Question bank generation method and device, storage medium and electronic equipment

Abstract

The invention provides a question bank generation method and device, a storage medium and electronic equipment. The method comprises the following steps: collecting a first feature data set containing influence factors of wrong question making of students by using an intelligent education terminal; a hierarchical clustering algorithm is utilized to screen out the characteristics of wrong questions made by the students due to objective factors from the first characteristic data set; utilizing a preset wrong question correlation evaluation model to evaluate correlation scores among the characteristics of the wrong questions made by the students caused by the objective factors, sorting the characteristics of the wrong questions made by the students caused by the objective factors from large to small according to the correlation scores, and determining knowledge points corresponding to the characteristic with the maximum correlation score as a first target knowledge point cluster; based on the basic knowledge point cluster and the first target knowledge point cluster, question banks corresponding to the questions of different difficulty levels are generated and mapped on a learning map, the learning map comprises one or more key learning grids, and the key learning grids are grids corresponding to the first target knowledge point cluster.

Description

Technical Field

[0001] The present disclosure relates to the field of smart education, and in particular to a question bank generation method, device, storage medium and electronic device. Background Art

[0002] In recent years, with the continuous advancement of computer technology and educational informatization, computer and artificial intelligence technology have been gradually applied to various daily educational and teaching activities. Improving the accuracy of knowledge points in knowledge point clusters can effectively help teachers improve the hit rate of important knowledge points, and at the same time recommend more accurate knowledge points to students to improve their learning effects.

[0003] In the current relevant technology, knowledge point clusters include basic knowledge point clusters and important knowledge point clusters, but the important knowledge point clusters do not eliminate the knowledge points affected by subjective reasons such as students' guessing, not doing, not knowing how to do, half-knowledge, misreading questions, lack of proficiency, etc. during the test-taking process. Therefore, it is impossible to effectively obtain the students' mastery of knowledge points, which affects the accuracy of key knowledge point screening, resulting in low accuracy of knowledge points in the question bank generated based on the basic knowledge point clusters and the important knowledge point clusters, and further resulting in the inability to dynamically adjust the difficulty of questions recommended to students according to their actual learning level. Summary of the invention

[0004] The present invention provides a method, device, storage medium and electronic device for generating a question bank.

[0005] According to a first aspect of the present disclosure, a method for generating a question bank is provided, the method comprising: The first feature data set containing the factors affecting students’ wrong answers is collected by using the smart education terminal. The factors affecting students’ wrong answers include subjective factors and objective factors. Hierarchical clustering algorithm is used to screen out the characteristics of objective factors that cause students to make mistakes in the questions from the first characteristic data set of factors affecting making mistakes in the questions; The preset wrong question correlation evaluation model is used to evaluate the correlation scores between the features of objective factors that cause students to make wrong questions, and the features of objective factors that cause students to make wrong questions are sorted from large to small according to the correlation scores, and the knowledge point corresponding to the feature with the largest correlation score is determined as the first target knowledge point cluster; Based on the basic knowledge point cluster and the first target knowledge point cluster, a question bank corresponding to questions of different difficulty levels is generated and mapped on the learning map. The learning map includes one or more key learning grids, which are grids corresponding to the first target knowledge point cluster. The basic knowledge point cluster is obtained by extracting knowledge points from the textbook based on the smart education terminal.

[0006] In some embodiments of the present disclosure, a first feature data set including factors affecting students' wrong answers is collected using a smart education terminal, including: Collect students' test-taking videos and test-taking status based on smart education terminals; The facial expression recognition algorithm and eye tracking algorithm are used to identify the subjective factor feature data set that causes students to make mistakes in the test-taking video footage, and the objective factor feature data set that causes students to make mistakes is obtained by using the students' test-taking situation. Based on the subjective factor feature data set and the objective factor feature data set, a first feature data set including factors influencing students' wrong answers is determined.

[0007] In some embodiments of the present disclosure, the subjective factor feature data set and the objective factor feature data set determine a first feature data set containing factors affecting students' wrong answers, including: The subjective factor feature data set and the objective factor feature data set are combined to obtain a second feature data set containing the factors affecting students' wrong answers; All feature data in the second feature data set are preprocessed and normalized to obtain a first feature data set containing factors affecting students' wrong answers.

[0008] In some embodiments of the present disclosure, a preset wrong question correlation evaluation model is used to evaluate the correlation scores between the features of objective factors that cause students to make wrong answers, including: Using a preset correlation algorithm to perform correlation evaluation on each wrong question influencing factor in the first feature data set, obtaining a first correlation analysis result between features in the first feature data set, and obtaining the correlation strength between the features of the wrong question influencing factors according to the first correlation analysis result; Based on the first correlation analysis result, generating a dendrogram from the first feature data set of factors affecting wrong answers according to a hierarchical clustering algorithm, determining the number of groups of features in the first feature data set according to the number of clusters in the dendrogram, and allocating the feature data in the first feature data set to corresponding groups to obtain at least two third feature data sets; The third feature data set is subjected to correlation evaluation using a pre-trained correlation evaluation model to obtain a second correlation analysis result between the third feature data sets, and a correlation score is obtained based on the first correlation analysis result and the second correlation analysis result.

[0009] In some embodiments of the present disclosure, a preset correlation algorithm is used to perform a correlation evaluation on each wrong question influencing factor in the first feature data set, to obtain a first correlation analysis result between features in the first feature data set, and to obtain the correlation strength between the wrong question influencing factor features according to the first correlation analysis result, including: The first correlation analysis result is calculated by a preset correlation algorithm:

[0010] in, Indicates the features in the first feature data set and Features The average value of represents the features in the first feature data set, Representation characteristics and Features The first correlation analysis results between; According to the first correlation analysis results, the correlation strength between the characteristics of the factors affecting wrong questions is obtained, among which, The value range is [0, 1].

[0011] In some embodiments of the present disclosure, a pre-trained correlation evaluation model is used to perform correlation evaluation on the third feature data set to obtain a second correlation analysis result between the third feature data sets, and a correlation score is obtained based on the first correlation analysis result and the second correlation analysis result, including: The second correlation analysis result is calculated by the pre-trained correlation evaluation model:

[0012] in, represents the second correlation analysis result, represents a constant term, represents the partial regression coefficient of the preset correlation evaluation model, Represents the characteristics of the third feature data set; A correlation score is obtained based on the first correlation analysis result and the second correlation analysis result.

