Intelligent learning situation analysis and interactive teaching optimization method and system and medium thereof

Abstract

This application discloses a method, system, and medium based on intelligent learning analysis and interactive teaching optimization, relating to the field of educational informatization technology. The method includes: recognizing and parsing images of teaching materials to generate structured homework data containing knowledge point tags; collecting student classroom behavior data, calculating focus deviation values ​​based on knowledge point tags, and triggering adjustments to teaching strategies; constructing a student ability assessment matrix, establishing a dynamic task grouping model based on the student ability assessment matrix and task attribute characteristics, and generating a task-learning progress report; collecting teacher physiological data, combining course information with the complexity index of the dynamic task grouping model to construct a teacher fatigue assessment model; optimizing the scheduling scheme by calling a scheduling optimization algorithm based on the fatigue assessment results; acquiring student learning trajectory data, using machine learning to generate personalized learning plans, and dynamically adjusting the weights of learning parameters through a feedback mechanism. This application improves the quality of interactive teaching and student learning efficiency.

Description

Technical Field

[0001] This application relates to the field of educational informatization technology, and in particular to a method, system and medium based on intelligent learning analysis and interactive teaching optimization. Background Technology

[0002] Currently, with the deepening development of educational informatization, intelligent teaching platforms have shown great potential in improving teaching efficiency and personalized instruction. However, existing technologies still have shortcomings in accurately generating student ability profiles and providing scientific feedback mechanisms.

[0003] Traditional teaching platforms rely on manual grading of assignments and manual analysis of student learning data. This is not only time-consuming and labor-intensive, but also makes it difficult to achieve real-time tracking and detailed analysis of students' learning status. As a result, the quality of interactive teaching is poor and it is not conducive to improving students' learning efficiency. Summary of the Invention

[0004] To improve the quality of interactive teaching and enhance students' learning efficiency, this application provides a method, system, and medium based on intelligent learning analysis and interactive teaching optimization.

[0005] Firstly, the objective of this invention is achieved through the following technical solution:

[0006] Based on intelligent learning analysis and interactive teaching optimization methods, including:

[0007] Text recognition and question parsing are performed on images of teaching materials to generate structured homework data containing question types and knowledge point tags;

[0008] Collect student classroom behavior data, calculate focus deviation value based on the knowledge point tags and attention assessment model, and trigger adjustments to teaching strategies.

[0009] Based on the error rate and knowledge mastery of students' historical homework data, and combined with the focus deviation value, a student ability assessment matrix is ​​constructed. Based on the student ability assessment matrix and task attribute characteristics, a dynamic task grouping model is established to generate a task-learning progress report.

[0010] Physiological data of teachers is collected by biosensors, and a teacher fatigue assessment model is constructed by combining course information with the complexity index of the dynamic task grouping model. Based on the fatigue assessment results, a scheduling optimization algorithm is called to optimize the scheduling scheme.

[0011] The system acquires student learning trajectory data, uses machine learning to generate personalized learning plans, and dynamically adjusts the weights of learning parameters based on the task-learning progress report through a feedback mechanism.

[0012] By adopting the above technical solutions, this application can be applied to industry-education integration scenarios such as vocational education and training centers and higher education institutions; it automatically parses teaching materials to improve homework grading efficiency, while the structured homework data provides standardized input for subsequent learning analysis; to optimize classroom interaction quality in real time, it dynamically adjusts teaching strategies through an attention assessment model, which helps improve students' focus in class; the dynamic task grouping model helps improve task matching, enabling refined allocation of learning resources; through teacher physiological data and fatigue assessment models, it effectively monitors teachers' physical condition, prevents a decline in teaching quality due to fatigue, and dynamically optimizes scheduling schemes, improving the overall stability of the teaching system; through personalized learning plans, it promotes personalized learning plans, utilizes learning algorithms to uncover students' learning patterns, and formulates learning paths that conform to individual characteristics. The personalized learning plans of this application support dynamic adjustment of parameters such as learning content and training intensity to adapt to the learning needs of different stages and improve students' learning efficiency. Thus, this application provides a teaching optimization solution with a higher degree of automation, more timely teaching response, stronger personalization, and more reasonable allocation of educational resources.

[0013] In a preferred embodiment, the structured homework data includes first homework data and second homework data; after performing text recognition and question parsing on the teaching aid images, the method further includes:

[0014] Based on the first assignment data and the second assignment data, learning analysis factors related to the mastery of knowledge points are obtained, wherein the learning analysis factors include a first learning analysis factor and a second learning analysis factor;

[0015] Based on the learning situation analysis factors, learning status influencing factors representing students' learning status are obtained, wherein the learning status influencing factors include a first learning status influencing factor and a second learning status influencing factor.

[0016] Based on the learning status influencing factors, students are divided into two groups: the first student group and the second student group.

[0017] Personalized teaching parameters are obtained based on the first student group and the second student group.

[0018] By adopting the above technical solution, the first assignment data is daily practice, and the second assignment data is exam assessment, distinguishing between the two types of assignment data to avoid data interference from different scenarios; by extracting learning analysis factors, capturing hard indicators such as knowledge point mastery and error rate, and soft indicators such as answering speed, a comprehensive assessment of students' true level is conducted; by using learning status influencing factors, abstract learning behaviors are transformed into quantifiable indicators (such as focus deviation value) to facilitate tracking students' classroom learning status, and automatically grouping students into classes and groups based on differences in learning status, avoiding one-size-fits-all teaching, allowing students of similar levels to learn from each other, and generating personalized teaching parameters such as personalized learning pace suggestions (such as frequency of redoing wrong questions) and resource recommendation weights for each student.

[0019] In a preferred embodiment of this application, the process of acquiring the first job data and the second job data includes:

[0020] Based on the images in the teaching materials, obtain multiple-choice and subjective-choice homework data;

[0021] Based on the multiple-choice and subjective question data, obtain the standard answer matching results;

[0022] Based on the standard answer matching results and the question type, obtain the first assignment data and the second assignment data.

[0023] By adopting the above technical solutions, different scoring standards are used for objective questions (multiple choice) and subjective questions (fill in the blank / essay), thereby improving the accuracy of grading; by matching standard answers and analyzing question types, the accuracy of answer comparison is ensured, such as math questions being accurate to four decimal places.

