Course quality supervision analysis method and system based on big data

Abstract

The invention relates to the technical field of course supervision, in particular to a course quality supervision analysis method and system based on big data. The method comprises the following steps: acquiring course data of a course teaching platform to carry out course teaching planning design, and generating course teaching planning data; obtaining course teaching log data of the course teaching platform to perform learning attention deviation evaluation and course response evaluation during teaching of each course, and generating response evaluation data corresponding to teaching of each course; and obtaining historical teaching information data to carry out course teaching result analysis and course teaching quality evaluation, carrying out course structure adjustment processing on the course teaching planning data, generating course teaching planning data after structure adjustment, and carrying out updating processing on the course data of the course teaching platform. According to the invention, real-time assessment of course quality and course optimization are realized through a whole-process supervision mechanism of behavior deviation, course response and teaching achievements.

Description

Technical Field

[0001] This invention relates to the field of course supervision technology, and in particular to a course quality supervision and analysis method and system based on big data. Background Technology

[0002] With the widespread deployment of teaching platforms and intelligent education systems, online teaching platforms can continuously collect multi-source data, including basic course data, course teaching logs, student behavior trajectories, and learning interaction records, constructing a large-scale big data resource system for course teaching. However, in existing technologies, course quality monitoring and analysis methods lack the ability to model the correlation between students' learning behavior and teaching elements such as course structure, teaching pace, and interaction patterns. This makes it difficult to identify behavioral deviation trends and teaching response characteristics during the learning process, resulting in a lag in the response of course teaching quality optimization strategies and a relative delay in the effective iteration of course structure and quality improvement. Summary of the Invention

[0003] Based on this, the present invention provides a course quality supervision and analysis method and system based on big data to solve at least one of the above-mentioned technical problems.

[0004] To achieve the above objectives, a big data-based method for course quality supervision and analysis includes the following steps:

[0005] Step S1: Obtain course data from the course teaching platform; design course teaching plans based on the course data, and generate course teaching plan data;

[0006] Step S2: Obtain course teaching log data from the course teaching platform; based on the course teaching log data and course teaching plan data, conduct learning attention deviation assessments for each course teaching session, and generate learning attention deviation data corresponding to each course teaching session; based on the learning attention deviation data corresponding to each course teaching session, conduct course response assessments for the corresponding courses within the course teaching plan data, and generate response assessment data corresponding to each course teaching session.

[0007] Step S3: Obtain historical teaching information data; based on the historical teaching information data and the learning attention deviation data corresponding to each course, analyze the teaching outcomes of the courses and generate teaching outcome data;

[0008] Step S4: Based on the teaching outcome data and the response evaluation data corresponding to each course, evaluate the course teaching planning data to generate course teaching quality data;

[0009] Step S5: Based on the course teaching quality data and the learning attention deviation data corresponding to each course, adjust the course teaching plan data to generate the adjusted course teaching plan data and update the course data on the course teaching platform.

[0010] Furthermore, step S1 includes the following steps:

[0011] Step S11: Obtain course data from the course teaching platform;

[0012] Step S12: Analyze the course adaptation groups based on the course data and generate course adaptation group data;

[0013] Step S13: Design the total course volume based on the target audience of the course and generate total course volume data;

[0014] Step S14: Based on the data of the target group for the course and the total number of courses, design the course teaching plan and generate course teaching plan data.

[0015] Furthermore, step S2 includes the following steps:

[0016] Step S21: Obtain course teaching log data from the course teaching platform;

[0017] Step S22: Identify student learning behavior status during course instruction based on course teaching log data, and generate student learning behavior status data;

[0018] Step S23: Based on student learning behavior status data and course teaching plan data, conduct learning attention deviation assessment during the teaching of each course, and generate learning attention deviation data corresponding to each course teaching.

[0019] Step S24: Based on the learning attention deviation data corresponding to each course, conduct a course response evaluation on the corresponding course teaching within the course teaching plan data, and generate response evaluation data for each course teaching.