[0013] In some embodiments of the present disclosure, after evaluating the correlation scores between the features of objective factors that cause students to make mistakes in questions using a preset wrong question correlation evaluation model, and sorting the features of objective factors that cause students to make mistakes in questions from large to small according to the correlation scores, and determining the knowledge point corresponding to the feature with the largest correlation score as the first target knowledge point cluster, the method further includes: Determine a basic knowledge point cluster based on the extraction of knowledge points in the textbook text, determine a first important knowledge point cluster based on the importance of knowledge points in the basic knowledge point cluster, and determine a second important knowledge point cluster based on the extraction of knowledge points in the history question bank and the difficulty level of the knowledge points in the history question bank; Determine the same knowledge points in the first target knowledge point cluster, the first important knowledge point cluster, and the second important knowledge point cluster as the second target knowledge point cluster; Based on the basic knowledge point cluster and the first target knowledge point cluster, a question bank corresponding to questions of different difficulty levels is generated and mapped on the learning map, including: Based on the basic knowledge point clusters and the second target knowledge point clusters, a question bank corresponding to questions of different difficulty levels is generated and mapped on the learning map.

[0014] According to a second aspect of the present disclosure, a question bank generating device is provided, the device comprising: A collection unit, used for collecting a first feature data set including factors affecting students' wrong answers by using a smart education terminal, where the factors affecting students' wrong answers include subjective factors and objective factors; A screening unit, used for screening out the features of objective factors that cause students to make mistakes in the questions from the first feature data set of factors affecting making mistakes in the questions by using a hierarchical clustering algorithm; A determination unit is used to evaluate the correlation scores between the features of objective factors that cause students to make mistakes in the questions by using a preset wrong question correlation evaluation model, and sort the features of objective factors that cause students to make mistakes in the questions from large to small according to the correlation scores, and determine the knowledge point corresponding to the feature with the largest correlation score as the first target knowledge point cluster; A generation unit is used to generate a question bank corresponding to questions of different difficulty levels based on the basic knowledge point cluster and the first target knowledge point cluster and map them on a learning map. The learning map includes one or more key learning grids, which are grids corresponding to the first target knowledge point cluster. The basic knowledge point cluster is obtained by extracting knowledge points from the textbook based on the smart education terminal.

[0015] According to a third aspect of the present disclosure, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the method described in the first aspect is implemented.

[0016] According to a fourth aspect of the present disclosure, an electronic device is provided, comprising a storage medium, a processor, and a computer program stored on the storage medium and executable on the processor, wherein the processor implements the method described in the first aspect when executing the computer program.

[0017] The question bank generation method, device, storage medium and electronic device provided by the present invention collect a first feature data set containing factors affecting students' wrong answers by using a smart education terminal, wherein the factors affecting students' wrong answers include subjective factors and objective factors; a hierarchical clustering algorithm is used to screen out the features of the objective factors that cause students to make wrong answers from the first feature data set of factors affecting students' wrong answers; a preset wrong answer correlation evaluation model is used to evaluate the correlation scores between the features of the objective factors that cause students to make wrong answers, and the features of the objective factors that cause students to make wrong answers are sorted from large to small according to the correlation scores, and the knowledge points corresponding to the features with the largest correlation scores are determined as the first target knowledge point clusters; based on the basic knowledge point clusters and the first target knowledge point clusters, the first target knowledge point clusters are clustered. The knowledge point clusters are marked, and the question banks corresponding to the questions of different difficulty levels are generated and mapped on the learning map. The learning map includes one or more key learning grids, and the key learning grids are the grids corresponding to the first target knowledge point cluster. The basic knowledge point cluster is obtained by extracting the knowledge points in the textbook based on the smart education terminal. By taking the knowledge points with the greatest impact on students' wrong answers among the objective factors as the knowledge points in the first target knowledge point cluster, the accuracy of the first target knowledge point cluster can be improved, thereby improving the accuracy of the question banks corresponding to the questions of different difficulty levels generated based on the basic knowledge point cluster and the first target knowledge point cluster. The difficulty of the questions recommended to students can be dynamically adjusted according to their actual learning level.

[0018] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The accompanying drawings are used to better understand the present solution and do not constitute a limitation of the present disclosure. Figure 1 A flowchart of a method for generating a question bank provided by an embodiment of the present disclosure; Figure 2 A flowchart of a method for obtaining a first feature data set provided by an embodiment of the present disclosure; Figure 3 A schematic diagram of a knowledge point network provided by an embodiment of the present disclosure; Figure 4 A schematic diagram of the screening process provided by the embodiment of the present disclosure; Figure 5 A schematic diagram of question bank marking provided by an embodiment of the present disclosure; Figure 6 A schematic diagram of the structure of a question bank generating device provided by an embodiment of the present disclosure; Figure 7 A schematic diagram of the hardware structure of an electronic device provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0020] The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0021] In recent years, with the continuous advancement of computer technology and educational informatization, computer and artificial intelligence technology have been gradually applied to various daily educational and teaching activities. Improving the accuracy of knowledge points in knowledge point clusters can effectively help teachers improve the hit rate of important knowledge points, and at the same time recommend more accurate knowledge points to students to improve their learning effects.

[0022] In the current relevant technology, knowledge point clusters include basic knowledge point clusters and important knowledge point clusters, but the important knowledge point clusters do not eliminate the knowledge points affected by subjective reasons such as students' guessing, not doing, not knowing how to do, half-knowledge, misreading questions, lack of proficiency, etc. during the test-taking process. Therefore, it is impossible to effectively obtain the students' mastery of knowledge points, which affects the accuracy of key knowledge point screening, resulting in low accuracy of knowledge points in the question bank generated based on the basic knowledge point clusters and the important knowledge point clusters, and further resulting in the inability to dynamically adjust the difficulty of questions recommended to students according to their actual learning level.

[0023] In order to solve the problems in the related art, the question bank generation method provided by the present disclosure collects a first feature data set containing factors affecting students' wrong answers by using a smart education terminal, wherein the factors affecting students' wrong answers include subjective factors and objective factors; a hierarchical clustering algorithm is used to screen out the features of the objective factors that cause students to make wrong answers from the first feature data set of factors affecting students' wrong answers; a preset wrong answer correlation evaluation model is used to evaluate the correlation scores between the features of the objective factors that cause students to make wrong answers, and the features of the objective factors that cause students to make wrong answers are sorted from large to small according to the correlation scores, and the knowledge points corresponding to the features with the largest correlation scores are determined as the first target knowledge point clusters; based on the basic knowledge point clusters and the first target knowledge point clusters, the clusters are sorted into a set of clusters. The knowledge point clusters are marked, and the question banks corresponding to the questions of different difficulty levels are generated and mapped on the learning map. The learning map includes one or more key learning grids, and the key learning grids are the grids corresponding to the first target knowledge point cluster. The basic knowledge point cluster is obtained by extracting the knowledge points in the textbook based on the smart education terminal. By taking the knowledge points with the greatest impact on students' wrong answers among the objective factors as the knowledge points in the first target knowledge point cluster, the accuracy of the first target knowledge point cluster can be improved, thereby improving the accuracy of the question banks corresponding to the questions of different difficulty levels generated based on the basic knowledge point cluster and the first target knowledge point cluster. The difficulty of the questions recommended to students can be dynamically adjusted according to their actual learning level.

