Personalized teaching and training auxiliary method and system, terminal equipment and medium

Abstract

The embodiment of the invention provides a personalized teaching and training auxiliary method and system, terminal equipment and a medium, and belongs to the technical field of data processing. The method comprises the following steps: collecting historical question-doing data and learning process data corresponding to a target student; performing data analysis on the historical exercise data and the learning process data to obtain an initial mastering result of the target student under the target knowledge point; performing question recommendation on the target student according to the initial mastering result to obtain a target question corresponding to the target student; obtaining a target answering result corresponding to the target question, and performing data evaluation on the target answering result to obtain a corresponding target mastering result of the target student under the target knowledge point; performing learning planning on the target student according to the target mastering result to obtain an initial learning scheme corresponding to the target student; obtaining a wrong question set corresponding to the target student according to the initial learning scheme, and performing data analysis on the wrong question set to obtain learning blind spots; and adjusting the initial learning scheme according to the learning blind spots to obtain a target learning scheme.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a personalized teaching and training auxiliary method, system, terminal device and medium. Background Art

[0002] With the continuous advancement of science and technology, the medical industry has an increasing demand for medical talents. As a key link in cultivating these talents, medical education has attracted much attention for its teaching methods and effects. Traditional medical education mainly relies on teachers' lectures and students' passive acceptance. Although this method can convey a lot of knowledge, it is often difficult to stimulate students' interest and potential. At the same time, it is also insufficient in evaluating and giving feedback on students' learning effects, which limits learning efficiency and quality. To solve these problems, the field of medical education has begun to introduce emerging technologies, such as multimedia teaching, online teaching, and virtual reality teaching. These new technologies can enhance the fun and interactivity of teaching, improve students' learning interest and effect, and monitor and evaluate the learning process in real time, providing more accurate feedback. However, current technologies still cannot be personalized according to students' individual differences, and it is difficult to effectively guide and intervene in students' learning behaviors, which seriously affects students' learning efficiency and quality. Summary of the invention

[0003] The main purpose of the embodiments of the present invention is to provide a personalized teaching and training assistance method, system, terminal device and medium, aiming to solve the problem that the related technology cannot make personalized adjustments according to the individual differences of students, and it is difficult to effectively guide and intervene in students' learning behavior, thereby seriously affecting students' learning efficiency and quality.

[0004] In a first aspect, an embodiment of the present invention provides a personalized teaching and training auxiliary method, including:

[0005] Collect relevant learning data corresponding to the target students, wherein the relevant learning data includes historical question-solving data and learning process data;

[0006] Performing data analysis on the historical question-solving data and the learning process data to obtain the initial mastery results corresponding to the target students under the target knowledge points;

[0007] Recommending questions to the target students according to the initial mastery results to obtain target questions corresponding to the target students, wherein the target questions include true-or-false questions, single-choice questions and multiple-choice questions;

[0008] Obtaining a target answer result corresponding to the target question, and performing data evaluation on the target answer result to obtain a target mastery result corresponding to the target student under the target knowledge point;

[0009] Performing a learning plan for the target student according to the target mastering result to obtain an initial learning plan corresponding to the target student;

[0010] Obtaining a set of wrong questions corresponding to the target student according to the initial learning plan, and performing data analysis on the set of wrong questions to obtain a learning blind spot corresponding to the target student;

[0011] The initial learning plan is adjusted according to the learning blind spots to obtain a target learning plan corresponding to the target student.

[0012] In a second aspect, an embodiment of the present invention provides a personalized teaching and training auxiliary system, including:

[0013] A data acquisition module is used to collect relevant learning data corresponding to the target students, wherein the relevant learning data includes historical question-solving data and learning process data;

[0014] A data analysis module is used to analyze the historical question-solving data and the learning process data to obtain the initial mastery results of the target students under the target knowledge points;

[0015] A topic recommendation module is used to recommend topics to the target students according to the initial mastery results, and obtain target topics corresponding to the target students, wherein the target topics include true-or-false questions, single-choice questions and multiple-choice questions;

[0016] A data evaluation module is used to obtain a target answer result corresponding to the target question, and to perform data evaluation on the target answer result to obtain a target mastery result corresponding to the target student under the target knowledge point;

[0017] A program recommendation module is used to make a learning plan for the target student according to the target mastering result, and obtain an initial learning program corresponding to the target student;

[0018] A wrong question analysis module, used to obtain a wrong question set corresponding to the target student according to the initial learning plan, and perform data analysis on the wrong question set to obtain a learning blind spot corresponding to the target student;

[0019] The program adjustment module is used to adjust the initial learning program according to the learning blind spots to obtain the target learning program corresponding to the target students.

[0020] In a third aspect, an embodiment of the present invention further provides a terminal device, comprising a processor, a memory, a computer program stored in the memory and executable by the processor, and a data bus for realizing connection and communication between the processor and the memory, wherein when the computer program is executed by the processor, the steps of any one of the personalized teaching and training assistance methods provided in the specification of the present invention are implemented.

[0021] In a fourth aspect, an embodiment of the present invention further provides a storage medium for computer-readable storage, characterized in that the storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps of any personalized teaching and training assistance method provided in the specification of the present invention.

[0022] The embodiment of the present invention provides a personalized teaching and training auxiliary method, system, terminal device and medium, the method includes: by analyzing the historical question-solving data and learning process data of the target student, the target student's knowledge mastery and learning habits can be understood, so that the target student's initial mastery result corresponding to the target knowledge point, and then the target student's initial mastery result is recommended according to the target student's initial mastery result, which can effectively improve the pertinence and effectiveness of the exercise, help students to focus on the target knowledge point, and thus improve the overall learning effect. Then obtain the target answer result corresponding to the target question, perform data evaluation on the target answer result to obtain the target mastery result corresponding to the target student under the target knowledge point, and make a learning plan for the target student according to the target mastery result to obtain the initial learning plan corresponding to the target student; obtain the wrong question set corresponding to the target student according to the initial learning plan, and perform data analysis on the wrong question set to obtain the learning blind spot corresponding to the target student; adjust the initial learning plan according to the learning blind spot to obtain the target learning plan corresponding to the target student. Analyzing the wrong questions and identifying the blind spots in learning can help the target students find their own omissions and misunderstandings in learning, and make targeted improvements, so as to adjust the initial learning plan according to the blind spots in learning, ensure that the target learning plan can adapt to the actual learning situation of the target students, continuously optimize the learning strategy, and improve the learning effect. Finally, through personalized and targeted target learning plans, the target students can study more efficiently, save time and energy, and achieve better learning results in a shorter period of time. It also solves the problem that related technologies cannot be personalized according to individual differences of students, and it is difficult to effectively guide and intervene in students' learning behaviors, which seriously affects students' learning efficiency and quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0024] Figure 1 A flowchart of a personalized teaching and training auxiliary method provided by an embodiment of the present invention;

[0025] Figure 2A schematic diagram of the module structure of a personalized teaching and training auxiliary system provided by an embodiment of the present invention;

[0026] Figure 3 A schematic block diagram of the structure of a terminal device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0027] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0028] The flowcharts shown in the accompanying drawings are only examples and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may also be decomposed, combined or partially merged, so the actual execution order may change according to actual conditions.

[0029] It should be understood that the terms used in this specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the specification of the present invention and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0030] The embodiments of the present invention provide a personalized teaching and training assistance method, system, terminal device and medium. The personalized teaching and training assistance method can be applied to a terminal device, which can be an electronic device such as a tablet computer, a laptop computer, a desktop computer, a personal digital assistant and a wearable device. The terminal device can be a server or a server cluster.

[0031] Some embodiments of the present invention are described in detail below in conjunction with the accompanying drawings. In the absence of conflict, the following embodiments and features in the embodiments can be combined with each other.