[0020] Furthermore, step S23 includes the following steps:

[0021] Step S231: Analyze the content sequence of the course based on the course teaching plan data to generate course content sequence data;

[0022] Step S232: Analyze the complexity of the course content based on the order data of the course content, and generate course content complexity data;

[0023] Step S233: Based on the complexity data of the course content, deduce the logical connection of the course content sequence data to generate logical connection deduction data of the course content;

[0024] Step S234: Based on the logical connection of the course content, derive data to identify the interaction in the course teaching and generate course teaching interaction data;

[0025] Step S235: Calculate the course learning cycle based on student learning behavior status data and generate course learning cycle data;

[0026] Step S236: Based on the course teaching interaction data and course learning cycle data, conduct an assessment of the learning attention deviation during the teaching of each course, and generate learning attention deviation data corresponding to each course.

[0027] Furthermore, step S236 includes the following steps:

[0028] Based on the data on course teaching interaction, the interaction patterns of each course teaching are identified, and corresponding interaction pattern data for each course teaching is generated.

[0029] Based on the course learning cycle, conduct learning depth value-added analysis for each course teaching, and generate learning depth value-added data corresponding to each course teaching.

[0030] Based on the learning depth value-added data and interaction mode data corresponding to each course, the interaction frequency analysis of each course teaching is carried out to generate interaction frequency data corresponding to each course teaching.

[0031] Based on the interaction frequency data corresponding to each course, the learning attention deviation during each course teaching is assessed, and the learning attention deviation data corresponding to each course teaching is generated.

[0032] Furthermore, step S24 includes the following steps:

[0033] Step S241: Monitor student learning behavior fluctuations based on the learning attention deviation data corresponding to each course, and generate behavior fluctuation data;

[0034] Step S242: Locate the content of interest courses based on behavioral fluctuation data and generate interest course content data;

[0035] Step S243: Based on the interest course content data, perform course-related content retrieval and identification to generate course-related content retrieval data;

[0036] Step S244: Based on the data retrieved from the course-related content, conduct a course response evaluation on the corresponding course teaching within the course teaching plan data, and generate response evaluation data for each course teaching.

[0037] Furthermore, step S3 includes the following steps:

[0038] Step S31: Obtain historical teaching information data;

[0039] Step S32: Plot the course's teaching history stage outcome curves based on historical teaching information data, and generate teaching history stage outcome curve data;

[0040] Step S33: Assess students' mastery of course knowledge based on the learning attention deviation data corresponding to each course, and generate course knowledge mastery data;

[0041] Step S34: Calculate the course teaching outcome deviation based on course knowledge mastery data and course teaching outcome curve data at historical teaching stages, and generate course teaching outcome deviation data;

[0042] Step S35: Analyze the course teaching outcomes based on the deviation data of course teaching outcomes and generate teaching outcome data.

[0043] Furthermore, step S4 includes the following steps:

[0044] Step S41: Analyze the focus of each course based on the response evaluation data corresponding to the teaching of each course, and generate course focus data;

[0045] Step S42: Assess students' understanding of course knowledge based on teaching outcome data and generate course knowledge understanding data;

[0046] Step S43: Based on the course focus content data and course knowledge comprehension data, perform course teaching matching deviation diagnosis and generate course teaching matching deviation data;

[0047] Step S44: Evaluate the course teaching quality based on the course teaching matching deviation data and generate course teaching quality data.

[0048] Furthermore, step S5 includes the following steps:

[0049] Step S51: Analyze the fit between students and teachers' course pace based on the course teaching quality data, and generate course pace fit data;

[0050] Step S52: Analyze the relevant interest courses based on the learning attention deviation data corresponding to each course, and generate relevant interest course data;

[0051] Step S53: Based on the course rhythm fit data and related interest course data, perform course teaching collaboration matching to generate course teaching collaboration data;

[0052] Step S54: Based on the course teaching collaboration data, perform structural adjustment processing on the course teaching planning data to generate structurally adjusted course teaching planning data and update the course data on the teaching interaction platform.