[0024] The following describes the question bank generation method, device, electronic device and storage medium of the embodiments of the present disclosure with reference to the accompanying drawings.

[0025] Figure 1 The following is a flow chart of a method for generating a question bank provided by an embodiment of the present disclosure. Figure 1 As shown, the method includes: Step 101: using a smart education terminal to collect a first feature data set including factors affecting students’ wrong answers, where the factors affecting students’ wrong answers include subjective factors and objective factors; In some embodiments, the smart education terminal can be implemented based on hardware devices (such as tablets, smart whiteboards), or it can be based on software applications and installed on ordinary computers or mobile devices. Usually these terminals can be connected to cloud services to access rich educational resources and perform data synchronization and other functions.

[0026] In some embodiments, a first feature data set including factors affecting students' incorrect answers can be obtained from log data of a learning management system.

[0027] In some embodiments, since there are many factors that affect students' incorrect answers, including reading the questions incorrectly, writing the questions incorrectly, and really not knowing how to do them, it is impossible to effectively obtain students' mastery of knowledge points. Therefore, it is necessary to filter out knowledge points that students really cannot do from a feature data set that includes factors that affect students' incorrect answers, so that teachers can understand students' mastery of knowledge points.

[0028] In some embodiments, subjective factors may include emotional characteristics including positive, bland, negative, etc.; concentration characteristics including focus, distraction, susceptibility to external factors, etc.; question-solving style characteristics including carefulness, carelessness, lack of patience, etc.; objective factors may include knowledge point mastery characteristics including proficiency, proficiency, unproficiency, etc.; question-solving efficiency characteristics including fast, medium, slow, etc.; question difficulty level characteristics: simple, ordinary, difficult, etc.

[0029] Step 102, using a hierarchical clustering algorithm to filter out the features of objective factors that cause students to make mistakes in the questions from the first feature data set of factors affecting making mistakes in the questions; In some embodiments, a bottom-up or top-down hierarchical clustering method may be selected. Taking bottom-up hierarchical clustering as an example, each factor affecting wrong answers may be first regarded as a separate category, and then the most similar categories may be merged in turn.

[0030] In some embodiments, before performing hierarchical clustering, a distance metric may be selected to measure the similarity or difference between different features. Commonly used distance metrics include Euclidean distance, Manhattan distance, and the like.

[0031] In some embodiments, improving the features for screening out objective factors that cause students to make mistakes in answering questions can eliminate the influence of subjective factors on making mistakes in answering questions, reduce data interference, and accurately determine whether students have mastered the corresponding knowledge points.

[0032] Step 103: Use a preset wrong question correlation evaluation model to evaluate the correlation scores between the features of objective factors that cause students to make wrong questions, and sort the features of objective factors that cause students to make wrong questions from large to small according to the correlation scores, and determine the knowledge point corresponding to the feature with the largest correlation score as the first target knowledge point cluster; In some embodiments, the preset wrong question relevance assessment model may be a pre-trained relevance assessment model.

[0033] In some embodiments, the correlation score may be directly obtained by a similarity algorithm, or may be obtained by an average value corresponding to the correlations obtained by multiple similarity algorithms, or may be obtained by a weighted sum of the correlations obtained by multiple similarity algorithms.

[0034] In some embodiments, by sorting the features that are objective factors causing students to make mistakes in answering questions from large to small according to their relevance scores, the speed of screening the features with the largest subsequent relevance scores can be increased.

[0035] In some embodiments, the feature with the largest correlation score is the feature that has the greatest impact on students' wrong answers among the objective factors. By determining the knowledge points corresponding to the objective factors that have the greatest impact on students' wrong answers as the first target knowledge point cluster, the knowledge points that cause students to answer wrong answers due to subjective factors can be eliminated, and the knowledge points corresponding to the objective factors that have the greatest impact on students' wrong answers can be obtained. The accuracy of the knowledge points in the obtained first target knowledge point cluster is higher.

[0036] Step 104: Based on the basic knowledge point cluster and the first target knowledge point cluster, generate a question bank corresponding to questions of different difficulty levels and map them on a learning map. The learning map includes one or more key learning grids, which are grids corresponding to the first target knowledge point cluster. The basic knowledge point cluster is obtained by extracting knowledge points from the textbook based on the smart education terminal.

[0037] In some embodiments, a learning map is a visualization tool that can be presented in the form of a network diagram or a flow chart. The learning map is used to help students understand and navigate complex knowledge systems, where each knowledge point cluster can be represented as a grid in the learning map, such as a basic knowledge point cluster and a first target knowledge point cluster each corresponding to a learning grid.

[0038] In some embodiments, the learning map is used to reflect the learning grid corresponding to the current user. The learning grids can be linked in sequence according to the order in which the students do the questions to obtain the learning path of the current user. The learning path is generated based on the learning map according to the student's learning goals, the student's current knowledge level, and the student's learning progress. Through the combination of knowledge maps and learning paths, key knowledge points are identified and monitored when students enter the learning scope of these knowledge points. When students encounter difficulties in their learning progress, the system will match users with similar learning paths to promote interaction and mutual assistance between students, thereby improving the learning experience.

[0039] In some embodiments, by performing detailed classification of the knowledge points in the basic knowledge point cluster and the first target knowledge point cluster, and determining the difficulty level of the knowledge points according to the degree of refinement, the systematicness and hierarchy of the knowledge points are ensured, thereby providing a basis for the subsequent generation of questions in the question bank. The questions in the question bank correspond to the important knowledge points in the textbook content. This application does not limit the types of questions in the question bank, which may be one or more of multiple-choice questions, fill-in-the-blank questions, short-answer questions, essay questions, etc.

[0040] In some embodiments, important knowledge point clusters that are core concepts or principles with wide application are marked in the knowledge graph, and different colors or symbols can be used to distinguish between common knowledge points and first target knowledge points. A learning map is designed based on the knowledge graph to ensure that the relationship between knowledge points is clearly visible and the first target knowledge point cluster is highlighted.

[0041] In some embodiments, the knowledge points can be distributed on a two-dimensional grid, with each knowledge point cluster occupying a cell. The size of the cell can be adjusted according to the complexity and importance of the knowledge point. The cell where the first target knowledge point cluster is located can be marked with a special identifier to form a focused learning grid.

