0), {b1, c2} (1.05 > 1.0), and {a1, b1} (1.1 > 1.0). These three combinations are determined as the second target knowledge point combinations.

[0132] In this embodiment of the disclosure, the updated target question bank can be a question bank adapted to the personalized needs of students by selecting questions corresponding to the second target knowledge point combination from the initial target question bank. This can be used to avoid students being exposed to scattered knowledge point questions and to focus on core related content.

[0133] For example, from the initial target question bank for junior high school mathematics, questions corresponding to the second target knowledge point combinations can be selected. For example, questions corresponding to {a1,c2} include: multiple choice questions on functions and algebraic operations, and problem-solving questions on functions and algebraic operations (15 questions in total); questions corresponding to {b1,c2} include: fill-in-the-blank questions on geometry proofs and functions, and problem-solving questions on geometry proofs and functions (12 questions in total); questions corresponding to {a1,b1} include: multiple choice questions on algebraic operations and geometry proofs, and fill-in-the-blank questions on algebraic operations and geometry proofs (10 questions in total).

[0134] As an alternative, the above 37 questions can be compiled into an updated target question bank. This question bank only contains questions on the core related knowledge points that Student B needs to focus on practicing, helping Student B to focus on knowledge points and avoid doing the same questions repeatedly.

[0135] When performing the action of "updating the knowledge point association network based on the target association strength to obtain the target knowledge point association network corresponding to the student, and updating the target question bank based on the target knowledge point association network, the following methods can be used, but are not limited to": the method includes: evaluating the relevance of the student's learning materials and multiple combinations of second target knowledge points; determining the target learning materials that meet the relevance conditions from the student's learning materials based on the relevance evaluation results, and determining the target learning materials as the student's target learning content; generating recommendation information for the target learning content, which is used to recommend the target learning content to the student.

[0136] In this embodiment of the disclosure, the relevance evaluation can be to calculate the degree of correlation between the set of knowledge points covered by the learning materials and the combination of the second target knowledge points (relevance = the sum of the correlation strength between the knowledge points covered by the materials and the knowledge points in the combination), which can be used to determine the suitability of the learning materials for students; the learning materials can be text or video materials containing knowledge point explanations, example analysis, and exercise summaries, which can be used to help students consolidate knowledge points.

[0137] For example, the correlation evaluation is used to calculate the degree of correlation between the set of knowledge points covered by the learning materials and the combination of the second target knowledge points. Specifically, if the combination of the second target knowledge points for student B is {{a1,c2},{b1,c2},{a1,b1}}, there are 4 sets of learning materials: Material R1: covering {a1} (Lecture notes on basic algebraic operations); Material R2: covering {b1,c2} (Lecture notes on geometric proofs and functions); Material R3: covering {a1,c2} (Special topic analysis on algebra and functions); Material R4: covering {c3} (Summary of basic probability).

[0138] Furthermore, the correlation between the calculated data and the second target knowledge point combination is analyzed:

[0139] The correlation between R1 and {a1,c2} = basic correlation value between a1 and a1 (1.0) + correlation strength between a1 and c2 (1.2) = 2.2;

[0140] The correlation between R2 and {b1,c2} = basic correlation value between b1 and b1 (1.0) + correlation strength between b1 and c2 (1.05) + basic correlation value between c2 and c2 (1.0) = 3.05;

[0141] The correlation between R3 and {a1,c2} = basic correlation value between a1 and a1 (1.0) + correlation strength between a1 and c2 (1.2) + basic correlation value between c2 and c2 (1.0) = 3.2;

[0142] R4 is not related to the second target knowledge point combination, and the correlation is 0.

[0143] In this embodiment of the disclosure, the relevance condition can be a preset relevance threshold, which can be used to filter out learning materials suitable for students; the target learning materials can be learning materials that meet the relevance condition, and the target learning content can be the core content such as knowledge point explanations and example analysis covered by the target learning materials, which can be used to clarify the key review direction for students.

[0144] For example, if the relevance condition is a correlation degree > 2.5, then target learning materials that meet the condition can be selected from the learning resource library based on a correlation degree > 2.5. For example, if the correlation degree between content A in R2 and {b1,c2} is 3.05, then content A can be identified as the target learning content; for example, if the correlation degree between content B in R3 and {a1,c2} is 3.2, then content B can be identified as the target learning content, ensuring that the learning content is highly matched with the combination of the second target knowledge points.

[0145] In this embodiment of the disclosure, the recommendation information may include prompts such as the name of the target learning material, a brief description of the core content, and the access path, which can be used to guide students to quickly obtain and learn the target content and improve learning efficiency.