[0053] Furthermore, the present invention also provides a big data-based course quality supervision and analysis system for executing the big data-based course quality supervision and analysis method described above. This big data-based course quality supervision and analysis system includes:

[0054] The course planning module is used to acquire course data from the course teaching platform; design course teaching plans based on the course data; and generate course teaching plan data.

[0055] The course teaching attention response analysis module is used to acquire course teaching log data from the course teaching platform; based on the course teaching log data and course teaching plan data, it evaluates the learning attention deviation during each course teaching and generates learning attention deviation data corresponding to each course teaching; based on the learning attention deviation data corresponding to each course teaching, it evaluates the course teaching response within the course teaching plan data and generates response evaluation data corresponding to each course teaching.

[0056] The course teaching outcome analysis module is used to acquire historical teaching information data; based on the historical teaching information data and the learning attention deviation data corresponding to each course, the teaching outcome of the course is analyzed and teaching outcome data is generated.

[0057] The course quality assessment module is used to assess the course teaching quality based on teaching outcome data and response assessment data corresponding to each course, and to generate course teaching quality data.

[0058] The course structure adjustment module is used to adjust the course structure based on course teaching quality data and learning attention deviation data corresponding to each course, generate adjusted course teaching plan data, and update the course data on the course teaching platform.

[0059] The beneficial effects of this invention are:

[0060] 1. The course quality supervision and analysis method based on big data proposed in this invention, compared with existing technologies, has the following advantages: By acquiring course data from the course teaching platform and conducting in-depth analysis of this data, including elements such as course objectives, student knowledge reserves, and the suitability of teaching resources, a course teaching plan that meets the actual needs of students and the laws of teaching can be formulated. Based on course teaching log data and course teaching plan data, the evaluation of learning attention deviation during each course's teaching can promptly identify problems such as students' lack of concentration and low participation in the learning process. Through in-depth mining and analysis of this data, the generated learning attention deviation data can accurately pinpoint specific courses, teaching links, and affected student groups, providing teachers with clear directions for adjusting teaching strategies. Furthermore, based on the learning attention deviation data, the course teaching corresponding to the course teaching plan data is evaluated to generate response evaluation data, which can comprehensively assess the adaptability and effectiveness of the teaching plan in actual teaching. Through this dynamic evaluation mechanism, teachers can promptly identify shortcomings in the teaching process, quickly make adjustments, improve the pertinence and flexibility of teaching, enhance students' learning interest and participation, and thus improve teaching effectiveness. Analyzing course teaching outcomes based on historical teaching data and corresponding attention deviation data for each course allows for in-depth analysis of teaching effectiveness from multiple perspectives. By comparing teaching outcomes across different periods and classes, successful experiences and existing problems in the teaching process can be summarized. Furthermore, combining attention deviation data allows for analysis of the impact of students' lack of concentration on teaching outcomes, identifying key factors affecting course teaching quality. Evaluating course teaching quality based on teaching outcome data and corresponding response assessment data for each course involves a comprehensive, objective, and systematic evaluation of course teaching, considering aspects such as the achievement of teaching objectives, the rationality of teaching content, the applicability of teaching methods, the smoothness of the teaching process, and student learning outcomes. For different courses, in-depth analysis of the causes allows for targeted course improvements. Finally, adjusting the course structure based on course teaching quality data and corresponding attention deviation data for each course enables precise optimization of course teaching. During the course structure adjustment process, teaching objectives can be re-examined and adjusted based on the results of teaching quality assessments to ensure that the objectives align with students' actual needs and developmental goals. Addressing attention deficit issues, teaching content and methods can be optimized to enhance the fun and appeal of instruction, thereby increasing student interest and participation. Generating revised course teaching plan data and updating course data on the teaching platform allows for timely application of the optimized plan to actual teaching, achieving dynamic adjustment and continuous improvement of the teaching process.By modeling the relationship between students' learning behavior and teaching elements such as curriculum structure, teaching pace, and interaction patterns, and by identifying behavioral deviation trends and teaching response characteristics during the learning process, we can achieve effective iteration of curriculum structure and improvement of curriculum quality.