[0042] In some embodiments, a navigation path may be added to the learning map to indicate the learning sequence from the basic knowledge point to the first target knowledge point, which may help students better plan their learning path.

[0043] In some embodiments, by determining key learning grids on the learning map, key knowledge points can be highlighted while providing students with a clear learning path, thereby improving student learning outcomes.

[0044] As a possible implementation, Figure 2 The flowchart of a method for generating a question bank shown in the figure, based on the above embodiment, uses a smart education terminal to collect a first feature data set containing factors affecting students' wrong answers, including the following steps: Step 201: Collect the video images of students doing the questions and the status of the students doing the questions based on the smart education terminal; In some embodiments, through the image acquisition device installed in the classroom, the smart education terminal can record the entire process of students doing the questions in real time, including students' writing movements, thinking time, facial expressions, etc.

[0045] In some embodiments, the smart education terminal can automatically record the specific questions and answers completed by each student, the time taken to complete each question, whether the answer was modified and the number of times the answer was modified, whether the prompts were checked, and other information to obtain the student's performance.

[0046] Step 202: using facial expression recognition algorithm and eye tracking algorithm to identify the subjective factor feature data set that causes students to make mistakes in the test-taking video, and using the students' test-taking situation to obtain the objective factor feature data set that causes students to make mistakes; In some embodiments, a facial expression recognition algorithm (Single Shot Multibox Detector, SSD) can be a deep learning-based method for detecting and classifying the emotional states of students during the problem-solving process, such as positive, indifferent, negative, confused, frustrated, confident, etc. Specifically, it can mark the emotional changes that occur within a specific time period, especially the facial expressions of students when facing difficult problems.

[0047] In some embodiments, eye tracking algorithms are used to monitor students' gaze movement patterns, including the location of students' gaze points, the path of scanning questions, etc., and by analyzing these movement patterns and question content, such as whether students spend too much time on irrelevant information or skip key parts, to analyze whether students are focused and careful when doing the questions.

[0048] In some embodiments, eye tracking can be performed using a dedicated eye tracker, a high-resolution camera, and an image processing algorithm.

[0049] Step 203: Based on the subjective factor feature data set and the objective factor feature data set, determine a first feature data set containing factors affecting students' wrong answers.

[0050] In some embodiments, subjective factors (such as emotional characteristics, concentration characteristics, and test-taking style characteristics) and objective factors (such as knowledge point mastery characteristics and test-taking efficiency characteristics) are combined to form a first feature set. By determining a first feature data set that includes factors affecting students' incorrect answers, the impact of both subjective and objective factors on students' incorrect answers can be considered at the same time.

[0051] As a possible implementation, based on the above embodiment, the specific process of determining the first feature data set containing the factors affecting students' wrong answers based on the subjective factor feature data set and the objective factor feature data set includes: The subjective factor feature data set and the objective factor feature data set are combined to obtain a second feature data set containing the factors affecting students' wrong answers; All feature data in the second feature data set are preprocessed and normalized to obtain a first feature data set containing factors affecting students' wrong answers.

[0052] In some embodiments, since the data in the subjective factor feature data set and the objective factor feature data set are collected, there may be outliers and missing values. Therefore, preprocessing is required before performing wrong question analysis to ensure the quality and consistency of the data. Specifically, an isolation forest can be used to detect outliers and decide whether to delete or correct these outliers. It is possible to delete records with missing values ​​or use interpolation methods to fill in missing values.

[0053] In some embodiments, the second feature data set includes all feature data in the subjective factor feature data set and the objective factor feature data set, including abnormal values ​​and missing values.

[0054] In some embodiments, normalization processing is used to map all feature data in the subjective factor feature data set and the objective factor feature data set to the same range to make data of different scales comparable. This application does not limit the method of normalization processing.

[0055] As a possible implementation, based on the above embodiment, a preset wrong question correlation evaluation model is used to evaluate the correlation scores between the features of the objective factors that cause students to make wrong answers, including: Using a preset correlation algorithm to perform correlation evaluation on each wrong question influencing factor in the first feature data set, obtaining a first correlation analysis result between features in the first feature data set, and obtaining the correlation strength between the features of the wrong question influencing factors according to the first correlation analysis result; In some embodiments, the preset correlation algorithm can be a Pearson correlation coefficient distance algorithm or a Spearman rank correlation coefficient. As long as it is an algorithm that can obtain the correlation between two features, it can be applied here. The preset correlation algorithm in this application takes the Pearson correlation coefficient distance algorithm as an example. The Pearson correlation coefficient distance algorithm can be used to obtain the correlation between two features in the first feature data set. The size of the correlation is used to measure the correlation strength between the two features.

[0056] In some embodiments, since it is a normal phenomenon that the difficulty of a question leads to slow test-taking efficiency, it is necessary to first find the correlation between the features in the first feature data set.

[0057] Based on the first correlation analysis result, generating a dendrogram from the first feature data set of factors affecting wrong answers according to a hierarchical clustering algorithm, determining the number of groups of features in the first feature data set according to the number of clusters in the dendrogram, and allocating the feature data in the first feature data set to corresponding groups to obtain at least two third feature data sets; In some embodiments, the first correlation analysis result is expressed in the form of a correlation matrix, and the correlation matrix can be used as the input of a hierarchical clustering algorithm to obtain a corresponding dendrogram, wherein a cluster in the dendrogram represents a group of strongly correlated features. Based on the dendrogram, the nodes in the dendrogram can be automatically segmented using the elbow method or the silhouette coefficient method to obtain the number of clusters.

[0058] In some embodiments, each third feature data set corresponds to a cluster.

[0059] In some embodiments, based on the first correlation analysis result, a tree diagram is generated for the first feature data set of factors affecting wrong answers according to a hierarchical clustering algorithm to obtain the correlation distribution between the evaluation index features. The shortest distance linking method can be selected: d(Ccluster,Dcluster)=min{d(c,d):c∈C,d∈D}, where d(Ccluster,Dcluster) represents the distance between cluster C and cluster D, and c,d represent a point c in cluster C and a point d in cluster D.

[0060] The third feature data set is subjected to correlation evaluation using a pre-trained correlation evaluation model to obtain a second correlation analysis result between the third feature data sets, and a correlation score is obtained based on the first correlation analysis result and the second correlation analysis result.

[0061] In some embodiments, a pre-trained correlation evaluation model is used to obtain the correlation between the third data sets.

[0062] In some embodiments, the correlation distribution between the third feature sets is generated based on the first correlation analysis results to determine whether the feature sets are correlated, thereby solving the problem of low accuracy and lack of global perspective in statistical calculation and averaging based on the correlation analysis results.