[0032] Please refer to Figure 1 , Figure 1 A flowchart of a personalized teaching and training assistance method provided in an embodiment of the present invention.

[0033] like Figure 1 As shown, the personalized teaching and training assistance method includes steps S101 to S107.

[0034] Step S101: Collect relevant learning data corresponding to the target student, wherein the relevant learning data includes historical question-solving data and learning process data.

[0035] Exemplarily, through big data analysis technology, relevant learning data of target students are collected and sorted, and relevant learning data include but are not limited to historical question-solving data and learning process data corresponding to target students, wherein historical question-solving data include question information corresponding to basic knowledge points and user answer information corresponding to the target students’ answers to the question information, and correct answer information corresponding to the question information, etc. Learning process data include but are not limited to learning effects, learning habits, and answer time information corresponding to the question information, etc.

[0036] For example, if a target knowledge point depends on multiple basic knowledge points, when evaluating the target students' initial mastery of the target knowledge point, we can rely on the target students' historical question-solving data and learning process data corresponding to the multiple basic knowledge points for preliminary evaluation.

[0037] Step S102: perform data analysis on the historical question-solving data and the learning process data to obtain the initial mastery results corresponding to the target students under the target knowledge points.

[0038] Exemplarily, historical test data includes data on questions related to basic knowledge points completed by the target students, such as answering time, accuracy, and wrong question records. Historical test data can reflect the target students' mastery of basic knowledge points and their answering habits. Learning process data covers the target students' study time, study frequency, completed learning modules, and interactive activities participated in. Historical test data helps understand the target students' learning behavior, learning involvement, and participation in learning activities.

[0039] Exemplarily, historical question-solving data and learning process data are analyzed using a classification model to predict the target student's mastery level on basic knowledge points. Thus, the initial mastery level corresponding to each basic knowledge point associated with the target knowledge point can be obtained. The initial mastery level can be a quantitative score (such as mastery percentage) or a qualitative description (such as good, medium, insufficient). Then, according to the importance of each basic knowledge point in the target knowledge point, the initial mastery level is weighted. For example, if a basic knowledge point is particularly critical in the study of the target knowledge point, a higher weight can be assigned to it. The weighted basic knowledge point mastery level is summed to obtain the target student's initial mastery result on the target knowledge point. The initial mastery result is used to characterize the difficulty level when the target student can answer questions in the target knowledge point or the predicted user level of the target student under the target knowledge point.

[0040] In some embodiments, the data analysis of the historical problem-solving data and the learning process data to obtain the initial mastery result of the target student under the target knowledge point includes: obtaining the associated knowledge points corresponding to the target knowledge point in the historical problem-solving data and the historical scores corresponding to the historical problem-solving data; clustering the historical problem-solving data according to the historical scores and the associated knowledge points to obtain the data clustering results corresponding to the historical problem-solving data; obtaining the associated process data corresponding to each first sub-clustering result in the data clustering results from the learning process data; obtaining the first associated data corresponding to the associated knowledge point in the knowledge mastery state and obtaining the associated process data; obtaining the first associated data corresponding to the associated knowledge point in the knowledge mastery state; obtaining the associated process data corresponding to the first sub-clustering result in the data clustering result; obtaining the associated process data corresponding to the first sub-clustering result in the data clustering state ... second associated data corresponding to a knowledge point in a state where the knowledge is not mastered; using the first associated data and the second associated data to determine a distinction coefficient corresponding to the associated knowledge point for dividing the knowledge mastered state and the knowledge not mastered state; determining a distinction threshold corresponding to the associated knowledge point for dividing the knowledge mastered state and the knowledge not mastered state according to the distinction coefficient; calculating a target value corresponding to the associated process data according to the distinction coefficient, and determining the knowledge mastery result corresponding to the target student under the associated knowledge point according to the distinction threshold and the target value; determining the initial mastery result corresponding to the target knowledge point according to the knowledge mastery result corresponding to the associated knowledge point.

[0041] Exemplarily, the associated knowledge points are the same as the basic knowledge points. The knowledge content involved in the historical question-solving data in the question-solving process is disassembled to obtain the relevant knowledge information corresponding to the historical question-solving data, and then the associated knowledge points corresponding to the target knowledge points in the historical question-solving data and the historical scores corresponding to the historical question-solving data are obtained from the relevant knowledge information.

[0042] For example, the target knowledge point is based on associated knowledge point 1 and associated knowledge point 2. After the knowledge content involved in the historical question-solving data in the process of answering questions is disassembled, it is determined whether the historical question-solving data involves associated knowledge point 1 or associated knowledge point 2.

[0043] Exemplarily, the historical test data is preliminarily clustered according to the associated knowledge points to obtain the first clustering result, that is, each cluster in the first clustering result contains the same associated knowledge points. Then, each cluster in the first clustering result is clustered again according to the score distance between the historical scores, so as to obtain the data clustering result corresponding to the historical test data, that is, each cluster in the data clustering result contains the same associated knowledge points and similar historical scores.

[0044] Exemplarily, the historical question-solving data and the learning process data have a one-to-one correspondence, and the associated process data corresponding to each first sub-clustering result in the data clustering result is obtained from the learning process data.

[0045] Exemplarily, based on the first associated data corresponding to other students who have learned the associated knowledge points in a state of knowledge mastery and the second associated data corresponding to the associated knowledge points in a state of knowledge non-mastery, that is, the first associated data is the representative data corresponding to the associated knowledge points in a state of knowledge mastery, and the second associated data is the representative data corresponding to the associated knowledge points in a state of knowledge non-mastery.

[0046] Exemplarily, a support vector machine is used to determine the discrimination coefficient corresponding to each data feature in the first associated data and the second associated data, that is, each data feature in the first associated data and the second associated data can effectively separate the knowledge not mastered state and the knowledge mastered state of the associated knowledge point under the discrimination coefficient.

[0047] Exemplarily, the first associated data and the second associated data include information such as the student's learning time, learning frequency, completed learning modules, participated interactive activities, and accuracy of related questions on the associated knowledge points, and then the sorted first associated data and the second associated data are merged into a training data set, and the label of each data point (knowledge mastery state and knowledge non-mastery state) is included. The training data set is input into the support vector machine model for training. The model will learn how to distinguish between knowledge mastery state and knowledge non-mastery state based on data features. The support vector machine model divides the data into two categories by finding an optimal decision boundary (hyperplane). The coefficient of the decision boundary (i.e., the coefficient of the hyperplane) can be used as a discrimination coefficient.

[0048] Exemplarily, the distinction coefficient is calculated respectively with the first associated data and the second associated data to obtain the calculated values ​​corresponding to each first associated data and each second associated data, thereby determining the distinction threshold corresponding to the associated knowledge point division into the knowledge mastery state and the knowledge non-mastery state according to the calculated values.

[0049] Exemplarily, the discrimination coefficient is used for the associated process data to obtain the target value corresponding to each associated process data, so as to compare the discrimination threshold and the target value, and when the target value is greater than or equal to the discrimination threshold, it is determined that the mastery result of the associated knowledge point identified by the target student in the associated process data is the knowledge mastery state, and when the target value is less than the discrimination threshold, it is determined that the mastery result of the associated knowledge point identified by the target student in the associated process data is the knowledge non-mastery state; thereby, the mastery results of all associated process data are statistically analyzed, and then when more than 90% of the mastery results of all associated process data are the knowledge mastery state, the knowledge mastery result corresponding to the target student under the associated knowledge point is the knowledge mastery state, and when the mastery results of all associated process data are less than 90% and more than 80%, the knowledge mastery result corresponding to the target student under the associated knowledge point is the knowledge basic mastery state; otherwise, the knowledge mastery result corresponding to the target student under the associated knowledge point is the knowledge non-mastery state.