[0061] 2. The big data-based course quality supervision and analysis system proposed in this invention consists of a course planning module, a course teaching attention response analysis module, a course teaching outcome analysis module, a course quality evaluation module, and a course structure adjustment module. It can realize any big data-based course quality supervision and analysis method described in this invention. It is used to combine the operations between the computer programs running on each module to realize the big data-based course quality supervision and analysis method. The internal structure of the system cooperates with each other, which can greatly reduce repetitive work and manpower input, and can quickly and effectively provide a more accurate and efficient big data-based course quality supervision and analysis process, thereby simplifying the operation process of the big data-based course quality supervision and analysis system. Attached Figure Description

[0062] Figure 1 This is a flowchart illustrating the steps of a course quality supervision and analysis method based on big data according to the present invention.

[0063] Figure 2 for Figure 1 A detailed flowchart illustrating the implementation steps of step S1.

[0064] Figure 3 for Figure 1 A detailed flowchart illustrating the implementation steps of step S2.

[0065] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0066] The technical method of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without inventive effort are within the scope of protection of this invention.

[0067] Furthermore, the accompanying drawings are merely illustrative of the invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor methods and / or microcontroller methods.

[0068] It should be understood that although the terms "first," "second," etc., may be used herein to describe various units, these units should not be limited by these terms. These terms are used merely to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, a first unit may be referred to as a second unit, and similarly, a second unit may be referred to as a first unit. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0069] To achieve the above objectives, please refer to Figures 1 to 3 This invention provides a course quality supervision and analysis method based on big data, comprising the following steps:

[0070] Step S1: Obtain course data from the course teaching platform; design course teaching plans based on the course data, and generate course teaching plan data;

[0071] Step S2: Obtain course teaching log data from the course teaching platform; based on the course teaching log data and course teaching plan data, conduct learning attention deviation assessments for each course teaching session, and generate learning attention deviation data corresponding to each course teaching session; based on the learning attention deviation data corresponding to each course teaching session, conduct course response assessments for the corresponding courses within the course teaching plan data, and generate response assessment data corresponding to each course teaching session.

[0072] Step S3: Obtain historical teaching information data; based on the historical teaching information data and the learning attention deviation data corresponding to each course, analyze the teaching outcomes of the courses and generate teaching outcome data;

[0073] Step S4: Based on the teaching outcome data and the response evaluation data corresponding to each course, evaluate the course teaching planning data to generate course teaching quality data;

[0074] Step S5: Based on the course teaching quality data and the learning attention deviation data corresponding to each course, adjust the course teaching plan data to generate the adjusted course teaching plan data and update the course data on the course teaching platform.

[0075] In the embodiments of this invention, please refer to Figure 1 The diagram shown illustrates the steps of the course quality supervision and analysis method based on big data according to the present invention. In this example, the course quality supervision and analysis method based on big data includes the following steps:

[0076] Step S1: Obtain course data from the course teaching platform; design course teaching plans based on the course data, and generate course teaching plan data;

[0077] In this embodiment of the invention, data scraping tools are used to extract course data from the database of the course teaching platform according to established data interface specifications. This data covers the teaching objectives stipulated in the course syllabus, the teaching content corresponding to each chapter, the preset teaching schedule, and detailed information on suitable teaching resources such as courseware and videos. Data mining algorithms in data analysis software, such as the Apriori association rule mining algorithm, are used to perform in-depth analysis of the course data. The relationship between teaching objectives and corresponding teaching content is analyzed, and the difficulty and focus of the teaching content are adjusted according to students' professional background and academic foundation. For example, if students in a certain major have a weak foundation in mathematics, and the course involves a lot of mathematical derivation, the explanation of basic mathematical knowledge is appropriately increased, and the teaching pace is slowed down. Based on the analysis results, a detailed course teaching plan is formulated, clarifying the specific teaching tasks of each teaching stage, the teaching methods used (e.g., lectures combined with case analysis for theoretical courses, and project-driven methods for practical courses), and the precise allocation of teaching time, ultimately generating course teaching plan data.