[0024] In a preferred embodiment of this application, the step of obtaining the learning status influencing factors based on the learning situation analysis factors specifically includes:

[0025] Obtain preset knowledge-related factors, and based on the preset knowledge-related factors, filter the first learning situation analysis factors to obtain the first learning state influencing factors, wherein the first learning state influencing factors include the first knowledge point mastery, the first error rate, and the first answering speed;

[0026] Based on the factors influencing the first learning state, obtain the first learning state influencing factor;

[0027] Based on the preset knowledge-related factors, the second learning situation analysis factors are screened to obtain the second learning state influencing factors, wherein the second learning state influencing factors include the second knowledge point mastery, the second error rate, the second answering speed, and the historical error repetition rate;

[0028] Based on the factors influencing the second learning state, obtain the influencing factor of the second learning state.

[0029] By adopting the above technical solutions, key learning factors are screened, irrelevant data such as general font size is filtered out, and core indicators reflecting learning ability are retained. By distinguishing between two types of learning status factors, such as the first type (basic version) which focuses on real-time performance (error rate) and the second type (advanced version) which incorporates historical data (repetition rate of wrong questions), the system can adapt to the needs of different teaching stages.

[0030] In a preferred embodiment of this application: the attention evaluation model is a multimodal attention evaluation model; the method further includes:

[0031] Based on the body movement characteristics and eye focus trajectory in the classroom behavior data, a multimodal attention assessment model is constructed; continuous classroom segments are segmented and labeled based on the sliding window algorithm, and the student attention fluctuation index is calculated in each time period;

[0032] Based on the covariance analysis of the student attention fluctuation index and the corresponding knowledge point difficulty coefficient matrix of the teaching materials, a set of teaching strategy adjustment instructions is dynamically generated.

[0033] By employing the aforementioned technical solutions, the multimodal attention assessment model monitors students' body movements (such as nodding / turning their heads) and gaze focus, more accurately determining whether students are distracted than a single camera. Through segmented annotation via a sliding window (e.g., dividing a 45-minute class into 3-minute segments), it detects real-time fluctuations in attention (e.g., the whole class becoming distracted when explaining difficult problems). Teachers can then make targeted adjustments: slowing down the pace of challenging content or introducing interactive elements. Covariance analysis generates strategies to analyze the difficulty of knowledge points and students' attention fluctuation indices, allowing for dynamic adjustments to knowledge explanation strategies. For example, 3D models can be used to assist in understanding complex math problems.

[0034] In a preferred example, the process of constructing the student ability assessment matrix includes:

[0035] The decay coefficients of the first error rate and the second error rate are calculated using a time series analysis model, and the weights of knowledge point mastery are adjusted in conjunction with the forgetting curve theory.

[0036] The first focus deviation value is graded and quantified by using fuzzy logic algorithm, and a three-dimensional evaluation matrix including cognitive input, knowledge absorption rate and error correction efficiency is established.

[0037] The three-dimensional evaluation matrix is ​​dynamically grouped based on the k-means++ clustering algorithm to generate student clusters with similar learning characteristics.

[0038] Based on the similarity of the learning trajectories of the student clusters, the first learning state influence factor and the second learning state influence factor are correlated.

[0039] By adopting the above technical solutions, the approximate error rate of students is analyzed based on time series data to accurately identify students' knowledge gaps; attention is quantified through fuzzy logic to avoid subjective judgment, so teachers do not have to worry about whether "a student spacing out for 10 seconds counts as being distracted"; a three-dimensional evaluation matrix + K-means++ clustering generates student clusters: accurately classifying student groups with different learning characteristics, laying the foundation for personalized teaching.

[0040] In a preferred embodiment of this application, the method for establishing the dynamic task grouping model includes:

[0041] Based on the cluster center coordinates of the student clusters, the cosine similarity matrix between the task attribute features and the student feature vectors is calculated; the improved k-medoids algorithm is used to optimize the clustering of the cosine similarity matrix to generate a task grouping scheme containing multiple grouping schemes.

[0042] A grouping stability prediction model is established using reinforcement learning algorithms, and the confidence probability distribution of each grouping scheme is output.

[0043] Based on the confidence probability distribution and the class hour allocation constraints in the course information, an adaptive teaching resource push strategy is generated through a multi-objective decision tree.

[0044] By adopting the above technical solutions, the cosine similarity matrix is ​​used to calculate the matching degree between tasks and students: measuring the degree of fit between each student and the task; the improved k-medoids algorithm optimizes grouping, making the grouping results more meaningful for practical teaching; the grouping stability prediction model is used to evaluate the feasibility of different grouping schemes in advance; and the multi-objective decision tree generates a resource push strategy to intelligently recommend the optimal teaching resource allocation method under constraints such as time, number of students, and task complexity.

[0045] Secondly, the objective of this invention is achieved through the following technical solution:

[0046] A system based on intelligent learning analysis and interactive teaching optimization, applied to the aforementioned method based on intelligent learning analysis and interactive teaching optimization, the system comprising:

[0047] The teaching materials digitization module is used to perform text recognition and question analysis on teaching materials images, and generate structured homework data containing question types and knowledge point tags;

[0048] The classroom behavior collection and analysis module is used to collect student classroom behavior data, calculate the focus deviation value based on the knowledge point tags and the attention assessment model, and trigger adjustments to teaching strategies.

[0049] The student ability assessment and task grouping module is used to construct a student ability assessment matrix based on the error rate and knowledge mastery in students' historical homework data, combined with the focus deviation value, and to establish a dynamic task grouping model based on the student ability assessment matrix and task attribute characteristics, and generate a task-learning progress report.

[0050] The scheduling optimization module is used to collect teachers' physiological data through biosensors, combine course information with the complexity index of the dynamic task grouping model, and construct a teacher fatigue assessment model; based on the fatigue assessment results, the scheduling optimization algorithm is called to optimize the scheduling scheme.

[0051] The personalized learning plan generation module is used to acquire students' learning trajectory data, generate personalized learning plans using machine learning, and dynamically adjust the weights of learning parameters based on the task-learning progress report through a feedback mechanism.