[0050] Exemplarily, the influence of each associated knowledge point on the target knowledge point is analyzed. This can be achieved through statistical analysis or modeling techniques. For example, regression analysis can be used to determine the contribution of each associated knowledge point to the mastery of the target knowledge point, thereby determining a quantitative influence index for each associated knowledge point to reflect its importance in the target knowledge point. According to the influence of each associated knowledge point, a corresponding target weight is assigned to it. The higher the influence of the associated knowledge point, the greater its weight should be.

[0051] Exemplarily, the determined target weight is used to perform a weighted summation on the knowledge mastery results of each associated knowledge point. A preset score threshold is determined to distinguish the mastery status of the target knowledge point. For example, a passing score or a standard score can be set. The calculated mastery score of the target knowledge point is compared with the preset value to determine whether the target knowledge point has reached the preset mastery standard. Based on the comparison result, the initial mastery status of the target knowledge point is determined. For example, if the mastery score of the target knowledge point is higher than the preset value, it is considered "mastered"; otherwise, it is considered "not mastered".

[0052] Specifically, by analyzing the target students' performance on related knowledge points, the accuracy and comprehensiveness of the assessment of their mastery of the target knowledge points can be improved, thereby providing support for the subsequent formulation of personalized learning plans for the target students to improve teaching effectiveness and student learning outcomes.

[0053] In some embodiments, the historical question-solving data is clustered according to the historical scores and the associated knowledge points to obtain data clustering results corresponding to the historical question-solving data, including: obtaining importance levels of the associated knowledge points to the historical question-solving data, the importance levels being used to characterize the degree of influence of the associated knowledge points on answering the historical question-solving data; initially clustering the historical question-solving data according to the importance levels and the associated knowledge points to obtain initial clustering results corresponding to the historical question-solving data; and segmenting each second sub-clustering result in the initial clustering result according to the historical score to obtain the data clustering results corresponding to the historical question-solving data.

[0054] Exemplarily, the importance level of related knowledge points in students' answering history problem-solving data is determined based on expert experience or student experience. For example, if the question cannot be answered due to lack of related knowledge points when answering the history problem-solving data, then the importance level of the related knowledge points in the history problem-solving data is determined to be very important; if there is other alternative knowledge to answer the question when there is lack of related knowledge points when answering the history problem-solving data, then the importance level of the related knowledge points in the history problem-solving data is determined to be important; if there is a lot of other alternative knowledge to answer the question when there is lack of related knowledge points when answering the history problem-solving data, then the importance level of the related knowledge points in the history problem-solving data is determined to be generally important, and so on.

[0055] Exemplarily, the importance level corresponding to each associated knowledge point corresponding to the historical question-solving data is obtained, so that the associated knowledge points and their corresponding importance levels are determined as data features, and then a clustering algorithm (such as K-means, hierarchical clustering, etc.) is used to preliminarily cluster the historical question-solving data based on the associated knowledge points and importance levels to obtain the initial clustering results corresponding to the historical question-solving data. Thus, the questions are clustered from the perspective of the similarity between the questions in the historical question-solving data, so as to conduct a more in-depth analysis.

[0056] For example, in the initial clustering results, each second sub-clustering result corresponds to a different historical score, and then the historical scores in the second sub-clustering results are normalized to obtain the normalized scores corresponding to the second sub-clustering results, and then the second sub-clustering results are segmented again according to the normalized scores, for example, by setting segmentation criteria. For example, some thresholds (such as high scores, medium scores, and low scores) are set to divide each second sub-cluster, so that each second sub-cluster is further segmented into smaller clusters according to the set segmentation criteria, and then the data clustering results are obtained, which can provide support for the subsequent analysis of the target students' mastery of various related knowledge points.

[0057] Specifically, we first cluster the data at the question level through the related knowledge points involved in the history test data, and then cluster the data from the student's perspective according to the historical scores of the target students in answering the questions, so as to obtain the data clustering results corresponding to the history test data. This can provide good support for more accurate acquisition of the target students' knowledge mastery results on related knowledge points in the subsequent process, and can also provide an important basis for subsequent personalized tutoring, thereby supporting more accurate educational intervention measures.

[0058] In some embodiments, determining the distinction threshold corresponding to the associated knowledge point for dividing the knowledge mastery state and the knowledge non-mastery state based on the distinction coefficient includes: calculating the first representation value corresponding to each of the first associated data based on the distinction coefficient, and calculating the second representation value corresponding to each of the second associated data based on the distinction coefficient; determining the distinction threshold corresponding to the associated knowledge point based on the first representation value and the second representation value.

[0059] Exemplarily, the discrimination coefficient is applied to the data feature corresponding to each first associated data to obtain the first characterization value corresponding to each first associated data, and the discrimination coefficient is applied to the data feature corresponding to each second associated data to obtain the second characterization value corresponding to each second associated data.

[0060] Exemplarily, the first characterization values ​​are summed to obtain a first summation value and the second characterization values ​​are summed to obtain a second summation value, and then the first summation value and the second summation value are added to obtain a target summation value, and then a first quantity corresponding to the first associated data and a second quantity corresponding to the second associated data are obtained, the first quantity and the second quantity are summed to obtain a target quantity, and finally the target summation value is divided by the target quantity to obtain a distinction threshold corresponding to the associated knowledge point.

[0061] Step S103: recommending questions to the target students based on the initial mastering results, and obtaining target questions corresponding to the target students, wherein the target questions include true-or-false questions, single-choice questions and multiple-choice questions.

[0062] For example, if the initial mastery results indicate that the target students have a low level of mastery of the target knowledge points, then when recommending questions to the target students, the questions will be relatively basic; if the initial mastery results indicate that the target students have a high level of mastery of the target knowledge points, then when recommending questions to the target students, the questions will tend to be more intensive and improved, and the target questions include true or false questions, single-choice questions, and multiple-choice questions.

[0063] For example, target questions include but are not limited to classic question types such as true or false questions, single-choice questions, multiple-choice questions, case questions, etc. For example, if the initial mastery result is a high degree of mastery, a small number of true or false questions, single-choice questions, and multiple-choice questions can be given, but more case questions can be given, so as to strengthen the target students' improvement of the target knowledge points.

[0064] In some embodiments, recommending topics to the target student based on the initial mastery result to obtain the target topics corresponding to the target student includes: determining the topic level corresponding to the target student under the target knowledge point based on the initial mastery result; determining a history student and the history learning data corresponding to the history student, and determining the related students corresponding to the target student based on the related learning data and the history learning data; obtaining the history learning topics corresponding to the related students under the target knowledge point and the level information corresponding to the history learning topics; screening the history learning topics according to the topic level and the level information to obtain the target topics corresponding to the target student.

[0065] Exemplarily, the initial mastery results of the target students are analyzed, and the initial mastery results reflect the mastery degree of the target knowledge points by the target students. Thus, according to the initial mastery results of the target students, the question levels related to the target knowledge points are set. The question levels may be based on factors such as the difficulty of the questions and the complexity of the target knowledge points involved. For example, the question levels can be divided into basic, advanced and advanced levels, and mapped to the initial mastery results of the target students.

[0066] Exemplarily, history students who have learned relevant knowledge points related to the target knowledge point are obtained, and then historical learning data corresponding to the history students are obtained, so that similarity is calculated between the relevant learning data and the historical learning data, and then related students corresponding to the target student among the history students are obtained.

[0067] Exemplarily, the history learning topics corresponding to the target knowledge points of the relevant students and the grade information corresponding to the history learning topics are obtained from the database, so as to filter out the history learning topics corresponding to the topic grades from the grade information and then determine the filtered history learning topics as the target titles corresponding to the target students.

[0068] Step S104: obtaining a target answer result corresponding to the target question, and performing data evaluation on the target answer result to obtain a target mastery result corresponding to the target student under the target knowledge point.