[0146] For example, the following recommendation information is generated for student B: Dear student, based on your learning trajectory and analysis of your weak knowledge points, the following key learning materials are recommended for you:

[0147] (1) Learning material R2, which contains A content, helps to master the core related knowledge points;

[0148] (2) Learning material R3, which includes content B, can enhance the ability to apply knowledge points. Click on the [Learning Materials] section to view the full content;

[0149] It should be noted that in related technologies, the question bank contains a large number of questions for each knowledge point's current question type, and students often skip chapters when practicing questions. The lack of connection between practice questions results in students only learning questions for each knowledge point's current question type, leading to a relatively scattered learning experience. This application's embodiment, based on the learning trajectory in the learning map, can link knowledge points, solving the problem of single question types. Combining the learning trajectory with the association network enables dynamic updates to the question bank, allowing students to focus on core knowledge points. It can also recommend review materials suitable for students' current learning stage, effectively solving the problem of lack of connection in review materials, making the review materials match the students' learning status.

[0150] Compared to existing technologies, this disclosure extracts clusters of basic and key knowledge points from textbooks, generates a target question bank with varying difficulty and question types, and constructs an initial knowledge point association network. It then combines this with students' trajectories on a learning map to generate target association strength, updating the network and question bank to determine suitable target learning content, helping students focus on core knowledge points and improve learning efficiency. Furthermore, this method can accurately update the association network, calculate personalized association strengths to meet student needs, and filter suitable materials through relevance evaluation of learning materials and knowledge point combinations, further assisting students in efficiently consolidating core knowledge points and improving review effectiveness.

[0151] Furthermore, as Figure 1 and Figure 3 The specific implementation of the method shown in this embodiment provides a learning content determination device based on a smart education platform, such as... Figure 6 As shown, the device includes: an extraction module 31, a generation module 32, and an update module 33.

[0152] The extraction module 31 is configured to extract basic knowledge point clusters and key knowledge point clusters from student textbooks, and generate a target question bank based on the basic knowledge point clusters and key knowledge point clusters with different difficulty levels and question types.

[0153] The generation module 32 is configured to generate a knowledge point association network corresponding to the target question bank based on the association relationship and association strength between the knowledge points corresponding to each question in the target question bank.

[0154] The generation module 32 is also configured to generate the target association strength between the target knowledge points learned by the student based on the student's learning trajectory in the learning map. The learning map is obtained by mapping the basic knowledge point clusters and key knowledge point clusters in multiple learning grids arranged according to a predetermined relationship.

[0155] The update module 33 is configured to update the knowledge point association network based on the target association strength to obtain the target knowledge point association network corresponding to the student, and update the target question bank based on the target knowledge point association network. The target question bank is used to determine the user's target learning content.

[0156] In some embodiments of this disclosure, the generation module 32 is specifically configured to: determine a set of knowledge point combinations corresponding to the target question bank based on the knowledge point combinations corresponding to each question in the target question bank; determine the relationship between knowledge points based on the set of knowledge point combinations; determine the number of questions corresponding to each knowledge point in the knowledge point combination and the number of combinations with other knowledge points; perform a weighted analysis on the number of questions and the number of combinations based on the target decay parameter at the current time to obtain the relationship strength between knowledge points corresponding to the set of knowledge point combinations at the current time; and generate a knowledge point association network corresponding to the target question bank based on the relationship and relationship strength between knowledge points corresponding to the set of knowledge point combinations.

[0157] In some embodiments of this disclosure, the generation module 32 is further configured to: mark the questions corresponding to the basic knowledge point clusters in the target question bank with a first identifier, and mark the questions corresponding to the key knowledge point clusters with a second identifier; map the questions marked with the first identifier and the questions marked with the second identifier to multiple learning grids in a preset map according to the order of the chapters in the textbook to obtain a learning map, wherein each learning grid in the learning map corresponds to all the questions of a knowledge point cluster, and the learning grid corresponding to the questions marked with the second identifier is a key learning grid.

[0158] In some embodiments of this disclosure, the extraction module 31 is specifically configured to: extract the student's learning trajectory from the learning map; select multiple first target knowledge point combinations corresponding to the learning trajectory from the knowledge point combination set; determine the knowledge point transfer relationship corresponding to the multiple first target knowledge point combinations; determine the association strength increment corresponding to the multiple first target knowledge point combinations based on the transfer relationship; and determine the target association strength based on the association strength increment and the association strength.

[0159] In some embodiments of this disclosure, the update module 33 is specifically configured to: determine multiple combinations of second target knowledge points whose target association strength meets the association conditions from the target knowledge point association network; select questions corresponding to multiple combinations of second target knowledge points from the target question bank; and determine the selected questions as the updated question bank of the target question bank.

[0160] In some embodiments of this disclosure, the generation module 32 is further configured to: perform a relevance evaluation based on the student's learning materials and multiple combinations of second target knowledge points; determine target learning materials that meet the relevance conditions from the student's learning materials according to the relevance evaluation results, and determine the target learning materials as the student's target learning content; generate recommendation information for the target learning content, which is used to recommend the target learning content to the student.