[0052] Thirdly, the objective of this invention is achieved through the following technical solution:

[0053] A computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps described above based on the intelligent learning analysis and interactive teaching optimization method.

[0054] Fourthly, the objective of this invention is achieved through the following technical solution:

[0055] A computer program product includes a computer program / instructions that, when executed by a processor, implement the steps described above based on the intelligent learning analysis and interactive teaching optimization method.

[0056] In summary, this application includes at least one of the following beneficial technical effects:

[0057] 1. Dynamically optimize the course scheduling scheme to improve the stability of the overall teaching system; promote personalized learning plans, utilize learning and mining to discover students' learning patterns, and formulate learning paths that meet individual characteristics. The personalized learning plan of this application supports dynamic adjustment of parameters such as learning content and training intensity to adapt to the learning needs of different stages and improve students' learning efficiency.

[0058] 2. By extracting factors for learning analysis, we can capture hard indicators such as knowledge mastery and error rate, as well as soft indicators such as answering speed, to comprehensively assess students' true level. By using learning status influencing factors, we can transform abstract learning behaviors into quantifiable indicators (such as focus deviation value) to track students' classroom learning status and automatically group students according to differences in learning status, avoiding a one-size-fits-all teaching approach. Attached Figure Description

[0059] Figure 1This is a flowchart of an embodiment of the intelligent learning analysis and interactive teaching optimization method in this application;

[0060] Figure 2 This is a flowchart following step S1 in an embodiment of the intelligent learning analysis and interactive teaching optimization method of this application. Detailed Implementation

[0061] The present application will be further described in detail below with reference to the accompanying drawings.

[0062] In one embodiment, such as Figure 1 As shown, this application discloses a method for optimizing teaching based on intelligent learning analysis and interactive teaching, which specifically includes the following steps:

[0063] S1: Perform text recognition and question parsing on images of teaching materials to generate structured homework data containing question types and knowledge point tags.

[0064] In this embodiment, images of teaching materials are acquired to obtain teaching material images; text recognition uses OCR technology to identify text, formulas, charts, and other data in the teaching material images and convert them into structured homework data in JSON format; question analysis refers to using a knowledge graph (such as the GeoGebra math knowledge base) to analyze the knowledge points tested in the questions; the structured homework data includes fields such as question ID, question type (multiple choice / fill in the blank / calculation / subjective / objective), knowledge point tag (People's Education Press Mathematics - Functions), and difficulty level (levels 1-5).

[0065] S2: Collect student classroom behavior data, calculate focus deviation value based on knowledge point tags and attention assessment model, and trigger adjustment of teaching strategies.

[0066] In this embodiment, the attention assessment model is an LSTM neural network based on multimodal data (eye movement + posture); the attention deviation value refers to the difference between the actual attention duration and the theoretical attention duration of the course (formula: deviation value = theoretical duration × 0.7 - actual attention duration).

[0067] Specifically, eye-tracking heatmaps (such as the percentage of the blackboard area being gazed at) and posture angles are recorded every 15 seconds using TobiiPro Glasses3 (sampling rate 120Hz); student sitting posture is captured using OpenPose (if the head pitch angle is >15° and the duration of gazing at the blackboard area within 15 seconds does not exceed 5 seconds, it is considered distracted); when the deviation value is >12 minutes, the system automatically pushes micro-lesson videos (such as function graph transformation animations) and sends a reminder to the teacher's tablet (or other teaching terminal used by the teacher): for example, "The average concentration of students in Class 3 has decreased, it is recommended to increase interactive Q&A."

[0068] Furthermore, the process of constructing the student ability assessment matrix includes:

[0069] S100: The decay coefficients of the first and second error rates are calculated using a time series analysis model, and the weights of knowledge mastery are adjusted in conjunction with the forgetting curve theory.

[0070] In this embodiment, the forgetting curve parameters are as follows: based on the Ebbinghaus forgetting curve model, the knowledge retention rate decay coefficient λ = e^-kt (where k is the memory decay constant). Time series analysis uses the ARIMA model to fit the error rate trend, calculating the short-term volatility coefficient α and the long-term trend coefficient β. The weighting correction formula is as follows:

[0071] E1 represents the first error rate; E2 represents the second error rate.

[0072] S200: The first focus deviation value is graded and quantified by fuzzy logic algorithm to establish a three-dimensional evaluation matrix that includes cognitive input, knowledge absorption rate and error correction efficiency.

[0073] In this embodiment, the first focus deviation value (A1) refers to the percentage of time spent with attention lost in classroom behavior data; error correction efficiency (R) is the ratio of the correctness rate of redoing incorrect questions to the time spent (e.g., correcting 80% of errors within 10 minutes). Focus level is graded as follows: low (A1<0.3), medium (0.3≤A1<0.7), and high (A1≥0.7); correction efficiency is graded as follows: inefficient (R<0.5), medium (0.5≤R<0.8), and efficient (R≥0.8).

[0074] Specifically, cognitive engagement is normalized to a value range of 0-1, knowledge absorption rate is expressed as a percentage, and error correction efficiency is a comprehensive indicator of the speed and accuracy of error correction, expressed as a score of 0-100.

[0075] S300: Based on the k-means++ clustering algorithm, the three-dimensional evaluation matrix is ​​dynamically grouped to generate student clusters with similar learning characteristics.

[0076] In this embodiment, the initial optimization of the k-means++ clustering algorithm refers to selecting the point farthest from the existing cluster center as the new center; the iteration termination condition is that the cluster center movement distance is < ε (e.g., 0.001) or the number of iterations is ≥ 100; for example, the k-means++ clustering algorithm generates 3 clusters, including cluster 1 (highly efficient learners): high cognitive engagement (> 0.8) and high absorption rate (> 0.7);

[0077] Cluster 2 (intermediate learners): medium cognitive engagement (0.4-0.6), medium absorption rate (0.4-0.6); Cluster 3 (group to be improved): low correction efficiency (<0.4).

[0078] S400: Based on the similarity of the learning trajectories of student clusters, associate the influence factors of the first learning state with those of the second learning state.