[0069] Exemplarily, the target answer results of the target students corresponding to the target questions are collected, and then the accuracy of the target answer results is evaluated to determine the target mastery results of the target students under the target knowledge points.

[0070] For example, when the accuracy rate is greater than or equal to the preset value, it is determined that the target mastery result corresponding to the target student under the target knowledge point is the same as or higher than the initial mastery result; when the accuracy rate is less than the preset value, it is determined that the target mastery result corresponding to the target student under the target knowledge point is lower than the initial mastery result.

[0071] Step S105: Perform a learning plan for the target student according to the target mastering result to obtain an initial learning plan corresponding to the target student.

[0072] Exemplarily, the target mastery results are used to clarify the learning needs of the target students on the target knowledge points. For example, the target knowledge points need to be strengthened and improved or the target knowledge points need to be learned step by step from the basics. Therefore, according to the target mastery results of the target students, the corresponding learning goals of the target students are set. For example, the short-term goal may be to master the target knowledge points proficiently, and the long-term goal may be to improve the overall subject ability. Therefore, a model suitable for learning planning is selected, such as a goal-oriented model, a capability matching model, or a personalized learning model. These models should be able to handle the mastery of the target students and the set learning goals. Therefore, according to the selected model, the framework of the learning planning model is constructed. Therefore, the target mastery results and learning goals of the target students are input into the learning planning model, and the learning planning model can be a neural network model or a deep learning model, so that the learning planning model will generate a corresponding initial learning plan based on the target mastery results and learning goals.

[0073] Alternatively, students similar to the target student are found from historical data based on the target mastering results and learning objectives, so that the learning plan of the students similar to the target student is determined as the initial learning plan corresponding to the target student.

[0074] Step S106: Obtain a set of wrong questions corresponding to the target student according to the initial learning plan, and perform data analysis on the set of wrong questions to obtain the learning blind spots corresponding to the target student.

[0075] Exemplarily, the target student is taught according to the initial learning plan and the corresponding wrong question data when the target student takes a question test under the initial learning plan is obtained, so that the wrong question data is combined into a wrong question set corresponding to the target student.

[0076] Exemplarily, the relevant knowledge points corresponding to each wrong question data in the wrong question set are obtained, and then the relevant knowledge points corresponding to the wrong question data in the wrong question set are counted, so that the knowledge points with the highest statistical number of relevant knowledge points are determined as the learning blind spots corresponding to the target students.

[0077] In some embodiments, the performing data analysis on the set of wrong questions to obtain the learning blind spots corresponding to the target students includes: using a text representation model to obtain an initial representation vector corresponding to each wrong question in the set of wrong questions; performing similarity calculation on the wrong questions according to the initial representation vector to obtain a plurality of adjacent questions corresponding to each wrong question; obtaining the distance between the wrong question and the adjacent questions according to the initial representation vector to obtain a distance value; determining the degree of correlation between the wrong question and the adjacent questions according to the distance value to obtain a target correlation value corresponding to the wrong question; performing data clustering on the set of wrong questions according to the target correlation value to obtain a target clustering result corresponding to the set of wrong questions; summarizing the same knowledge points for each third sub-clustering result in the target clustering result to obtain the learning blind spots corresponding to the target student; wherein the target correlation value is obtained according to the following formula:

[0078]

[0079] Among them, corr i represents the target correlation value corresponding to the i-th wrong question, N represents the number of questions corresponding to the adjacent questions, and dis ij represents the distance value between the i-th wrong question and the j-th adjacent question, M represents the number of related questions corresponding to the j-th adjacent question in the wrong question set, and dis mj Represents the distance value between the jth adjacent question and the mth related question.

[0080] Exemplarily, a suitable text representation model is selected according to the requirements. For example, a pre-trained language model (such as BERT, GPT, RoBERTa, etc.) or other specialized text representation models can be selected. Thus, the text corresponding to each wrong question is input into the selected text representation model to generate the corresponding initial representation vector.

[0081] Exemplarily, initial characterization vectors corresponding to any two wrong questions are obtained, and then the similarity of the wrong questions is calculated using cosine similarity based on the initial characterization vectors to obtain the similarity value between each wrong question and other questions in the wrong question set except the wrong question, and then the questions corresponding to the top few similarity values ​​are determined as adjacent questions corresponding to the wrong questions.

[0082] Exemplarily, the distance between the wrong question and each adjacent question is obtained by using cosine similarity according to the initial representation vector to obtain a distance value; thus, the degree of correlation between the wrong question and the adjacent questions is determined according to the distance value to obtain a target correlation value corresponding to the wrong question. The target correlation value is obtained according to the following formula:

[0083]

[0084] Among them, corr i represents the target correlation value corresponding to the i-th wrong question, N represents the number of questions corresponding to the adjacent questions, and ij represents the distance between the i-th wrong question and the j-th adjacent question, M represents the number of related questions corresponding to the j-th adjacent question in the wrong question set, and dis mj Represents the distance value between the jth adjacent question and the mth related question.

[0085] Exemplarily, M represents the number of related questions corresponding to the jth adjacent question in the wrong question set, that is, the number of adjacent questions corresponding to the jth adjacent question in the wrong question set when the jth adjacent question is a wrong question. The distance value between the wrong question and each adjacent question is converted, so as to To represent the dense information of wrong questions in a local range, and then according to To represent the dense information of wrong questions in the range of adjacent questions, the target correlation value corresponding to the wrong questions in the wrong question set is characterized according to the dense information in the local range and the dense information in the range of adjacent questions. That is, the target correlation value is used to characterize the comprehensive density of wrong questions in the wrong question set. In other words, the distribution of wrong questions in the wrong question set is quantified according to the target correlation value, which provides support for subsequent data clustering.

[0086] Exemplarily, the wrong question set is clustered according to the target correlation value, and the wrong questions corresponding to when the difference between the target correlation values ​​is less than or equal to the preset correlation value are determined as a cluster, thereby obtaining the target clustering result corresponding to the wrong question set.

[0087] Exemplarily, the same knowledge points are summarized for each third sub-clustering result in the target clustering result, and then the knowledge points corresponding to each summarized third sub-clustering result are determined as the learning blind spots corresponding to the target students.

[0088] Specifically, the distribution of wrong questions in the wrong question set can be better quantified according to the target correlation value, so that the wrong questions can be clustered more accurately, providing good support for the subsequent discovery of learning blind spots.

[0089] In some embodiments, clustering the wrong question set according to the target relevance value to obtain the target clustering result corresponding to the wrong question set includes: determining any two wrong questions in the wrong question set as the first question and the second question, and obtaining the first associated question corresponding to the first question and the second associated question corresponding to the second question; obtaining the first representation vector corresponding to the first question and the second representation vector corresponding to the second question by using the text representation model; determining the association distance between the first question and the second question according to the calculation of the first representation vector and the second representation vector; performing intersection processing on the first associated question and the second associated question to obtain the target associated question, and obtaining the target number corresponding to the target associated question; obtaining the first relevance value corresponding to the first question and the second relevance value corresponding to the second question from the target relevance value; determining the target similarity between the first question and the second question according to the association distance, the target number, the first relevance value and the second relevance value; clustering the wrong question set according to the target similarity to obtain the target clustering result corresponding to the wrong question set; wherein the target similarity is obtained according to the following formula:

[0090]

[0091] Among them, corr r represents the first correlation value corresponding to the rth first question, corr t represents the second correlation value corresponding to the t-th second question, corr r By substituting the rth first question into corr i In the t By substituting the tth second question into corr i Obtained in, dis rt represents the correlation distance between the rth first question and the tth second question, sum represents the target number, min(corr r ,corr t ) indicates obtaining the minimum value between the first correlation value and the second correlation value, max(corr r ,corr t ) indicates obtaining the maximum value between the first correlation value and the second correlation value.