[0161] It should be noted that other corresponding descriptions of the functional units involved in the learning content determination device based on the smart education platform provided in this embodiment can be found in [reference]. Figure 1 and Figure 3 The corresponding description in [the document] will not be repeated here.

[0162] Based on the above, Figure 1 and Figure 3 Accordingly, this embodiment also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described method. Figure 1 and Figure 3 The method shown.

[0163] Based on this understanding, the technical solution disclosed herein can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, or portable hard drive) and includes several instructions to cause a computer device (such as a personal computer, server, or network device) to execute the methods of various implementation scenarios of this disclosure.

[0164] like Figure 7 The diagram shown is a hardware structure schematic of an electronic device according to the present invention, comprising:

[0165] At least one processor 401; and,

[0166] Memory 402 is communicatively connected to at least one processor 401; wherein,

[0167] The memory 402 stores instructions that can be executed by at least one processor, which enables the at least one processor to perform the learning aid method as described above.

[0168] Figure 7 Take a processor 401 as an example.

[0169] The electronic device may also include an input device 403 and a display device 404.

[0170] The processor 401, memory 402, input device 403, and display device 404 can be connected via a bus or other means. Figure 7 Taking the example of a connection between China and Israel via a bus.

[0171] Memory 402, as a non-volatile computer-readable storage medium, can be used to store non-volatile software programs, non-volatile computer-executable programs, and modules, such as the program instructions / modules corresponding to the learning assistance method in the embodiments of this disclosure, for example, Figure 1 and Figure 3 The method flow is shown. The processor 401 executes various functional applications and data processing by running non-volatile software programs, instructions, and modules stored in the memory 402, thereby implementing the learning assistance method in the above embodiments.

[0172] Memory 402 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created based on the use of the learning-aiding method, etc. Furthermore, memory 402 may include high-speed random access memory and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other non-volatile solid-state storage device. In some embodiments, memory 402 may optionally include memory remotely located relative to processor 401, and these remote memories may be connected via a network to the apparatus performing the learning-aiding method. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0173] The input device 403 can receive user clicks and generate signal inputs related to user settings and function control of the learning assistance method. The display device 404 may include a display screen or other display device.

[0174] One or more modules are stored in memory 402, and when run by one or more processors 401, the learning assistance method in any of the above method embodiments is executed.

[0175] Optionally, the aforementioned physical devices may also include a user interface, a network interface, a camera, radio frequency (RF) circuitry, sensors, audio circuitry, a Wi-Fi module, etc. The user interface may include a display screen, input units such as a keyboard, etc., and optional user interfaces may also include USB interfaces, card reader interfaces, etc. The network interface may optionally include standard wired interfaces, wireless interfaces (such as Wi-Fi interfaces), etc.

[0176] Those skilled in the art will understand that the physical device structure provided in this embodiment does not constitute a limitation on the physical device, and may include more or fewer components, or combine certain components, or have different component arrangements.

[0177] The storage medium may also include an operating system and a network communication module. The operating system is a program that manages the hardware and software resources of the aforementioned physical device, supporting the operation of information processing programs and other software and / or programs. The network communication module is used to enable communication between the various components within the storage medium, as well as communication with other hardware and software in the information processing physical device.

[0178] Through the above description of the embodiments, those skilled in the art can clearly understand that this disclosure can be implemented using software plus necessary general-purpose hardware platforms, or it can be implemented using hardware. By applying the solution of this embodiment, compared with the prior art, this disclosure extracts basic and key knowledge point clusters from textbooks, generates a target question bank with multiple difficulties and question types, and constructs an initial knowledge point association network; it combines the student's trajectory in the learning map to generate target association strength, updates the network and question bank to determine suitable target learning content, and helps students focus on core knowledge points and improve learning efficiency. At the same time, this method can accurately update the association network, calculate personalized association strength to meet student needs, and can also screen suitable materials through the relevance evaluation of learning materials and knowledge point combinations, further helping students efficiently consolidate core knowledge points and improve review effectiveness.

[0179] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0180] The above description is merely a specific embodiment of this disclosure, enabling those skilled in the art to understand or implement it. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this disclosure. Therefore, this disclosure is not to be limited to the embodiments described herein, but is to be accorded the widest scope consistent with the principles and novel features claimed herein.