[0078] Step S2: Obtain course teaching log data from the course teaching platform; based on the course teaching log data and course teaching plan data, conduct learning attention deviation assessments for each course teaching session, and generate learning attention deviation data corresponding to each course teaching session; based on the learning attention deviation data corresponding to each course teaching session, conduct course response assessments for the corresponding courses within the course teaching plan data, and generate response assessment data corresponding to each course teaching session.

[0079] In this embodiment of the invention, a log parsing tool is used to analyze the course teaching log data recorded on the course teaching platform. The log data includes information such as the time students log into the course platform, the frequency and content of their participation in class discussions, the time and quality of their homework submissions, and their answers to online tests. A learning attention deviation assessment model is constructed, using time series analysis methods, such as the ARIMA model, to analyze the sequence of students' behavioral data during the course teaching period. For example, if students experience a sharp decrease in the number of times they log into the platform, consistently low participation in class discussions, or a significant increase in delayed or error-prone homework submissions within a specific time period, it can be determined that students' learning attention has deviated during that period. For each course, the assessment model outputs corresponding learning attention deviation data, such as the percentage of periods of focused attention and the peak periods of inattentiveness. A course response assessment model is constructed, using the Analytic Hierarchy Process (AHP) to determine the weights of assessment indicators, which cover the attractiveness of teaching methods to students' attention and the matching degree between teaching progress and students' learning abilities. The learning attention deviation data is then substituted into the model to calculate the response assessment value of each course's teaching plan in actual teaching, generating response assessment data for each course, thereby judging the actual implementation effect of the course teaching plan.

[0080] Step S3: Obtain historical teaching information data; based on the historical teaching information data and the learning attention deviation data corresponding to each course, analyze the teaching outcomes of the courses and generate teaching outcome data;

[0081] In this embodiment of the invention, historical teaching information data is obtained from the school's teaching management database, including information such as the distribution of exam scores for each course in past semesters, student evaluations, and teacher teaching summaries. Combined with the generated learning attention deviation data corresponding to each course, correlation analysis, such as Pearson correlation coefficient analysis, is used to explore the relationship between student learning attention deviation and teaching outcomes. For example, the analysis found that in a certain course, students' attention was concentrated on the explanation of complex theories, and the exam scores for this part of the knowledge points were generally low, indicating that attention deviation has a negative impact on teaching outcomes. Using classification algorithms in data mining, such as decision tree algorithms, the course teaching outcomes are classified into categories such as excellent, good, average, and poor based on the historical teaching information data and the learning attention deviation data, generating teaching outcome data to provide a basis for subsequent teaching improvement.

[0082] Step S4: Based on the teaching outcome data and the response evaluation data corresponding to each course, evaluate the course teaching planning data to generate course teaching quality data;

[0083] In this embodiment of the invention, a course teaching quality evaluation model is constructed, employing the fuzzy comprehensive evaluation method. Indicators such as the excellent rate of exam scores and student satisfaction with teaching evaluations from the teaching outcome data, and indicators such as the adaptability of teaching methods and the rationality of teaching progress from the response evaluation data, are comprehensively calculated according to different weights. The weight of each indicator is determined by combining expert scoring with historical data verification. For example, the weight of the excellent rate of exam scores is set to 0.4, and the weight of the adaptability of teaching methods is set to 0.3. The teaching outcome data and response evaluation data of each course are substituted into the model to calculate the comprehensive teaching quality score for each course. Teaching quality levels are then divided according to the score range, such as 90-100 points for excellent, 80-89 points for good, etc., generating course teaching quality data that comprehensively and objectively reflects the level of course teaching quality.

[0084] Step S5: Based on the course teaching quality data and the learning attention deviation data corresponding to each course, adjust the course teaching plan data to generate the adjusted course teaching plan data and update the course data on the course teaching platform.