[0092] Exemplarily, any two wrong questions are determined as the first question and the second question from the wrong question set, and a first related question corresponding to the first question and a second related question corresponding to the second question are obtained, wherein the first related questions are multiple adjacent questions corresponding to the first question, and the second related questions are multiple adjacent questions corresponding to the second question.

[0093] Exemplarily, a first representation vector corresponding to the first topic and a second representation vector corresponding to the second topic are obtained using a text representation model, so as to determine the association distance between the first topic and the second topic by calculating the cosine similarity between the first representation vector and the second representation vector.

[0094] Exemplarily, the first related topic and the second related topic are subjected to intersection processing to obtain a target related topic, that is, the obtained target related topic exists in the first related topic and also in the second related topic, and then data statistics are performed on the target related topic to obtain the corresponding target quantity.

[0095] Exemplarily, the first correlation value corresponding to the first question and the second correlation value corresponding to the second question are obtained according to the above calculation formula of the target correlation value; thereby the target similarity between the first question and the second question is obtained according to the following formula:

[0096]

[0097] Among them, corr r represents the first correlation value corresponding to the first question of the rth order, corr t represents the second correlation value corresponding to the t-th second question, corr r By substituting the rth first question into corr i In the t By substituting the tth second question into corr i Obtained in, dis rt represents the correlation distance between the rth first question and the tth second question, sum represents the number of targets, min(corr r ,corr t ) indicates obtaining the minimum value between the first correlation value and the second correlation value, max(corr r ,corr t ) indicates obtaining the maximum value between the first correlation value and the second correlation value.

[0098] For example, by calculating the minimum and maximum values ​​of the correlation values ​​of the first question and the second question, the relative importance and similarity of the two questions on the same knowledge point can be measured. This helps to accurately evaluate the correlation between questions, rather than relying solely on absolute values. Combining the correlation distance with the minimum and maximum correlation values ​​can more comprehensively capture the true similarity between questions, avoid the deviation of a single dimension, and make the similarity calculation more fair and reasonable. This improves the reliability and practicality of the similarity measurement.

[0099] Exemplarily, according to the target similarity calculation method, the similarity between each pair of wrong questions in the wrong question set is calculated. These similarity values ​​are organized into a matrix, wherein each element of the matrix represents the similarity value between a pair of wrong questions. Thus, the wrong question set is clustered using the target similarity according to K-means clustering or hierarchical clustering to obtain the target clustering result corresponding to the wrong question set.

[0100] Specifically, the similarity between wrong questions can be accurately measured according to the target similarity, thereby improving the effect of subsequent clustering analysis and making the clustering results more in line with the actual situation, thereby ensuring a comprehensive and accurate analysis of the wrong question set, thereby providing strong data support for teaching and learning.

[0101] Step S107: adjusting the initial learning plan according to the learning blind spots to obtain a target learning plan corresponding to the target student.

[0102] Exemplarily, learning blind spots are weak links found by target students in the process of learning target knowledge points. Therefore, the content of the initial learning plan is adjusted according to the identified learning blind spots. This may include increasing the explanation of relevant knowledge points, providing more exercises or supplementing learning resources. According to the students' blind spots, the goals of the learning plan are modified to make them more in line with the students' actual needs. Then all adjusted content, goals and teaching methods are integrated into a new learning plan. This should include a detailed learning plan, specific course arrangements and evaluation methods. In this way, the target learning plan corresponding to the target students is obtained.

[0103] For example, the acquisition of a target learning plan is a dynamic and iterative process, which involves identifying learning blind spots, evaluating and adjusting existing learning plans, developing and implementing new learning plans, and continuous monitoring and optimization. This ensures that the learning plan matches the actual needs of students, effectively addresses their weak links in learning, and thus improves learning outcomes.

[0104] The embodiment of the present invention provides a personalized teaching and training auxiliary method, system, terminal device and medium, the method includes: by analyzing the historical question-solving data and learning process data of the target student, the target student's knowledge mastery and learning habits can be understood, so that the target student's initial mastery result corresponding to the target knowledge point, and then the target student's initial mastery result is recommended according to the target student's initial mastery result, which can effectively improve the pertinence and effectiveness of the exercise, help students to focus on the target knowledge point, and thus improve the overall learning effect. Then obtain the target answer result corresponding to the target question, perform data evaluation on the target answer result to obtain the target mastery result corresponding to the target student under the target knowledge point, and make a learning plan for the target student according to the target mastery result to obtain the initial learning plan corresponding to the target student; obtain the wrong question set corresponding to the target student according to the initial learning plan, and perform data analysis on the wrong question set to obtain the learning blind spot corresponding to the target student; adjust the initial learning plan according to the learning blind spot to obtain the target learning plan corresponding to the target student. Analyzing the wrong questions and identifying the blind spots in learning can help the target students find their own omissions and misunderstandings in learning, and make targeted improvements, so as to adjust the initial learning plan according to the blind spots in learning, ensure that the target learning plan can adapt to the actual learning situation of the target students, continuously optimize the learning strategy, and improve the learning effect. Finally, through personalized and targeted target learning plans, the target students can study more efficiently, save time and energy, and achieve better learning results in a shorter period of time. It also solves the problem that related technologies cannot be personalized according to individual differences of students, and it is difficult to effectively guide and intervene in students' learning behaviors, which seriously affects students' learning efficiency and quality.

[0105] See also Figure 2 , Figure 2A personalized teaching and training auxiliary system 200 is provided in the embodiment of the present application, and the personalized teaching and training auxiliary system 200 includes a data acquisition module 201, a data analysis module 202, a topic recommendation module 203, a data evaluation module 204, a solution recommendation module 205, a wrong question analysis module 206, and a solution adjustment module 207, wherein the data acquisition module 201 is used to collect relevant learning data corresponding to the target student, and the relevant learning data includes historical question-solving data and learning process data; the data analysis module 202 is used to perform data analysis on the historical question-solving data and the learning process data to obtain the initial mastery result corresponding to the target student under the target knowledge point; the topic recommendation module 203 is used to recommend topics to the target student based on the initial mastery result, and obtain the corresponding The target questions include true-or-false questions, single-choice questions and multiple-choice questions; a data evaluation module 204 is used to obtain the target answer results corresponding to the target questions, and perform data evaluation on the target answer results to obtain the target mastery results corresponding to the target students under the target knowledge points; a program recommendation module 205 is used to make a learning plan for the target students according to the target mastery results, and obtain the initial learning plan corresponding to the target students; an error analysis module 206 is used to obtain the set of error questions corresponding to the target students according to the initial learning plan, and perform data analysis on the set of error questions to obtain the learning blind spots corresponding to the target students; a program adjustment module 207 is used to adjust the initial learning plan according to the learning blind spots, and obtain the target learning plan corresponding to the target students.

[0106] In some implementations, the data analysis module 202 performs, during the process of analyzing the historical question-solving data and the learning process data to obtain the initial mastery result corresponding to the target student under the target knowledge point:

[0107] Obtaining the associated knowledge points corresponding to the target knowledge points in the historical test data and the historical scores corresponding to the historical test data;

[0108] Performing data clustering on the historical test data according to the historical scores and the associated knowledge points to obtain data clustering results corresponding to the historical test data;

[0109] Obtaining, from the learning process data, associated process data corresponding to each first sub-clustering result in the data clustering result;

[0110] Obtaining first associated data corresponding to the associated knowledge point in a state where the knowledge is mastered and obtaining second associated data corresponding to the associated knowledge point in a state where the knowledge is not mastered;

[0111] Determine, by using the first associated data and the second associated data, a distinction coefficient corresponding to dividing the associated knowledge point into the knowledge mastered state and the knowledge not mastered state;

[0112] Determine, according to the discrimination coefficient, a discrimination threshold corresponding to dividing the associated knowledge point into the knowledge mastered state and the knowledge not mastered state;

[0113] Calculating the target value corresponding to the associated process data according to the discrimination coefficient, and determining the knowledge mastery result corresponding to the target student under the associated knowledge point according to the discrimination threshold and the target value;

[0114] The initial mastering result corresponding to the target knowledge point is determined according to the knowledge mastering result corresponding to the associated knowledge point.