[0079] In this embodiment, the DTW (Dynamic Time Warping) algorithm is used to match the temporal similarity of students' historical learning paths, where the similarity calculation formula is:

[0080] Where T1 and T2 are time-series learning path data, representing the learning path data of students A and B respectively, analyzing the changes in differences among students in the same task group. This is represented as T1 = [t...]. 11 , t 12 , ..., t 1n ], T2=[t 21 , t 22 , ..., t 2n ]. Euclidean distance sum of squares is the sum of squares of the numerical differences between two time series at the same time point, reflecting the overall degree of difference.

[0081] S3: Based on the error rate and knowledge mastery of students' historical homework data, and combined with the focus deviation value, construct a student ability assessment matrix. Based on the student ability assessment matrix and task attribute characteristics, establish a dynamic task grouping model and generate a task-learning progress report.

[0082] In this embodiment, the ability assessment matrix is ​​a three-dimensional matrix (cognitive engagement × knowledge absorption rate × error correction efficiency). Cognitive engagement = focus deviation value × number of classroom interactions; knowledge absorption rate = (current accuracy rate - accuracy rate one week ago) / knowledge point complexity coefficient; error correction efficiency = accuracy rate of redoing incorrect questions / redoing time; the dynamic task grouping model is a clustering model based on the improved k-medoids algorithm (with added course complexity constraints); the task-learning progress report includes course information, task completion status, recommended learning resource links, next assessment time, and learning progress.

[0083] In vocational education scenarios, dynamic task grouping models prioritize matching skills-based tasks (such as machine tool operation simulations), and task attribute features include equipment type code (CNC-01) and safety operation standard matching degree.

[0084] Specifically, the knowledge mastery level is based on the value after the forgetting curve correction. The k-medoids algorithm (the initial center is selected from the sample with the largest silhouette coefficient) is used to group tasks. The grouping examples are as follows: Group A (characteristics: error rate <20% and high engagement), the teaching strategy is challenging inquiry tasks; Group B (characteristics: error rate 20%-50%), the teaching strategy is tiered consolidation exercises; Group C (characteristics: error rate >50%), the teaching strategy is to fill in the gaps in basic concepts.

[0085] S4: Collect teachers' physiological data through biosensors, combine course information with the complexity index of the dynamic task grouping model, and construct a teacher fatigue assessment model; based on the fatigue assessment results, call the scheduling optimization algorithm to optimize the scheduling plan.

[0086] In this embodiment, biosensor data refers to heart rate variability (HRV) and electrical conductance (EDA) collected by the smart bracelet; the complexity index refers to the gradient value of the Pareto front in the task grouping model (reflecting the teaching load).

[0087] Specifically, the teacher fatigue assessment model is an LSTM model, and the detection data are [HRV mean, EDA rate of change, continuous teaching duration], outputting fatigue levels (0-3). The scheduling optimization adopts a genetic algorithm, which encodes chromosomes based on the course time period (08:00-12:00), teacher allocation, and classroom resources. The corresponding fitness function is 1 / (fatigue level × number of classroom conflicts), and the output includes the fatigue assessment results containing the teacher fatigue level. An example of the optimization result is: the original three consecutive math classes on Wednesday afternoons are changed to math → physical education → math.

[0088] S5: Acquire student learning trajectory data, use machine learning to generate personalized learning plans, and dynamically adjust the weights of learning parameters based on task-learning progress reports through a feedback mechanism.

[0089] In this embodiment, the learning trajectory data includes clickstream data (such as the number of times micro-lesson videos are paused), resource dwell time, homework data, learning habits, and the rate of change in knowledge mastery; the personalized learning plan includes daily tasks, course resource recommendations, time arrangements, and learning progress feedback; parameter weight adjustment refers to the Q-learning reward function in reinforcement learning (such as +10 points for completing a high-difficulty task).

[0090] Specifically, abnormal clickstream data is cleaned (e.g., 5 consecutive clicks within 5 seconds are considered abnormal); a knowledge mastery time series graph is constructed with nodes as knowledge points and edge weights as the rate of change of knowledge mastery; Transformer-XL is used to handle long-term learning behaviors (e.g., window length = 30 days) to track students' learning tasks in terms of knowledge (e.g., if a student has mastered "linear functions", the next step is to learn "quadratic functions") and difficulty matching (if a student always calculates coefficients incorrectly, the system will recommend more basic practice questions, such as drawing parabolic graphs first, and then learning vertex formulas); and then, based on the previous analysis, a personalized learning plan is generated.

[0091] The personalized learning plan's daily tasks include the day's course schedule, and dynamically adjusts the weight of learning parameters including: error alerts, status monitoring, and progress feedback. Error alerts mean that if a student gets three similar questions wrong in a row, the system will automatically insert a "knowledge gap remediation lesson." Status monitoring means that if a student is frequently distracted during class (e.g., the camera captures them turning their head to look out the window), the system will recommend highly interactive tasks (e.g., quiz games). Progress feedback means that if a student completes all tasks ahead of schedule, the system will "add" more challenging questions; if progress is behind, the content will be simplified or review time will be increased.

[0092] Furthermore, when the classroom camera is obstructed by more than 30%, it automatically switches to the desktop camera; when the biosensor is disconnected, historical fatigue data interpolation is enabled to complete the data.

[0093] In one embodiment, such as Figure 2 As shown, the structured homework data includes first homework data and second homework data; after step S1, the intelligent learning analysis and interactive teaching optimization method further includes:

[0094] S11: Based on the data from the first and second assignments, obtain the learning analysis factors related to the mastery of knowledge points. These learning analysis factors include the first and second learning analysis factors.

[0095] In this embodiment, the first assignment data refers to daily classroom practice data (such as in-class quizzes and homework), focusing on process-oriented learning performance; the second assignment data refers to periodic exam / test data, focusing on outcome-oriented knowledge mastery assessment; the learning analysis factors are a set of quantitative indicators reflecting students' learning status; the first learning analysis factor is a short-term learning status indicator generated based on daily assignment data (such as multiple-choice questions, in-class quizzes, etc.), reflecting students' real-time performance and potential problems in recent learning; the second learning analysis factor is a long-term ability assessment indicator generated based on periodic exam data (such as monthly exams, mid-term exams, etc.), reflecting students' knowledge system completeness and deep thinking ability.