[0085] In this embodiment of the invention, based on the quality level and score in the course teaching quality data, and the problems reflected by the learning attention deviation data, a course structure optimization algorithm, such as a genetic algorithm, is used to adjust the course teaching plan data. If a course has a low teaching quality score, and the learning attention deviation data shows that students are not concentrating during practical teaching, the algorithm can be used to optimize the content and organization of practical teaching, such as increasing the fun of practical projects and shortening the duration of each practical session. The adjusted course teaching plan data is then generated. Using a data update tool, and following the data update interface specifications of the course teaching platform, the adjusted course teaching plan data is synchronized to the course teaching platform, overwriting the original course data and ensuring that subsequent teaching is conducted according to the optimized plan.

[0086] Furthermore, step S1 includes the following steps:

[0087] Step S11: Obtain course data from the course teaching platform;

[0088] Step S12: Analyze the course adaptation groups based on the course data and generate course adaptation group data;

[0089] Step S13: Design the total course volume based on the target audience of the course and generate total course volume data;

[0090] Step S14: Based on the data of the target group for the course and the total number of courses, design the course teaching plan and generate course teaching plan data.

[0091] As an embodiment of the present invention, reference Figure 2 As shown, Figure 1 A detailed flowchart of step S1 is shown below. In this embodiment, step S1 includes the following steps:

[0092] Step S11: Obtain course data from the course teaching platform;

[0093] In this embodiment of the invention, a data acquisition device is used to directly connect to the database of the course teaching platform according to a pre-defined API interface protocol to extract course data. The acquired data includes basic course attributes, such as course name, subject category, and credit settings; course content details, including course outline and distribution of knowledge points in each chapter; teaching resource information, such as the format and duration of accompanying courseware and teaching videos; and student registration information, covering enrollment year, major category, completed course grades, and learning style assessment results (such as visual, auditory, and kinesthetic learning preferences). During the data extraction process, ETL (Extract-Transform-Load) technology is used to clean the raw data, remove duplicate records, and correct data fields with incorrect formats, such as standardizing date formats and subject category names, to ensure that the acquired course data is accurate, complete, and meets the requirements for subsequent analysis.

[0094] Step S12: Analyze the course adaptation groups based on the course data and generate course adaptation group data;

[0095] In this embodiment of the invention, clustering analysis algorithms, such as K-Means clustering, are used to process student information in the acquired course data. Based on students' completed course grades as the core criterion, combined with assessment results of their major and learning style, students are divided into different groups. For example, for computer programming courses, students with excellent grades in programming language courses, strong logical thinking, a science background, and a visual learning style are grouped into one group; these students possess strong self-learning and abstract understanding abilities. Students with average grades, a humanities background, and a kinesthetic learning style are grouped into another group; these students may require more practical operation and intuitive examples to aid their learning. Through multiple iterations of calculating cluster centers, the clustering results are continuously optimized, ultimately generating course-adaptive group data. This data records the number of students and characteristic labels (e.g., "high foundation - science - visual" and "medium foundation - humanities - kinesthetic") for each group, clarifying the differences in knowledge reserves and learning preferences among different groups, and providing accurate group profiles for subsequent course design.

[0096] Step S13: Design the total course volume based on the target audience of the course and generate total course volume data;

[0097] In this embodiment of the invention, based on the generated course adaptation group data, quantitative analysis and prediction models, such as linear regression models, are used to determine the total number of courses. First, the number of students in each adaptation group is counted to analyze the degree of course demand from different groups. Assuming a certain group has a large number of students and diverse needs for course knowledge, such as the "intermediate-humanities-kinesthetic" group, whose students have an average foundation and prefer practical application, they may need more courses with varying difficulty levels and rich practical components to meet their learning needs. By using historical data on similar groups' course selection numbers and credit completion rates, a regression equation is constructed to predict the number of courses required by this group. Simultaneously, the prediction results are adjusted considering the school's teaching resource capacity, such as the number of teachers, teaching space, and equipment. For example, if it is predicted that a certain group needs 10 courses, but the school's faculty can only undertake the teaching of 8 courses, then the total number of courses is determined to be 8.