[0115] In some implementations, the data analysis module 202 performs, during the process of clustering the historical problem solving data according to the historical scores and the associated knowledge points to obtain the data clustering results corresponding to the historical problem solving data:

[0116] Obtaining the importance level of the associated knowledge point to the historical question-solving data, wherein the importance level is used to represent the degree of influence of the associated knowledge point on answering the historical question-solving data;

[0117] Performing initial clustering on the historical question-solving data according to the importance level and the associated knowledge points to obtain initial clustering results corresponding to the historical question-solving data;

[0118] Each second sub-clustering result in the initial clustering result is segmented according to the historical score to obtain the data clustering result corresponding to the historical question-solving data.

[0119] In some implementations, during the process of determining the distinction threshold corresponding to the associated knowledge point for dividing the knowledge mastered state and the knowledge not mastered state according to the distinction coefficient, the data analysis module 202 performs:

[0120] Calculate a first characterization value corresponding to each first associated data according to the discrimination coefficient, and calculate a second characterization value corresponding to each second associated data according to the discrimination coefficient;

[0121] A distinction threshold corresponding to the associated knowledge point is determined according to the first characterization value and the second characterization value.

[0122] In some implementations, the topic recommendation module 203 performs, during the process of recommending topics to the target student based on the initial mastering result and obtaining the target topic corresponding to the target student:

[0123] Determining the topic level corresponding to the target student under the target knowledge point according to the initial mastery result;

[0124] Determine a history student and the history learning data corresponding to the history student, and determine the related students corresponding to the target student according to the related learning data and the history learning data;

[0125] Obtaining the history learning topics corresponding to the target knowledge points and the grade information corresponding to the history learning topics of the relevant students;

[0126] The history learning topics are screened according to the topic levels and the level information to obtain the target topics corresponding to the target students.

[0127] In some implementations, the wrong question analysis module 206 performs, during the process of performing data analysis on the wrong question set to obtain the learning blind spots corresponding to the target student:

[0128] Using a text representation model to obtain an initial representation vector corresponding to each wrong question in the wrong question set;

[0129] Calculate the similarity of the wrong questions according to the initial representation vector to obtain a plurality of adjacent questions corresponding to each wrong question;

[0130] Obtaining the distance between the wrong question and the adjacent question according to the initial representation vector to obtain a distance value;

[0131] Determine the correlation between the wrong question and the adjacent question according to the distance value, and obtain a target correlation value corresponding to the wrong question;

[0132] Performing data clustering on the wrong question set according to the target correlation value to obtain a target clustering result corresponding to the wrong question set;

[0133] Summarizing the same knowledge points of each third sub-clustering result in the target clustering result to obtain the learning blind spots corresponding to the target student;

[0134] Wherein, the target related value is obtained according to the following formula:

[0135]

[0136] Among them, corr i represents the target correlation value corresponding to the i-th wrong question, N represents the number of questions corresponding to the adjacent questions, and dis ijrepresents the distance value between the i-th wrong question and the j-th adjacent question, M represents the number of related questions corresponding to the j-th adjacent question in the wrong question set, and dis mj Represents the distance value between the jth adjacent question and the mth related question.

[0137] In some implementations, the wrong question analysis module 206 performs, during the process of clustering the wrong question set according to the target related value to obtain the target clustering result corresponding to the wrong question set:

[0138] Determine any two wrong questions in the wrong question set as a first question and a second question, and obtain a first related question corresponding to the first question and a second related question corresponding to the second question;

[0139] Obtaining a first representation vector corresponding to the first question and a second representation vector corresponding to the second question using the text representation model;

[0140] Determine the association distance between the first topic and the second topic by calculating the first characterization vector and the second characterization vector;

[0141] Performing intersection processing on the first related topic and the second related topic to obtain a target related topic, and obtaining a target quantity corresponding to the target related topic;

[0142] Obtaining a first correlation value corresponding to the first topic and a second correlation value corresponding to the second topic from the target correlation value;

[0143] Determine the target similarity between the first topic and the second topic according to the association distance, the target number, and the first correlation value and the second correlation value;

[0144] Performing data clustering on the wrong question set according to the target similarity to obtain a target clustering result corresponding to the wrong question set;

[0145] The target similarity is obtained according to the following formula:

[0146]

[0147] Among them, corr r represents the first correlation value corresponding to the rth first question, corr t represents the second correlation value corresponding to the t-th second question, corr r By substituting the rth first question into corr i In the tBy substituting the tth second question into corr i Obtained in, dis rt represents the correlation distance between the rth first question and the tth second question, sum represents the target number, min(corr r ,corr t ) indicates obtaining the minimum value between the first correlation value and the second correlation value, max(corr r ,corr t ) indicates obtaining the maximum value between the first correlation value and the second correlation value.

[0148] In some embodiments, the personalized teaching and training assistance system 200 can be applied to a terminal device.

[0149] It should be noted that technical personnel in the relevant field can clearly understand that for the convenience and conciseness of description, the specific working process of the personalized teaching and training assistance system 200 described above can refer to the corresponding process in the aforementioned personalized teaching and training assistance method embodiment, and will not be repeated here.

[0150] See also Figure 3 , Figure 3 A schematic block diagram of the structure of a terminal device provided in an embodiment of the present invention.

[0151] like Figure 3 As shown, the terminal device 300 includes a processor 301 and a memory 302, and the processor 301 and the memory 302 are connected via a bus 303, such as an I2C (Inter-integrated Circuit) bus.

[0152] Specifically, the processor 301 is used to provide computing and control capabilities to support the operation of the entire terminal device. The processor 301 can be a central processing unit (CPU), and the processor 301 can also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc.

[0153] Specifically, the memory 302 may be a Flash chip, a read-only memory (ROM) disk, an optical disk, a USB flash drive, or a mobile hard disk.

[0154] Those skilled in the art will understand that Figure 3 The structure shown in the figure is only a block diagram of a partial structure related to the embodiment of the present invention, and does not constitute a limitation on the terminal device to which the embodiment of the present invention is applied. The specific server may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0155] Among them, the processor is used to run the computer program stored in the memory, and implement any one of the personalized teaching and training assistance methods provided by the embodiments of the present invention when executing the computer program.

[0156] In one embodiment, the processor is used to run a computer program stored in the memory, and implements the following steps when executing the computer program:

[0157] Collect relevant learning data corresponding to the target students, wherein the relevant learning data includes historical question-solving data and learning process data;

[0158] Performing data analysis on the historical question-solving data and the learning process data to obtain the initial mastery results corresponding to the target students under the target knowledge points;

[0159] Recommending questions to the target students according to the initial mastery results to obtain target questions corresponding to the target students, wherein the target questions include true-or-false questions, single-choice questions and multiple-choice questions;

[0160] Obtaining a target answer result corresponding to the target question, and performing data evaluation on the target answer result to obtain a target mastery result corresponding to the target student under the target knowledge point;

[0161] Performing a learning plan for the target student according to the target mastering result to obtain an initial learning plan corresponding to the target student;

[0162] Obtaining a set of wrong questions corresponding to the target student according to the initial learning plan, and performing data analysis on the set of wrong questions to obtain a learning blind spot corresponding to the target student;

[0163] The initial learning plan is adjusted according to the learning blind spots to obtain a target learning plan corresponding to the target student.