Claims

1. A method for determining learning content based on a smart education platform, characterized in that, The method comprises the following steps: extracting a basic knowledge point cluster and a key knowledge point cluster from a student textbook, and generating a question bank of different difficulty levels and different question types according to the basic knowledge point cluster and the key knowledge point cluster; generating a knowledge point association network corresponding to the question bank based on the association relationship and the association strength between the knowledge points corresponding to each question in the question bank; generating a target association strength between target knowledge points learned by a student based on a learning trajectory of the student in a learning map, the learning map being obtained by mapping the basic knowledge point cluster and the key knowledge point cluster in a plurality of learning grids arranged according to a predetermined relationship; updating the knowledge point association network according to the target association strength to obtain a target knowledge point association network corresponding to the student, and updating the question bank based on the target knowledge point association network, the question bank being used to determine the target learning content of the user; extracting the learning trajectory of the student from the learning map, and selecting a plurality of first target knowledge point combinations corresponding to the learning trajectory from a knowledge point combination set; determining a knowledge point transition relationship corresponding to the plurality of first target knowledge point combinations, and determining an association strength increment corresponding to the plurality of first target knowledge point combinations based on the transition relationship; determining the target association strength according to the association strength increment and the association strength.

2. The method of claim 1, wherein, The method comprises the following steps: determining a knowledge point combination set corresponding to the question bank based on the knowledge point combination corresponding to each question in the question bank, and determining the association relationship between the knowledge points based on the knowledge point combination set; determining the number of questions corresponding to each knowledge point in the knowledge point combination, and the combination times of each knowledge point in the knowledge point combination combined with other knowledge points to form a knowledge point combination; weighting and analyzing the number of questions and the combination times based on a target decay parameter at the current time to obtain the association strength between the knowledge points corresponding to the knowledge point combination set at the current time; generating the knowledge point association network corresponding to the question bank according to the association relationship and the association strength between the knowledge points corresponding to the knowledge point combination set.

3. The method of claim 2, wherein, Before the step of generating a target association strength between target knowledge points learned by a student based on a learning trajectory of the student in a learning map, the method further comprises the following steps: marking the questions corresponding to the basic knowledge point cluster in the question bank with a first mark, and marking the questions corresponding to the key knowledge point cluster with a second mark; mapping the questions marked with the first mark and the questions marked with the second mark in a plurality of learning grids in a preset map according to the order of chapters in the textbook to obtain the learning map, each learning grid in the learning map corresponding to all the questions of a knowledge point cluster, and the learning grid corresponding to the questions marked with the second mark being a key learning grid.

4. The method of claim 1, wherein, updating the question bank based on the target knowledge point association network comprises the following steps: determine a plurality of second target knowledge point combinations from the target knowledge point association network, wherein the target association strength of the plurality of second target knowledge point combinations meets the association condition; select a question corresponding to the plurality of second target knowledge point combinations from the target question bank, and update the target question bank by determining the selected question as the updated target question bank.

5. The method of claim 4, wherein, After the knowledge point association network is updated according to the target association strength to obtain the target knowledge point association network corresponding to the student, and the target question bank is updated based on the target knowledge point association network, the method further comprises: performing relevance evaluation based on the learning materials of the student and the plurality of second target knowledge point combinations; determining target learning materials meeting the relevance condition from the learning materials of the student according to the relevance evaluation result, and determining the target learning materials as the target learning content of the student; generating recommendation information of the target learning content, wherein the recommendation information is used to recommend the target learning content to the student. 6.A learning content determination device based on a smart education platform, characterized by comprising: comprises: an extraction module configured to extract a basic knowledge point cluster and a key knowledge point cluster from a student textbook, and generate a target question bank composed of questions of different difficulty levels and different question types according to the basic knowledge point cluster and the key knowledge point cluster; a generation module configured to generate a knowledge point association network corresponding to the target question bank based on the association relationship and the association strength between the knowledge points corresponding to each question in the target question bank; the generation module is further configured to generate a target association strength between target knowledge points learned by the student based on a learning trajectory of the student in a learning map, wherein the learning map is obtained by mapping the basic knowledge point cluster and the key knowledge point cluster in a plurality of learning grids arranged according to a predetermined relationship; extract the learning trajectory of the student from the learning map, and select a plurality of first target knowledge point combinations corresponding to the learning trajectory from a knowledge point combination set; determine a knowledge point transfer relationship corresponding to the plurality of first target knowledge point combinations, and determine an association strength increment corresponding to the plurality of first target knowledge point combinations based on the transfer relationship; determine the target association strength according to the association strength increment and the association strength; an update module configured to update the knowledge point association network according to the target association strength to obtain a target knowledge point association network corresponding to the student, and update the target question bank based on the target knowledge point association network, wherein the target question bank is used to determine the target learning content of the user.

7. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the method of any one of claims 1 to 5.

8. An electronic device comprising a storage medium, a processor, and a computer program stored on the storage medium and executable on the processor, characterized in that, The processor executes the computer program to implement the method of any one of claims 1 to 5.

9. A computer program product comprising a computer program, characterized in that, The computer program is executed by the processor to implement the method of any one of claims 1 to 5.

Images