[0063] In some embodiments, after obtaining the first correlation analysis result and the second correlation analysis result, the two correlation analysis results can be averaged to obtain a correlation score, or the two correlation analysis results can be weighted and summed to obtain a correlation score. The method of correlation scoring is not limited in this application.

[0064] As a possible implementation method, based on the above embodiment, a preset correlation algorithm is used to perform correlation evaluation on each wrong question influencing factor in the first feature data set, to obtain a first correlation analysis result between features in the first feature data set, and to obtain the correlation strength between the wrong question influencing factor features according to the first correlation analysis result, including: The first correlation analysis result is calculated by a preset correlation algorithm: Formula 1 In formula 1, Indicates the features in the first feature data set and Features The average value of represents the features in the first feature data set, Representation characteristics and Features The first correlation analysis results between; According to the first correlation analysis results, the correlation strength between the characteristics of the factors affecting wrong questions is obtained, among which, The value range is [0, 1].

[0065] In some embodiments, The closer the value is to 1, the weaker the correlation between the characteristics of the factors affecting wrong questions and the characteristics of wrong questions.

[0066] In some embodiments, by utilizing a preset correlation algorithm to perform a correlation evaluation on each factor affecting wrong questions in the first feature data set, a first correlation analysis result between the features in the first feature data set is obtained, thereby facilitating obtaining the correlation between the factors affecting wrong questions and the features of wrong questions. The larger the first correlation analysis result, the weaker the correlation between the features of the factors affecting wrong questions and the features of wrong questions.

[0067] As a possible implementation method, based on the above embodiment, a pre-trained correlation evaluation model is used to perform correlation evaluation on the third feature data set to obtain a second correlation analysis result between the third feature data sets, and a correlation score is obtained based on the first correlation analysis result and the second correlation analysis result, including: The second correlation analysis result is calculated by the pre-trained correlation evaluation model: Formula 2 In formula 2, represents the second correlation analysis result, represents a constant term, represents the partial regression coefficient of the preset correlation evaluation model, Represents the characteristics of the third feature data set; A correlation score is obtained based on the first correlation analysis result and the second correlation analysis result.

[0068] In some embodiments, by obtaining a correlation score based on the first correlation analysis result and the second correlation analysis result, it is possible to more finely screen out truly strongly correlated features.

[0069] As a possible implementation, based on the above embodiment, a preset wrong question correlation evaluation model is used to evaluate the correlation scores between the features of objective factors that cause students to make wrong questions, and the features of objective factors that cause students to make wrong questions are sorted from large to small according to the correlation scores, and the knowledge point corresponding to the feature with the largest correlation score is determined as the first target knowledge point cluster, and the question bank generation method further includes: Determine a basic knowledge point cluster based on the extraction of knowledge points in the textbook text, determine a first important knowledge point cluster based on the importance of knowledge points in the basic knowledge point cluster, and determine a second important knowledge point cluster based on the extraction of knowledge points in the history question bank and the difficulty level of the knowledge points in the history question bank; In some embodiments, the content in the textbook text can be uploaded to the first knowledge point analysis module, and the knowledge points corresponding to the content in the textbook text can be extracted using text analysis. Specifically, the knowledge points corresponding to the content in the textbook text can be extracted using keywords. Before performing text analysis, the content in the textbook text can also be preprocessed, such as word segmentation, removal of stop words, etc.

[0070] In some embodiments, knowledge points are graded according to the associations between knowledge points to obtain graded identifications corresponding to the knowledge points. Specifically, graded identification refers to assigning specific marks to knowledge points at each level in a multi-level classification structure for easy distinction and management. The graded identification of knowledge points can clearly display the hierarchical relationship of knowledge points, which is helpful for the construction of a knowledge point network. Knowledge points are used as nodes in the knowledge network, and knowledge points classified by chapters in the textbook are used as first-level classification nodes, corresponding to the main nodes in the knowledge point network; knowledge points classified by sections in the textbook are used as second-level classification nodes, corresponding to the slave nodes of the first-level classification nodes in the knowledge point network, and are used to represent each subsection in each chapter; knowledge points classified by unit results can be used as third-level classification nodes, and the corresponding knowledge points are slave nodes of the second-level classification nodes in the network, which are used to further subdivide the specific knowledge points under each unit. The associations between knowledge points refer to the above-mentioned chapters, sections and units.

[0071] In some embodiments, the grading identifiers corresponding to the knowledge points may be English letters, such as A for a first-level classification node, B for a second-level classification node, and C for a third-level classification node.

[0072] In some embodiments, the hierarchical identification corresponding to the knowledge point is encoded to obtain the identification code value of the knowledge point at each level. Specifically, each knowledge point corresponds to a unique identification code value, and the identification code value can be one or more combinations of multiple identification forms such as numbers, letters, special symbols, etc. When English letters are used as hierarchical identification, different knowledge points on the same hierarchical node can be encoded using numbers. For example, if there are 3 first-level classification nodes, they can be encoded as A1, A2, and A3 respectively. If there are two second-level classification nodes, they can be encoded as B1 and B2. At the same time, the first-level classification nodes associated with B1 and B2 also need to be encoded.

[0073] In some embodiments, based on the hierarchical identification corresponding to the knowledge point and the identification code value of the knowledge point at each level, the importance identification vector set corresponding to the knowledge point is determined, and the basic knowledge point cluster is generated according to the identification vector set. Specifically, the first-level classification node appears in the form of a single unit identification vector, and the multi-level classification node appears in the form of a combination of multiple units of identification vectors; the knowledge points in the textbook can be divided into n levels, such as Figure 3 As shown, Figure 3 Schematic diagram of the knowledge point network provided in the embodiment of the present disclosure. When n is 3, the corresponding knowledge point cluster U=[An, Bm, Ck]. Specifically, the first-level classification An (A in the figure) in the knowledge point network is used as the main node, and is classified by chapter to obtain An=[A1, A2...An]; the second-level classification Bm is used as the slave node of An, and is classified by section to obtain Bm=[A1b1, A1b2, A2b1...Anbm]; the third-level classification Ck is used as the slave node of Bm, and is classified by unit results to obtain Ck=[A1b1c1, A1b2c2, A1b2c3...Anbmck]. The corresponding identification vector is generated according to the identification code, and the knowledge point network is constructed based on the identification vector. The knowledge point network is used to represent the relationship between knowledge points, and then multiple knowledge point networks are combined into a comprehensive identification vector set, namely the basic knowledge point cluster Uori.