[0164] In some implementations, the processor 301 performs, during the process of performing data analysis on the historical question-solving data and the learning process data to obtain the initial mastery result corresponding to the target student under the target knowledge point:

[0165] Obtaining the associated knowledge points corresponding to the target knowledge points in the historical test data and the historical scores corresponding to the historical test data;

[0166] Performing data clustering on the historical test data according to the historical scores and the associated knowledge points to obtain data clustering results corresponding to the historical test data;

[0167] Obtaining, from the learning process data, associated process data corresponding to each first sub-clustering result in the data clustering result;

[0168] Obtaining first associated data corresponding to the associated knowledge point in a state where the knowledge is mastered and obtaining second associated data corresponding to the associated knowledge point in a state where the knowledge is not mastered;

[0169] Determine, by using the first associated data and the second associated data, a distinction coefficient corresponding to dividing the associated knowledge point into the knowledge mastered state and the knowledge not mastered state;

[0170] Determine, according to the discrimination coefficient, a discrimination threshold corresponding to dividing the associated knowledge point into the knowledge mastered state and the knowledge not mastered state;

[0171] Calculating the target value corresponding to the associated process data according to the discrimination coefficient, and determining the knowledge mastery result corresponding to the target student under the associated knowledge point according to the discrimination threshold and the target value;

[0172] The initial mastering result corresponding to the target knowledge point is determined according to the knowledge mastering result corresponding to the associated knowledge point.

[0173] In some implementations, the processor 301, in the process of clustering the historical problem-solving data according to the historical scores and the associated knowledge points to obtain the data clustering results corresponding to the historical problem-solving data, executes:

[0174] Obtaining the importance level of the associated knowledge point to the historical question-solving data, wherein the importance level is used to represent the degree of influence of the associated knowledge point on answering the historical question-solving data;

[0175] Performing initial clustering on the historical question-solving data according to the importance level and the associated knowledge points to obtain initial clustering results corresponding to the historical question-solving data;

[0176] Each second sub-clustering result in the initial clustering result is segmented according to the historical score to obtain the data clustering result corresponding to the historical question-solving data.

[0177] In some implementations, during the process of determining the distinction threshold corresponding to dividing the associated knowledge point into the knowledge mastered state and the knowledge not mastered state according to the distinction coefficient, the processor 301 executes:

[0178] Calculate a first characterization value corresponding to each first associated data according to the discrimination coefficient, and calculate a second characterization value corresponding to each second associated data according to the discrimination coefficient;

[0179] A distinction threshold corresponding to the associated knowledge point is determined according to the first characterization value and the second characterization value.

[0180] In some implementations, the processor 301, in the process of recommending a topic to the target student according to the initial mastering result and obtaining a target topic corresponding to the target student, executes:

[0181] Determining the topic level corresponding to the target student under the target knowledge point according to the initial mastery result;

[0182] Determine a history student and the history learning data corresponding to the history student, and determine the related students corresponding to the target student according to the related learning data and the history learning data;

[0183] Obtaining the history learning topics corresponding to the target knowledge points and the grade information corresponding to the history learning topics of the relevant students;

[0184] The history learning topics are screened according to the topic levels and the level information to obtain the target topics corresponding to the target students.

[0185] In some implementations, the processor 301, in the process of performing data analysis on the set of wrong questions to obtain the learning blind spots corresponding to the target student, executes:

[0186] Using a text representation model to obtain an initial representation vector corresponding to each wrong question in the wrong question set;

[0187] Calculate the similarity of the wrong questions according to the initial representation vector to obtain a plurality of adjacent questions corresponding to each wrong question;

[0188] Obtaining the distance between the wrong question and the adjacent question according to the initial representation vector to obtain a distance value;

[0189] Determine the correlation between the wrong question and the adjacent question according to the distance value, and obtain a target correlation value corresponding to the wrong question;

[0190] Performing data clustering on the wrong question set according to the target correlation value to obtain a target clustering result corresponding to the wrong question set;

[0191] Summarizing the same knowledge points of each third sub-clustering result in the target clustering result to obtain the learning blind spots corresponding to the target student;

[0192] Wherein, the target related value is obtained according to the following formula:

[0193]

[0194] Among them, corr i represents the target correlation value corresponding to the i-th wrong question, N represents the number of questions corresponding to the adjacent questions, and dis ij represents the distance value between the i-th wrong question and the j-th adjacent question, M represents the number of related questions corresponding to the j-th adjacent question in the wrong question set, and dis mj Represents the distance value between the jth adjacent question and the mth related question.

[0195] In some implementations, the processor 301, in the process of clustering the wrong question set according to the target correlation value to obtain the target clustering result corresponding to the wrong question set, executes:

[0196] Determine any two wrong questions in the wrong question set as a first question and a second question, and obtain a first related question corresponding to the first question and a second related question corresponding to the second question;

[0197] Obtaining a first representation vector corresponding to the first question and a second representation vector corresponding to the second question using the text representation model;

[0198] Determine the association distance between the first topic and the second topic by calculating the first characterization vector and the second characterization vector;

[0199] Performing intersection processing on the first related topic and the second related topic to obtain a target related topic, and obtaining a target quantity corresponding to the target related topic;

[0200] Obtaining a first correlation value corresponding to the first topic and a second correlation value corresponding to the second topic from the target correlation value;

[0201] Determine the target similarity between the first topic and the second topic according to the association distance, the target number, and the first correlation value and the second correlation value;

[0202] Performing data clustering on the wrong question set according to the target similarity to obtain a target clustering result corresponding to the wrong question set;

[0203] The target similarity is obtained according to the following formula:

[0204]

[0205] Among them, corr r represents the first correlation value corresponding to the rth first question, corr t represents the second correlation value corresponding to the t-th second question, corr r By substituting the rth first question into corr i In the t By substituting the tth second question into corr i Obtained in, dis rt represents the correlation distance between the rth first question and the tth second question, sum represents the target number, min(corr r ,corr t ) indicates obtaining the minimum value between the first correlation value and the second correlation value, max(corr r ,corr t ) indicates obtaining the maximum value between the first correlation value and the second correlation value.

[0206] It should be noted that technical personnel in the relevant field can clearly understand that for the convenience and conciseness of description, the specific working process of the terminal device described above can refer to the corresponding process in the aforementioned personalized teaching and training assistance method embodiment, and will not be repeated here.

[0207] An embodiment of the present invention also provides a storage medium for computer-readable storage, wherein the storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps of any personalized teaching and training assistance method provided in the description of the embodiment of the present invention.

[0208] The storage medium may be an internal storage unit of the terminal device in the aforementioned embodiment, such as a hard disk or memory of the terminal device. The storage medium may also be an external storage device of the terminal device, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the terminal device.

[0209] It will be appreciated by those skilled in the art that all or some of the steps, systems, and functional modules / units in the methods disclosed above may be implemented as software, firmware, hardware, and appropriate combinations thereof. In a hardware embodiment, the division between the functional modules / units mentioned in the above description does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed by several physical components in cooperation. Some or all physical components may be implemented as software executed by a processor, such as a central processing unit, a digital signal processor, or a microprocessor, or implemented as hardware, or implemented as an integrated circuit, such as an application-specific integrated circuit. Such software may be distributed on a computer-readable medium, which may include a computer storage medium (or non-transitory medium) and a communication medium (or temporary medium). As known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer. In addition, it is well known to those of ordinary skill in the art that communication media typically contain computer-readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

[0210] It should be understood that the term "and / or" used in the present specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, including these combinations. It should be noted that, in this article, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "including a..." does not exclude the presence of other identical elements in the process, method, article or system including the element.