[0096] Specifically, knowledge point association analysis includes: for multiple-choice questions: after matching the standard answer, the knowledge point score rate of each question is calculated (e.g., the correct score rate of set theory questions is 82%); for subjective questions: handwritten steps are identified through OCR and scored using the BERT model (e.g., geometry proof questions score 7 / 10).

[0097] The process of obtaining the first and second task data includes:

[0098] S101: Based on the images in the teaching materials, obtain the data for multiple-choice and subjective questions.

[0099] In this embodiment, the multiple-choice question data includes objective question data with question stems, options, and correct answer identifiers; the subjective question data includes non-standardized question data with question stems, student handwritten answers, and grading criteria. The assignment data is obtained using an image segmentation algorithm, where the image segmentation algorithm refers to a deep learning-based image region segmentation model (such as Mask R-CNN).

[0100] Specifically, the image preprocessing steps include: using OpenCV to perform grayscale and binarization processing on the teaching aid images, eliminating paper stains as interference, and locating the question borders using an edge detection algorithm (Canny operator).

[0101] When classifying question types, a CNN classification model is used (input: image of the question area, output: question type label), and then features are extracted. For example, features of multiple-choice questions include: rectangular option boxes, "ABCD" option labels, and the "()" symbol at the end of the question stem; features of subjective questions include: paragraph-style text, no option structure, and the presence of prompt words such as "Answer:".

[0102] S102: Obtain the standard answer matching results based on the multiple-choice question data and the subjective question data.

[0103] In this embodiment, the standard answer database contains a database of correct answers and scoring details for each question; the standard answer matching results are obtained by using an answer matching algorithm, which includes string similarity calculation (such as Levenshtein distance) and formula structure matching (such as MathML comparison).

[0104] Specifically, during answer extraction, for multiple-choice questions: the checkmarks in the option area are read directly (e.g., "●A" indicates that option A is selected); for subjective questions: student answers are identified through handwritten OCR (e.g., "y=2x+1"). Then, mathematical formula standardization and text answer normalization are performed, followed by answer matching calculation.

[0105] S103: Based on the standard answer matching results and question type, obtain the first assignment data and the second assignment data.

[0106] In this embodiment, both the first and second assignment data are associated with corresponding data tags, which include knowledge points, difficulty coefficients, and question type weights. Multiple-choice question features include accuracy, answering time, and option distractibility; subjective question features include step completeness score, key formula coverage, and logical coherence score.

[0107] S12: Based on the factors of learning situation analysis, obtain the learning status influencing factors that represent the students' learning status. Among them, the learning status influencing factors include the first learning status influencing factor and the second learning status influencing factor.

[0108] In this embodiment, the learning status influencing factor is a dynamic indicator system that characterizes the student's real-time learning status; the first learning status factor is a short-term status monitoring indicator based on daily practice; and the second learning status factor is a long-term ability assessment indicator based on exam data.

[0109] Specifically, based on the analysis of learning situations, factors influencing learning status are identified, including:

[0110] S121: Obtain preset knowledge-related factors, and based on the preset knowledge-related factors, filter the first learning situation analysis factors to obtain the first learning state influencing factors. Among them, the first learning state influencing factors include the mastery of the first knowledge point, the first error rate, and the first answering speed.

[0111] In this embodiment, the preset knowledge association factor refers to the predefined knowledge point association rules (such as "to master trigonometric functions, one must first understand trigonometric ratios"); the first knowledge point mastery refers to the student's real-time mastery of basic knowledge (such as the accuracy rate of drawing quadratic function graphs); the first error rate refers to the number of consecutive errors of the same type of question in recent assignments / exams; and the first answering speed refers to the average time spent per unit of question (such as ≤30 seconds per multiple-choice question is considered efficient).

[0112] Specifically, a knowledge graph based on teaching materials is established. If the knowledge graph shows "to master knowledge point A, you must first master knowledge point B", then the learning data related to B is selected first. For example, if a student has not mastered "simplification of fractions" (B), then the answer data of their fractional equation problem (A) is selected.

[0113] S122: Obtain the influencing factor of the first learning state based on the influencing factors of the first learning state.

[0114] In this embodiment, the learning status influencing factor is an algorithm output that converts the original learning data into quantifiable evaluation indicators; the first learning status influencing factor is a set of core indicators used to quantify students' short-term learning performance and real-time learning status, mainly generated based on daily homework and classroom behavior data.

[0115] Specifically, the factors influencing the first learning state include the mastery of the first knowledge point, the first error rate, the first answering speed, and the concentration.

[0116] S123: Based on the preset knowledge-related factors, screen the second learning situation analysis factors to obtain the second learning state influencing factors, which include the second knowledge point mastery, the second error rate, the second answering speed, and the historical error repetition rate.

[0117] In this embodiment, the second knowledge point mastery level refers to the level of knowledge points accumulated by students over a long period of time (such as the average score of the function chapter from the beginning of the semester to the present); the second error rate refers to the tendency of students to make repeated mistakes in advanced question types (such as losing points continuously on the final question); the second answering speed refers to the time management ability to complete complex problems (such as the ratio of the average time spent on problem-solving to the standard time); and the historical error repetition rate refers to the frequency of the recurrence of the same knowledge point error in multiple exams (such as the number of times the error of "vector perpendicular condition" appeared repeatedly in the past 3 exams).

[0118] S124: Obtain the influencing factor of the second learning state based on the influencing factors of the second learning state.

[0119] In this embodiment, the second learning state influencing factor is a set of core indicators used to evaluate students' long-term learning ability and the stability of their knowledge structure. It is mainly generated based on periodic exams, comprehensive test questions, or long-term learning trajectory data.

[0120] Specifically, the first learning state influencing factor focuses on the mastery of basic knowledge and the real-time learning status; the second learning state influencing factor includes the ability to integrate knowledge and the depth of thinking.

[0121] S13: Based on the factors influencing learning status, divide the students into two groups, resulting in the first student group and the second student group.

[0122] In this embodiment, the student group division is based on machine learning algorithms to divide students into groups with different learning characteristics; the first student group refers to students with similar short-term learning status (such as the group with a recent continuous decline in concentration); the second student group refers to students with similar long-term ability characteristics (such as the group with weak mastery of function knowledge).