[0074] In some embodiments, a pre-trained evaluation model can be used to evaluate the importance of each knowledge point in the basic knowledge point cluster, and then the first important knowledge point cluster is determined according to the importance, such as Figure 4 As shown, Figure 4 This is a schematic diagram of the screening process provided in the embodiment of the present disclosure, specifically screening out knowledge points with higher importance from the knowledge points in the basic knowledge point cluster Uori to obtain the first important knowledge point cluster Unew.

[0075] In some embodiments, based on all the knowledge points involved in the history question bank and the pre-trained important knowledge point evaluation model, the difficulty level score of each of the knowledge points involved in the history question bank is determined. Specifically, based on the history question bank, the basic knowledge point cluster Zori=[Wn, Rm, Hk] corresponding to the question bank can be generated according to the identification vector (Wn, Rm, Hk) corresponding to the knowledge point in the question bank. The difficulty level score of each of the knowledge points involved in the history question bank can be calculated by the pre-trained important knowledge point evaluation model, and its mathematical expression is as follows: Formula 3 In formula 3, Gj is the number of times the test point knowledge point appears, Gq is the number of times the error frequency of the test point knowledge point appears, Gr is the number of branches of the test point knowledge point, is the weight score corresponding to each indicator, Rate the difficulty level of the knowledge point.

[0076] Based on the difficulty level score of each knowledge point in all the knowledge points involved in the historical question bank and the second preset screening threshold, all the knowledge points involved in the historical question bank are screened, and the screened knowledge point set is determined as the second important knowledge point cluster. The second preset screening threshold is an initial value set for the difficulty level corresponding to the knowledge point. Specifically, the second preset screening threshold can be obtained by the same method as the first preset screening threshold, or by a method different from the first preset screening threshold, that is, one of the methods such as custom, average method, median, etc., or by a plurality of methods such as custom, average method, median, etc., and then averaging multiple values ​​obtained based on a plurality of methods. This embodiment does not limit this. A second preset screening threshold T2 can be determined. When the aforementioned yG>T2, important knowledge points are screened out to obtain a second important knowledge point cluster. The selection of the second preset screening threshold is related to the importance of the knowledge points in the second important knowledge point cluster. If the second preset screening threshold is too small, it will result in a large number of knowledge points in the second important knowledge point cluster, and the importance is not high. If the second preset threshold is too large, it will result in a small number of knowledge points in the second important knowledge point cluster, and the importance of the knowledge points in the second important knowledge point cluster is high, but some relatively important knowledge points may be excluded. Furthermore, for example, in the scenario of teacher preparation, a second preset threshold that is too large will cause the scope of teacher preparation to become smaller, which is insufficient to cover most of the important knowledge points. For example, in the review scenario, a second preset threshold that is too large will cause some important test points to not be covered.

[0077] Determine the same knowledge points in the first target knowledge point cluster, the first important knowledge point cluster, and the second important knowledge point cluster as the second target knowledge point cluster; In some embodiments, one of the algorithms such as Jaccard similarity algorithm, cosine similarity algorithm, Euclidean distance algorithm and Manhattan distance algorithm can be used to determine the same knowledge points in the first target knowledge point cluster, the first important knowledge point cluster and the second important knowledge point cluster, and then determine the second target knowledge point cluster.

[0078] Based on the basic knowledge point cluster and the first target knowledge point cluster, a question bank corresponding to questions of different difficulty levels is generated and mapped on the learning map, including: Based on the basic knowledge point clusters and the second target knowledge point clusters, a question bank corresponding to questions of different difficulty levels is generated and mapped on the learning map.

[0079] In some embodiments, Figure 5 As shown, Figure 5A schematic diagram of a question bank marking method provided in an embodiment of the present disclosure, in which a basic knowledge point cluster in the question bank is identified by a white mark as the first information, and a second target knowledge point cluster is identified by a black mark as the second information, and the question bank includes questions corresponding to level one (easy difficulty), level two (medium difficulty), and level three (difficult difficulty).

[0080] In some embodiments, since the second target knowledge point cluster is determined by the same knowledge points in the first target knowledge point cluster, the first important knowledge point cluster and the second important knowledge point cluster, it is more accurate to generate questions of different difficulty levels using the basic knowledge point cluster and the second target knowledge point cluster. Therefore, the question bank composed of questions of different difficulty levels is also more accurate. Finally, the more accurate question bank is mapped on the learning map, which can improve students' learning effects.

[0081] Corresponding to the above-mentioned question bank generation method, the present invention also provides a question bank generation device. Since the device embodiment of the present invention corresponds to the above-mentioned method embodiment, details not disclosed in the device embodiment can be referred to the above-mentioned method embodiment, and will not be repeated in the present invention.

[0082] Figure 6 A schematic diagram of a structure of a question bank generating device provided by an embodiment of the present disclosure is shown in FIG. Figure 6 As shown, the question bank generating device 600 includes: A collection unit 601 is used to collect a first feature data set including factors affecting students' wrong answers by using a smart education terminal, where the factors affecting students' wrong answers include subjective factors and objective factors; A screening unit 602 is used to screen out the features of objective factors that cause students to make mistakes in the questions from the first feature data set of factors affecting making mistakes in the questions by using a hierarchical clustering algorithm; A determination unit 603 is used to evaluate the correlation scores between the features of the objective factors that cause students to make mistakes in the questions using a preset wrong question correlation evaluation model, and sort the features of the objective factors that cause students to make mistakes in the questions from large to small according to the correlation scores, and determine the knowledge point corresponding to the feature with the largest correlation score as the first target knowledge point cluster; The generation unit 604 is used to generate a question bank corresponding to questions of different difficulty levels based on the basic knowledge point cluster and the first target knowledge point cluster and map them on the learning map. The learning map includes one or more key learning grids, which are grids corresponding to the first target knowledge point cluster. The basic knowledge point cluster is obtained by extracting knowledge points from the textbook based on the smart education terminal.

[0083] In some embodiments of the present disclosure, the collection unit 601 is used to: Collect students' test-taking videos and test-taking status based on smart education terminals; The facial expression recognition algorithm and eye tracking algorithm are used to identify the subjective factor feature data set that causes students to make mistakes in the test-taking video footage, and the objective factor feature data set that causes students to make mistakes is obtained by using the students' test-taking situation. Based on the subjective factor feature data set and the objective factor feature data set, a first feature data set including factors influencing students' wrong answers is determined.

[0084] In some embodiments of the present disclosure, the collection unit 601 is used to: The subjective factor feature data set and the objective factor feature data set are combined to obtain a second feature data set containing the factors affecting students' wrong answers; All feature data in the second feature data set are preprocessed and normalized to obtain a first feature data set containing factors affecting students' wrong answers.