[0211] The serial numbers of the embodiments of the present invention are only for description and do not represent the advantages and disadvantages of the embodiments. The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed by the present invention, and these modifications or replacements should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A personalized teaching and training auxiliary method, characterized in that: The method comprises: Collect relevant learning data corresponding to the target students, wherein the relevant learning data includes historical question-solving data and learning process data; Performing data analysis on the historical question-solving data and the learning process data to obtain the initial mastery results corresponding to the target students under the target knowledge points; Recommending questions to the target students according to the initial mastery results to obtain target questions corresponding to the target students, wherein the target questions include true-or-false questions, single-choice questions and multiple-choice questions; Obtaining a target answer result corresponding to the target question, and performing data evaluation on the target answer result to obtain a target mastery result corresponding to the target student under the target knowledge point; Performing a learning plan for the target student according to the target mastering result to obtain an initial learning plan corresponding to the target student; Obtaining a set of wrong questions corresponding to the target student according to the initial learning plan, and performing data analysis on the set of wrong questions to obtain a learning blind spot corresponding to the target student; The initial learning plan is adjusted according to the learning blind spots to obtain a target learning plan corresponding to the target student.

2. The method according to claim 1, characterized in that The performing of data analysis on the history test data and the learning process data to obtain the initial mastery result corresponding to the target student under the target knowledge point includes: Obtaining the associated knowledge points corresponding to the target knowledge points in the historical test data and the historical scores corresponding to the historical test data; Performing data clustering on the historical test data according to the historical scores and the associated knowledge points to obtain data clustering results corresponding to the historical test data; Obtaining, from the learning process data, associated process data corresponding to each first sub-clustering result in the data clustering result; Obtaining first associated data corresponding to the associated knowledge point in a state where the knowledge is mastered and obtaining second associated data corresponding to the associated knowledge point in a state where the knowledge is not mastered; Determine, by using the first associated data and the second associated data, a distinction coefficient corresponding to dividing the associated knowledge point into the knowledge mastered state and the knowledge not mastered state; Determine, according to the discrimination coefficient, a discrimination threshold corresponding to dividing the associated knowledge point into the knowledge mastered state and the knowledge not mastered state; Calculating the target value corresponding to the associated process data according to the discrimination coefficient, and determining the knowledge mastery result corresponding to the target student under the associated knowledge point according to the discrimination threshold and the target value; The initial mastering result corresponding to the target knowledge point is determined according to the knowledge mastering result corresponding to the associated knowledge point.

3. The method according to claim 2, characterized in that The performing data clustering on the historical test data according to the historical scores and the associated knowledge points to obtain data clustering results corresponding to the historical test data includes: Obtaining the importance level of the associated knowledge point to the historical question-solving data, wherein the importance level is used to represent the degree of influence of the associated knowledge point on answering the historical question-solving data; Performing initial clustering on the historical question-solving data according to the importance level and the associated knowledge points to obtain initial clustering results corresponding to the historical question-solving data; Each second sub-clustering result in the initial clustering result is segmented according to the historical score to obtain the data clustering result corresponding to the historical question-solving data.

4. The method according to claim 2, characterized in that: The determining, according to the distinction coefficient, a distinction threshold corresponding to dividing the associated knowledge point into the knowledge mastered state and the knowledge not mastered state comprises: Calculate a first characterization value corresponding to each first associated data according to the discrimination coefficient, and calculate a second characterization value corresponding to each second associated data according to the discrimination coefficient; A distinction threshold corresponding to the associated knowledge point is determined according to the first characterization value and the second characterization value.

5. The method according to claim 1, characterized in that The step of recommending a topic to the target student according to the initial mastering result to obtain a target topic corresponding to the target student includes: Determining the topic level corresponding to the target student under the target knowledge point according to the initial mastery result; Determine a history student and the history learning data corresponding to the history student, and determine the related students corresponding to the target student according to the related learning data and the history learning data; Obtaining the history learning topics corresponding to the target knowledge points and the grade information corresponding to the history learning topics of the relevant students; The history learning topics are screened according to the topic levels and the level information to obtain the target topics corresponding to the target students.

6. The method according to claim 1, characterized in that The performing data analysis on the set of wrong questions to obtain the learning blind spots corresponding to the target students includes: Using a text representation model to obtain an initial representation vector corresponding to each wrong question in the wrong question set; Calculate the similarity of the wrong questions according to the initial representation vector to obtain a plurality of adjacent questions corresponding to each wrong question; Obtaining the distance between the wrong question and the adjacent question according to the initial representation vector to obtain a distance value; Determine the correlation between the wrong question and the adjacent question according to the distance value, and obtain a target correlation value corresponding to the wrong question; Performing data clustering on the wrong question set according to the target correlation value to obtain a target clustering result corresponding to the wrong question set; Summarizing the same knowledge points of each third sub-clustering result in the target clustering result to obtain the learning blind spots corresponding to the target student; Wherein, the target related value is obtained according to the following formula: Among them, corr i represents the target correlation value corresponding to the i-th wrong question, N represents the number of questions corresponding to the adjacent questions, and dis ij represents the distance value between the i-th wrong question and the j-th adjacent question, M represents the number of related questions corresponding to the j-th adjacent question in the wrong question set, and dis mj Represents the distance value between the jth adjacent question and the mth related question.

7. The method according to claim 6, characterized in that The step of clustering the wrong question set according to the target related value to obtain a target clustering result corresponding to the wrong question set includes: Determine any two wrong questions in the wrong question set as a first question and a second question, and obtain a first related question corresponding to the first question and a second related question corresponding to the second question; Obtaining a first representation vector corresponding to the first question and a second representation vector corresponding to the second question using the text representation model; Determine the association distance between the first topic and the second topic by calculating the first characterization vector and the second characterization vector; Performing intersection processing on the first related topic and the second related topic to obtain a target related topic, and obtaining a target quantity corresponding to the target related topic; Obtaining a first correlation value corresponding to the first topic and a second correlation value corresponding to the second topic from the target correlation value; Determine the target similarity between the first topic and the second topic according to the association distance, the target number, and the first correlation value and the second correlation value; Performing data clustering on the wrong question set according to the target similarity to obtain a target clustering result corresponding to the wrong question set; The target similarity is obtained according to the following formula: Among them, corr r represents the first correlation value corresponding to the rth first question, corr t represents the second correlation value corresponding to the t-th second question, corr r By substituting the rth first question into corr i In the t By substituting the tth second question into corr i Obtained in, dis rt represents the correlation distance between the rth first question and the tth second question, sum represents the target number, min(corr r ,corr t ) indicates obtaining the minimum value between the first correlation value and the second correlation value, max(corr r ,corr t ) indicates obtaining the maximum value between the first correlation value and the second correlation value.

8. A personalized teaching and training auxiliary system, characterized in that: include: A data acquisition module is used to collect relevant learning data corresponding to the target students, wherein the relevant learning data includes historical question-solving data and learning process data; A data analysis module is used to analyze the historical question-solving data and the learning process data to obtain the initial mastery results of the target students under the target knowledge points; A topic recommendation module is used to recommend topics to the target students according to the initial mastery results, and obtain target topics corresponding to the target students, wherein the target topics include true-or-false questions, single-choice questions and multiple-choice questions; A data evaluation module is used to obtain a target answer result corresponding to the target question, and to perform data evaluation on the target answer result to obtain a target mastery result corresponding to the target student under the target knowledge point; A program recommendation module is used to make a learning plan for the target student according to the target mastering result, and obtain an initial learning program corresponding to the target student; A wrong question analysis module, used to obtain a wrong question set corresponding to the target student according to the initial learning plan, and perform data analysis on the wrong question set to obtain a learning blind spot corresponding to the target student; The program adjustment module is used to adjust the initial learning program according to the learning blind spots to obtain the target learning program corresponding to the target students.

9. A terminal device, characterized in that: The terminal device includes a processor and a memory; The memory is used to store computer programs; The processor is used to execute the computer program and implement the personalized teaching and training assistance method as described in any one of claims 1 to 7 when executing the computer program.

10. A computer storage medium for computer storage, characterized in that: The computer storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps of the personalized teaching and training assistance method described in any one of claims 1 to 7.

Images