[0123] Specifically, the labeled characteristics of the student group also include:

[0124] 0: 'High-risk group for attention deficit'

[0125] 1: 'Groups with weak knowledge of functions'

[0126] 2: 'A group with stable computing power'

[0127] 3: 'Potential Enhancement Group'

[0128] 4: 'Group with excellent overall abilities'.

[0129] Based on the above characteristics, in actual learning analysis, the labeling characteristics of the first and second student groups can be defined by ourselves, and the academic performance of different characteristic groups can be analyzed.

[0130] S14: Obtain personalized teaching parameters based on the first and second student groups.

[0131] In this embodiment, personalized teaching parameters include configuration parameters for teaching resource recommendation weights and intervention strategy priorities; the dynamic adaptation mechanism refers to adjusting parameter thresholds in real time based on group characteristics.

[0132] Specifically, the baseline settings for adjusting parameters are as follows: Recommended resource weights: {Videos: 0.3, Question Bank: 0.5, Gamification: 0.2}; Intervention trigger thresholds: {Focus < 60: High priority, Error rate > 30%: Medium priority}.

[0133] For example, we can optimize the weight of recommended learning resources based on the different groups to which students belong, and set corresponding teaching intervention reminder rules:

[0134] If it is Group 0 (students with low concentration and easy to get distracted):

[0135] The system places a greater emphasis on gamified learning content (50%), with a significant amount of question bank practice (40%) and fewer video explanations (10%). When a student's concentration drops below 70%, the system will issue a medium-priority reminder, prompting the teacher or the system to intervene appropriately.

[0136] If it's Group 1 (students who don't have a good grasp of function concepts):

[0137] The system primarily recommends video explanations (60%), supplemented by a certain amount of question bank practice (30%), with very little gamified content (10%). When the error rate of these students exceeds 25%, the system will issue a high-priority reminder and provide timely and focused tutoring.

[0138] In this embodiment, the attention evaluation model is a multimodal attention evaluation model; the method further includes:

[0139] S10: Construct a multimodal attention assessment model based on the body movement characteristics and eye focus trajectory in classroom behavior data.

[0140] In this embodiment, the multimodal attention model is a deep learning model that integrates multi-dimensional data such as visual (body movements) and visual (eye focus); body movement features refer to biological characteristics such as head tilt angle, gesture frequency, and torso tilt; eye focus trajectory refers to the movement path of the student's eyes between teaching materials, blackboard, and teacher.

[0141] Specifically, student body movement sequences were captured using an infrared camera (sampling rate ≥ 60fps), including head pitch angle, gesture frequency, and torso tilt. The body movement data was denoised using Kalman filtering, and the gaze trajectory data was normalized using Z-Score (μ = 0, σ = 1). Eye tracker data (sampling rate ≥ 120Hz) was collected, recording pupil position and fixation area (blackboard / teaching materials / teacher). When eye tracker data was lost for more than 10 seconds, historical attention mean interpolation was used. A Transformer multimodal fusion network was constructed, with the input layer containing a visual encoder (ViT) and a pose encoder (LSTM). The loss function used was cross-entropy loss, and the optimizer was AdamW. "The infrared camera used was a FLIRAX8 (640×480 resolution), the eye tracker was a Tobii Pro Glasses 3 (sampling rate 120Hz), and the data was stored in a local MySQL database (version 8.0)." "The visual encoder uses the ViT-B / 16 model (ImageNet pre-trained weights), the pose encoder LSTM has a hidden layer dimension of 256, and the Transformer has 8 multi-head attention heads."

[0142] S20: Based on the sliding window algorithm, segment and label continuous classroom segments, and calculate the student attention fluctuation index within each time period.

[0143] In this embodiment, the sliding window algorithm divides the continuous class time into segments of fixed length (e.g., every 5 minutes); the attention fluctuation index is used to quantify the statistical indicators of changes in students' concentration in different time periods.

[0144] Specifically, a sliding window algorithm (window length T_window = 300s, step size T_step = 30s) is used to divide continuous classroom behavior data into frames; each frame of data is labeled with an attention label:

[0145] 1 (Focus): Continuous gaze at teaching materials for ≥180s and gesture frequency ≤2 times / min;

[0146] 0 (Distraction): The duration of eye contact with supplementary teaching materials is ≥60 seconds or the rate of change in head pitch angle is ≥0.5 rad / s. 2 .

[0147] Calculate the rate of change of focus labels in adjacent windows and normalize the volatility index.

[0148] S30: Based on the covariance analysis of the student attention fluctuation index and the knowledge point difficulty coefficient matrix of the corresponding teaching materials, dynamically generate a set of teaching strategy adjustment instructions.

[0149] Specifically, the knowledge point difficulty coefficient matrix has preset difficulty levels for each knowledge point (e.g., easy: 1, difficult: 5); covariance analysis is used to quantify the correlation between attention fluctuations and knowledge point difficulty.

[0150] Specifically, a Knowledge Difficulty Matrix (KDM) is established, where each element Dij represents the difficulty level (level 1-5) of the i-th knowledge point in the j-th teaching stage.

[0151] Calculate the covariance between the attention fluctuation index and the difficulty of the knowledge point:

[0152]

[0153] Among them, V iT Let D be the attention fluctuation index for the i-th knowledge point in time period T. i Let T be the difficulty coefficient of the knowledge point in the time period T of the i-th knowledge point; and Cov represents the average value; n is the total number of segments into which class time is divided. Cov > 0 indicates that attention fluctuation is positively correlated with the difficulty of the knowledge points.

[0154] The rule base for generating teaching strategies includes:

[0155] covariance sign Correlation strength Types of teaching strategies >0.7 Strong positive correlation Reduce the difficulty of knowledge points and insert examples for explanation. <-0.5 Strong negative correlation Increase interactive Q&A to enhance participation. [-0.1，0.1] No significant correlation Maintain the current teaching pace

[0156] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0157] In one embodiment, a system based on intelligent learning analysis and interactive teaching optimization is provided, which corresponds to the method based on intelligent learning analysis and interactive teaching optimization in the above embodiments.