[0085] In some implementations of the present disclosure, the determining unit 603 is configured to: Using a preset correlation algorithm to perform correlation evaluation on each wrong question influencing factor in the first feature data set, obtaining a first correlation analysis result between features in the first feature data set, and obtaining the correlation strength between the features of the wrong question influencing factors according to the first correlation analysis result; Based on the first correlation analysis result, generating a dendrogram from the first feature data set of factors affecting wrong answers according to a hierarchical clustering algorithm, determining the number of groups of features in the first feature data set according to the number of clusters in the dendrogram, and allocating the feature data in the first feature data set to corresponding groups to obtain at least two third feature data sets; The third feature data set is subjected to correlation evaluation using a pre-trained correlation evaluation model to obtain a second correlation analysis result between the third feature data sets, and a correlation score is obtained based on the first correlation analysis result and the second correlation analysis result.

[0086] In some implementations of the present disclosure, the determining unit 603 is configured to: The first correlation analysis result is calculated by a preset correlation algorithm:

[0087] in, Indicates the features in the first feature data set and Features The average value of represents the features in the first feature data set, Representation characteristics and Features The first correlation analysis results between; According to the first correlation analysis results, the correlation strength between the characteristics of the factors affecting wrong questions is obtained, among which, The value range is [0, 1].

[0088] In some implementations of the present disclosure, the determining unit 603 is configured to: The second correlation analysis result is calculated by the pre-trained correlation evaluation model:

[0089] in, represents the second correlation analysis result, represents a constant term, represents the partial regression coefficient of the preset correlation evaluation model, Represents the characteristics of the third feature data set; A correlation score is obtained based on the first correlation analysis result and the second correlation analysis result.

[0090] In some embodiments of the present disclosure, the question bank generating device 600 further includes a second target knowledge point cluster determining unit, and the second target knowledge point cluster determining unit is used to: Determine a basic knowledge point cluster based on the extraction of knowledge points in the textbook text, determine a first important knowledge point cluster based on the importance of knowledge points in the basic knowledge point cluster, and determine a second important knowledge point cluster based on the extraction of knowledge points in the history question bank and the difficulty level of the knowledge points in the history question bank; Determine the same knowledge points in the first target knowledge point cluster, the first important knowledge point cluster, and the second important knowledge point cluster as the second target knowledge point cluster; In some embodiments of the present disclosure, the generating unit 604 is configured to: Based on the basic knowledge point clusters and the second target knowledge point clusters, a question bank corresponding to questions of different difficulty levels is generated and mapped on the learning map.

[0091] It should be noted that the above explanation of the method embodiment is also applicable to the device of this embodiment, and the principle is the same, which is not limited in this embodiment.

[0092] Based on the above Figures 1 to 5 The method shown in the figure, accordingly, this embodiment also provides a computer program product, including a computer program, which implements the above-mentioned Figures 1 to 5 The method shown.

[0093] Based on the above Figures 1 to 5 The method shown in the embodiment also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the above-mentioned Figures 1 to 5The method shown.

[0094] Based on this understanding, the technical solution of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, USB flash drive, mobile hard disk, etc.), including a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute the methods of various implementation scenarios of the present application.

[0095] like Figure 7 The figure shows a hardware structure diagram of an electronic device of the present invention, including: at least one processor 701; and, A memory 702 that is communicatively connected to at least one of the processors 701; wherein, The memory 702 stores instructions that can be executed by at least one of the processors. The instructions are executed by at least one of the processors so that the at least one processor can execute the question bank generation method as described above.

[0096] Figure 7 A processor 701 is taken as an example.

[0097] The electronic device may further include: an input device 703 and a display device 704 .

[0098] The processor 701, the memory 702, the input device 703 and the display device 704 may be connected via a bus or other means, and the figure takes the connection via a bus as an example.

[0099] The memory 702 is a non-volatile computer-readable storage medium that can be used to store non-volatile software programs, non-volatile computer executable programs and modules, such as program instructions / modules corresponding to the review content generation method in the embodiment of the present application, for example, Figures 1 to 5 The processor 701 executes various functional applications and data processing by running the non-volatile software programs, instructions and modules stored in the memory 702, that is, the question bank generation method in the above embodiment is implemented.

[0100] The memory 702 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required by at least one function; the data storage area may store data created according to the use of the review content generation method, etc. In addition, the memory 702 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage devices. In some embodiments, the memory 702 may optionally include a memory remotely arranged relative to the processor 701, and these remote memories may be connected to the device for executing the review content generation method via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0101] The input device 703 can receive user clicks and generate signal inputs related to user settings and function controls of the review content generation method. The display device 704 can include display devices such as a display screen.

[0102] The one or more modules are stored in the memory 702, and when executed by the one or more processors 701, the question bank generation method in any of the above method embodiments is executed.

[0103] Optionally, the above-mentioned physical device may also include a user interface, a network interface, a camera, a radio frequency (RF) circuit, a sensor, an audio circuit, a WI-FI module, etc. The user interface may include a display, an input unit such as a keyboard, etc., and the optional user interface may also include a USB interface, a card reader interface, etc. The network interface may optionally include a standard wired interface, a wireless interface (such as a WI-FI interface), etc.

[0104] Those skilled in the art will appreciate that the above-mentioned physical device structure provided in this embodiment does not constitute a limitation on the physical device, and may include more or fewer components, or a combination of certain components, or different arrangements of components.

[0105] 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 above-mentioned physical device, and supports the operation of the information processing program and other software and / or programs. The network communication module is used to realize the communication between the components inside the storage medium, and the communication with other hardware and software in the information processing physical device.

[0106] Through the description of the above implementation methods, the technicians in this field can clearly understand that the present application can be implemented by means of software plus the necessary general hardware platform, or by hardware. By applying the scheme of this embodiment, compared with the current existing technology, this embodiment obtains a first historical data set of multiple history students, and builds a learning performance prediction model based on the first historical data set, and the first historical data set includes the learning performance of multiple history students; obtains a second historical data set of multiple history students, and builds a learning performance prediction model based on the second historical data set, and the second historical data set includes the learning performance of multiple history students; based on the learning performance prediction model and the learning performance prediction model, a question bank generation model is constructed; based on the question bank generation model, the learning ability of the target student is determined, and the question bank generation model is constructed by constructing the learning performance prediction model and the learning performance prediction model, so that the question bank generation model can comprehensively evaluate the learning performance of the target student in terms of objective aspects and subjective aspects of learning performance, improve the accuracy and diversity of the question bank generation model, and help schools or parents to more intuitively and clearly understand the all-round development of students' learning ability, timely discover the reasons for the decline in students' grades, and intervene as soon as possible to guide targeted learning counseling programs.

[0107] It should be noted that, in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0108] The above is only a specific implementation of the present application, so that those skilled in the art can understand or implement the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments described herein, but will conform to the widest scope consistent with the principles and novel features applied for herein.