[0158] The system, based on intelligent learning analysis and interactive teaching optimization, includes modules for digitizing teaching materials, collecting and analyzing classroom behavior, assessing student abilities and grouping tasks, optimizing class scheduling, and generating personalized learning plans. Detailed descriptions of each module are as follows:

[0159] The teaching materials digitization module is used to perform text recognition and question analysis on teaching materials images, and generate structured homework data containing question types and knowledge point tags;

[0160] The classroom behavior collection and analysis module is used to collect student classroom behavior data, calculate focus deviation values ​​based on knowledge point tags and attention assessment models, and trigger adjustments to teaching strategies.

[0161] The student ability assessment and task grouping module is used to construct a student ability assessment matrix based on the error rate and knowledge mastery of students' historical homework data, combined with the attention deviation value. Based on the student ability assessment matrix and task attribute characteristics, a dynamic task grouping model is established to generate a task-learning progress report.

[0162] The scheduling optimization module collects teachers' physiological data through biosensors, combines course information with the complexity index of the dynamic task grouping model, and constructs a teacher fatigue assessment model. Based on the fatigue assessment results, it calls the scheduling optimization algorithm to optimize the scheduling scheme. The personalized learning plan generation module obtains students' learning trajectory data, uses machine learning to generate personalized learning plans, and dynamically adjusts the weights of learning parameters based on task-learning progress reports through a feedback mechanism.

[0163] For specific limitations regarding the system based on intelligent learning analysis and interactive teaching optimization, please refer to the limitations of the method based on intelligent learning analysis and interactive teaching optimization mentioned above, which will not be repeated here. Each module in the above-mentioned system based on intelligent learning analysis and interactive teaching optimization can be implemented in whole or in part through software, hardware, or a combination thereof. Each module can be embedded in the processor of the computer device in hardware form or independent of it, or it can be stored in the memory of the computer device in software form, so that the processor can call and execute the corresponding operations of each module.

[0164] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, the computer program performing the following steps when executed by a processor:

[0165] S1: Perform text recognition and question analysis on images of teaching materials to generate structured homework data containing question types and knowledge point tags;

[0166] S2: Collect student classroom behavior data, calculate focus deviation value based on knowledge point tags and attention assessment model, and trigger adjustment of teaching strategies;

[0167] S3: Based on the error rate and knowledge mastery of students' historical homework data, and combined with the focus deviation value, construct a student ability assessment matrix. Based on the student ability assessment matrix and task attribute characteristics, establish a dynamic task grouping model and generate a task-learning progress report.

[0168] S4: Collect teachers' physiological data through biosensors, combine course information with the complexity index of the dynamic task grouping model, and construct a teacher fatigue assessment model; based on the fatigue assessment results, call the scheduling optimization algorithm to optimize the scheduling plan;

[0169] S5: Acquire student learning trajectory data, use machine learning to generate personalized learning plans, and dynamically adjust the weights of learning parameters based on task-learning progress reports through a feedback mechanism.

[0170] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0171] In one embodiment, particularly according to an embodiment of the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present invention includes a computer program product comprising a computer program / instructions that, when executed by a processor, implement the steps of the method for optimizing intelligent learning and interactive teaching as described above. In such an embodiment, the computer program can be downloaded and installed from a network via a communication module, and / or installed from a removable medium. When the computer program is executed by a central processing unit (CPU), it performs the various functions defined in the present invention.

[0172] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is used as an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0173] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method based on intelligent learning analysis and interactive teaching optimization, characterized in that: include: Text recognition and question parsing are performed on images of teaching materials to generate structured homework data containing question types and knowledge point tags; Collect student classroom behavior data, calculate focus deviation value based on the knowledge point tags and attention assessment model, and trigger adjustments to teaching strategies. Based on the error rate and knowledge mastery of students' historical homework data, and combined with the attention deviation value, a student ability assessment matrix is ​​constructed. Based on the student ability assessment matrix and task attribute characteristics, a dynamic task grouping model is established to generate a task-learning progress report. Physiological data of teachers is collected by biosensors, and a teacher fatigue assessment model is constructed by combining course information with the complexity index of the dynamic task grouping model. Based on the fatigue assessment results, a scheduling optimization algorithm is called to optimize the scheduling scheme. Acquire student learning trajectory data, use machine learning to generate personalized learning plans, and dynamically adjust the weights of learning parameters based on the task-learning progress report through a feedback mechanism; The structured job data includes first job data and second job data; After performing text recognition and question parsing on the images of the teaching materials, the process also includes: Based on the first assignment data and the second assignment data, learning analysis factors related to the mastery of knowledge points are obtained, wherein the learning analysis factors include a first learning analysis factor and a second learning analysis factor; Based on the learning situation analysis factors, learning status influencing factors representing students' learning status are obtained, wherein the learning status influencing factors include a first learning status influencing factor and a second learning status influencing factor. Based on the learning status influencing factors, students are divided into two groups: the first student group and the second student group. Personalized teaching parameters are obtained based on the first student group and the second student group; The attention assessment model is a multimodal attention assessment model; the method further includes: Based on the body movement characteristics and eye focus trajectory in the classroom behavior data, a multimodal attention assessment model is constructed. The continuous classroom segments are segmented and labeled using the sliding window algorithm, and the student attention fluctuation index is calculated within each time period. Based on the covariance analysis of the student attention fluctuation index and the knowledge point difficulty coefficient matrix of the corresponding teaching materials, a set of teaching strategy adjustment instructions is dynamically generated. The step of obtaining learning status influencing factors based on the learning situation analysis factors specifically includes: Obtain preset knowledge-related factors, and based on the preset knowledge-related factors, filter the first learning situation analysis factors to obtain the first learning state influencing factors, wherein the first learning state influencing factors include the first knowledge point mastery, the first error rate, and the first answering speed; Based on the factors influencing the first learning state, obtain the first learning state influencing factor; Based on the preset knowledge-related factors, the second learning situation analysis factors are screened to obtain the second learning state influencing factors, wherein the second learning state influencing factors include the second knowledge point mastery, the second error rate, the second answering speed, and the historical error repetition rate; Based on the factors influencing the second learning state, obtain the influencing factor of the second learning state; The process of constructing the student ability assessment matrix includes: The decay coefficients of the first error rate and the second error rate are calculated using a time series analysis model, and the weights of knowledge point mastery are adjusted in conjunction with the forgetting curve theory. The first focus deviation value is graded and quantified by using fuzzy logic algorithm, and a three-dimensional evaluation matrix including cognitive input, knowledge absorption rate and error correction efficiency is established. The three-dimensional evaluation matrix is ​​dynamically grouped based on the k-means++ clustering algorithm to generate student clusters with similar learning characteristics. Based on the similarity of the learning trajectories of the student clusters, the first learning state influence factor and the second learning state influence factor are associated. The method for establishing the dynamic task grouping model includes: Based on the cluster center coordinates of the student clusters, calculate the cosine similarity matrix between the task attribute features and the student feature vectors. An improved k-medoids algorithm is used to cluster and optimize the cosine similarity matrix, generating a task grouping scheme with multiple grouping options; A grouping stability prediction model is established using reinforcement learning algorithms, and the confidence probability distribution of each grouping scheme is output. Based on the confidence probability distribution and the class hour allocation constraints in the course information, an adaptive teaching resource push strategy is generated through a multi-objective decision tree.