Claims

1. A method for generating a question bank, characterized in that: include: Using a smart education terminal to collect a first feature data set including factors affecting students' wrong answers, wherein the factors affecting students' wrong answers include subjective factors and objective factors; Using a hierarchical clustering algorithm, the characteristics of objective factors that cause students to make mistakes in the questions are screened out from the first characteristic data set of the factors affecting making mistakes in the questions; Using a preset wrong question correlation evaluation model, the correlation scores between the features of the objective factors that cause students to make wrong questions are evaluated, and the features of the objective factors that cause students to make wrong questions are sorted from large to small according to the correlation scores, and the knowledge point corresponding to the feature with the largest correlation score is determined as the first target knowledge point cluster; Based on the basic knowledge point cluster and the first target knowledge point cluster, a question bank corresponding to questions of different difficulty levels is generated and mapped on a learning map. The learning map includes one or more key learning grids, which are grids corresponding to the first target knowledge point cluster. The basic knowledge point cluster is obtained by extracting knowledge points from the textbook based on the smart education terminal.

2. The method according to claim 1, characterized in that The method of using the smart education terminal to collect a first feature data set including factors affecting students' wrong answers includes: Collect students' test-taking videos and test-taking status based on smart education terminals; A facial expression recognition algorithm and an eye tracking algorithm are used to identify a subjective factor feature data set that causes the student to make mistakes in the test-taking video of the student, and an objective factor feature data set that causes the student to make mistakes is obtained using the test-taking situation of the student; Based on the subjective factor characteristic data set and the objective factor characteristic data set, a first characteristic data set including factors influencing students' wrong answers is determined.

3. The method according to claim 2, characterized in that The determining, based on the subjective factor feature data set and the objective factor feature data set, of the first feature data set containing the factors affecting students' wrong answers includes: Combining the subjective factor feature data set and the objective factor feature data set to obtain a second feature data set containing factors affecting students' wrong answers; All feature data in the second feature data set are preprocessed and normalized to obtain the first feature data set containing the factors affecting students' wrong answers.

4. The method according to claim 1, characterized in that The using of a preset wrong question correlation evaluation model to evaluate the correlation scores between the features of the objective factors that cause students to make wrong questions includes: Using a preset correlation algorithm to perform correlation evaluation on each wrong question influencing factor in the first feature data set, obtaining a first correlation analysis result between features in the first feature data set, and obtaining the correlation strength between the features of the wrong question influencing factors according to the first correlation analysis result; Based on the first correlation analysis result, generating a dendrogram for the first feature data set of the factors affecting wrong answers according to a hierarchical clustering algorithm, determining the number of groups of features in the first feature data set according to the number of clusters in the dendrogram, and allocating the feature data in the first feature data set to corresponding groups to obtain at least two third feature data sets; The third feature data set is subjected to correlation evaluation using a pre-trained correlation evaluation model to obtain a second correlation analysis result between the third feature data sets, and the correlation score is obtained based on the first correlation analysis result and the second correlation analysis result.

5. The method according to claim 4, characterized in that The method of using a preset correlation algorithm to perform a correlation evaluation on each wrong answer factor in the first feature data set to obtain a first correlation analysis result between features in the first feature data set, and obtaining the correlation strength between the features of the wrong answer factor according to the first correlation analysis result, includes: The first correlation analysis result is calculated by a preset correlation algorithm: in, Indicates the features in the first feature data set and Features The average value of represents the features in the first feature data set, Representation characteristics and Features The first correlation analysis results between; The correlation strength between the characteristics of the factors affecting wrong answers is obtained according to the first correlation analysis result, wherein: The value range is [0, 1].

6. The method according to claim 4, characterized in that The using a pre-trained correlation evaluation model to perform correlation evaluation on the third feature data sets to obtain a second correlation analysis result between the third feature data sets, and obtaining the correlation score based on the first correlation analysis result and the second correlation analysis result, comprises: The second correlation analysis result is calculated by a pre-trained correlation evaluation model: in, represents the second correlation analysis result, represents a constant term, represents the partial regression coefficient of the preset correlation evaluation model, Represents the characteristics of the third feature data set; The correlation score is obtained based on the first correlation analysis result and the second correlation analysis result.

7. The method according to claim 1, characterized in that After evaluating the correlation scores between the features of the objective factors that cause students to make mistakes by using the preset wrong question correlation evaluation model, and sorting the features of the objective factors that cause students to make mistakes by sorting them from large to small according to the correlation scores, and determining the knowledge point corresponding to the feature with the largest correlation score as the first target knowledge point cluster, the method further includes: Determine a basic knowledge point cluster based on the extraction of knowledge points in the textbook text, determine a first important knowledge point cluster based on the importance of knowledge points in the basic knowledge point cluster, and determine a second important knowledge point cluster based on the extraction of knowledge points in the history question bank and according to the difficulty level of the knowledge points in the history question bank; Determine the same knowledge points in the first target knowledge point cluster, the first important knowledge point cluster and the second important knowledge point cluster as a second target knowledge point cluster; The step of generating a question bank corresponding to questions of different difficulty levels based on the basic knowledge point cluster and the first target knowledge point cluster and mapping the question bank on the learning map includes: Based on the basic knowledge point cluster and the second target knowledge point cluster, a question bank corresponding to questions of different difficulty levels is generated and mapped on the learning map.

8. A question bank generating device, characterized in that: include: A collection unit, configured to collect, by using a smart education terminal, a first feature data set including factors affecting students' mistakes in answering questions, wherein the factors affecting students' mistakes in answering questions include subjective factors and objective factors; A screening unit, used to screen out the features of objective factors that cause students to make mistakes in the questions from the first feature data set of the factors affecting making mistakes in the questions by using a hierarchical clustering algorithm; A determination unit is used to evaluate the correlation scores between the features of the objective factors that cause students to make mistakes in the questions by using a preset wrong question correlation evaluation model, and sort the features of the objective factors that cause students to make mistakes in the questions from large to small according to the correlation scores, and determine the knowledge point corresponding to the feature with the largest correlation score as the first target knowledge point cluster; A generation unit is used to generate a question bank corresponding to questions of different difficulty levels based on a basic knowledge point cluster and the first target knowledge point cluster and map them on a learning map, wherein the learning map includes one or more key learning grids, wherein the key learning grids are grids corresponding to the first target knowledge point cluster, and the basic knowledge point cluster is obtained by extracting knowledge points from the textbook based on the smart education terminal.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

10. An electronic device comprising a storage medium, a processor, and a computer program stored in the storage medium and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 7 is implemented.

Images