2. The method for optimizing teaching based on intelligent learning analysis and interactive teaching according to claim 1, characterized in that, The process of acquiring the first job data and the second job data includes: Based on the images in the teaching materials, obtain multiple-choice and subjective-choice homework data; Based on the multiple-choice and subjective question data, obtain the standard answer matching results; Based on the standard answer matching results and the question type, obtain the first assignment data and the second assignment data.

3. A system based on intelligent learning analysis and interactive teaching optimization, characterized in that: The system, applied to the intelligent learning analysis and interactive teaching optimization method as described in any one of claims 1 to 2, comprises: The teaching materials digitization module is used to perform text recognition and question analysis on teaching materials images, and generate structured homework data containing question types and knowledge point tags; The classroom behavior collection and analysis module is used to collect student classroom behavior data, calculate the focus deviation value based on the knowledge point tags and attention assessment model, and trigger adjustments to teaching strategies. The student ability assessment and task grouping module is used to construct a student ability assessment matrix based on the error rate and knowledge mastery in students' historical homework data, combined with the focus deviation value, and to establish a dynamic task grouping model based on the student ability assessment matrix and task attribute characteristics, and generate a task-learning progress report. The scheduling optimization module is used to collect teachers' physiological data through biosensors, combine course information with the complexity index of the dynamic task grouping model, and construct a teacher fatigue assessment model; based on the fatigue assessment results, the scheduling optimization algorithm is called to optimize the scheduling scheme. The personalized learning plan generation module is used to acquire student learning trajectory data, generate personalized learning plans using machine learning, and dynamically adjust the weights of learning parameters based on the task-learning progress report through a feedback mechanism. The structured homework data includes first homework data and second homework data; after performing text recognition and question parsing on the teaching aid images, it also includes: Based on the first assignment data and the second assignment data, learning analysis factors related to the mastery of knowledge points are obtained, wherein the learning analysis factors include a first learning analysis factor and a second learning analysis factor; Based on the learning situation analysis factors, learning status influencing factors representing students' learning status are obtained, wherein the learning status influencing factors include a first learning status influencing factor and a second learning status influencing factor. Based on the learning status influencing factors, students are divided into two groups: the first student group and the second student group. Personalized teaching parameters are obtained based on the first student group and the second student group; The attention assessment model is a multimodal attention assessment model; the method further includes: Based on the body movement characteristics and eye focus trajectory in the classroom behavior data, a multimodal attention assessment model is constructed. The continuous classroom segments are segmented and labeled using the sliding window algorithm, and the student attention fluctuation index is calculated within each time period. Based on the covariance analysis of the student attention fluctuation index and the knowledge point difficulty coefficient matrix of the corresponding teaching materials, a set of teaching strategy adjustment instructions is dynamically generated. The step of obtaining learning status influencing factors based on the learning situation analysis factors specifically includes: Obtain preset knowledge-related factors, and based on the preset knowledge-related factors, filter the first learning situation analysis factors to obtain the first learning state influencing factors, wherein the first learning state influencing factors include the first knowledge point mastery, the first error rate, and the first answering speed; Based on the factors influencing the first learning state, obtain the first learning state influencing factor; Based on the preset knowledge-related factors, the second learning situation analysis factors are screened to obtain the second learning state influencing factors, wherein the second learning state influencing factors include the second knowledge point mastery, the second error rate, the second answering speed, and the historical error repetition rate; Based on the factors influencing the second learning state, obtain the influencing factor of the second learning state; The process of constructing the student ability assessment matrix includes: The decay coefficients of the first error rate and the second error rate are calculated using a time series analysis model, and the weights of knowledge point mastery are adjusted in conjunction with the forgetting curve theory. The first focus deviation value is graded and quantified by using fuzzy logic algorithm, and a three-dimensional evaluation matrix including cognitive input, knowledge absorption rate and error correction efficiency is established. The three-dimensional evaluation matrix is ​​dynamically grouped based on the k-means++ clustering algorithm to generate student clusters with similar learning characteristics. Based on the similarity of the learning trajectories of the student clusters, the first learning state influence factor and the second learning state influence factor are associated. The method for establishing the dynamic task grouping model includes: Based on the cluster center coordinates of the student clusters, calculate the cosine similarity matrix between the task attribute features and the student feature vectors. An improved k-medoids algorithm is used to cluster and optimize the cosine similarity matrix, generating a task grouping scheme with multiple grouping options; A grouping stability prediction model is established using reinforcement learning algorithms, and the confidence probability distribution of each grouping scheme is output. Based on the confidence probability distribution and the class hour allocation constraints in the course information, an adaptive teaching resource push strategy is generated through a multi-objective decision tree.

4. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the intelligent learning analysis and interactive teaching optimization method as described in any one of claims 1 to 2.

5. A computer program product, comprising a computer program / instructions, characterized in that, When the computer program / instruction is executed by the processor, it implements the steps of the intelligent learning analysis and interactive teaching optimization method as described in any one of claims 1 to 2.

Images