[0098] Step S14: Based on the data of the target group for the course and the total number of courses, design the course teaching plan and generate course teaching plan data.

[0099] In this embodiment of the invention, course adaptation group data is combined with total course data, and a course planning algorithm is used for instructional planning and design. Differentiated teaching objectives, content, and methods are formulated for different course adaptation groups. For the "high-level foundation-science-visual learner" group, the teaching objective is to delve into cutting-edge knowledge, with content focusing on more challenging theoretical extensions and complex case analyses. The teaching method combines online self-study resources (such as in-depth academic papers and complex programming project examples) with offline group discussions. For the "intermediate foundation-humanities-kinesthetic learner" group, the teaching objective focuses on mastering basic knowledge and practical application, with content including extensive explanations of basic concepts and practical operation examples. The teaching method employs project-based learning, such as conducting actual programming project exercises and organizing field research. Based on the total course data, teaching time is allocated reasonably, determining the teaching duration and semester arrangement for each course. For example, more challenging courses are distributed across different semesters to avoid excessive concentration of student learning pressure. The final result is a detailed course teaching plan, which includes the syllabus, schedule, resource allocation list, and teaching method descriptions for each course, providing comprehensive guidance for the smooth implementation of the course.

[0100] Furthermore, step S2 includes the following steps:

[0101] Step S21: Obtain course teaching log data from the course teaching platform;

[0102] Step S22: Identify student learning behavior status during course instruction based on course teaching log data, and generate student learning behavior status data;

[0103] Step S23: Based on student learning behavior status data and course teaching plan data, conduct learning attention deviation assessment during the teaching of each course, and generate learning attention deviation data corresponding to each course teaching.

[0104] Step S24: Based on the learning attention deviation data corresponding to each course, conduct a course response evaluation on the corresponding course teaching within the course teaching plan data, and generate response evaluation data for each course teaching.

[0105] As an embodiment of the present invention, reference is made to... Figure 3 As shown, Figure 1 A detailed flowchart of step S2 is shown below. In this embodiment, step S2 includes the following steps:

[0106] Step S21: Obtain course teaching log data from the course teaching platform;

[0107] In this embodiment of the invention, a data extraction tool is used to establish a stable connection with the platform's log storage system, following the database communication protocol of the course teaching platform. The course teaching log data covers all student operation records during the course learning process, including the specific time of each login to the course platform, the duration of the session accurate to the second, the start and end times of accessing various teaching resources (such as teaching videos and electronic documents), the playback progress of videos, the number of pauses and pause times, the speaking time, the number of words spoken, and the number of replies to others' comments in online discussions, the time of submitting assignments, the number of words in the assignment content, the number of times the assignment was modified, and the answering time of online tests, the answering time for each question, and information on answer modification traces. During the data extraction process, data filtering technology is used to filter log data within a specific course teaching cycle according to a time range, eliminating invalid operation records, such as redundant records caused by repeated page refreshes or accidental operations, to ensure that the obtained course teaching log data is authentic and valid, accurately reflecting the student's learning process.

[0108] Step S22: Identify student learning behavior status during course instruction based on course teaching log data, and generate student learning behavior status data;

[0109] In this embodiment of the invention, a Hidden Markov Model (HMM) from pattern recognition algorithms is used to analyze the acquired course teaching log data. Taking student access to teaching resources as an example, if a student frequently switches between different teaching videos within a short period of time, with viewing progress generally below 30%, and does not participate in any discussions or submit assignments, the state transition probability calculation of the HMM can identify the student as being in a state of superficial browsing and inattentive learning behavior. Conversely, if a student watches teaching videos sequentially, with a total viewing time exceeding 80% of the total video duration, pauses multiple times to take notes, actively participates in discussions and replies to others' comments, and submits high-quality assignments on time, the model determines that the student is in a state of deep learning and focused engagement. By extracting features and modeling states of various learning behaviors from the log data, student learning behavior state data is generated for each student at different course teaching periods, including learning behavior state labels (such as "focused learning," "superficial browsing," and "passive learning"), behavior duration, and other information, comprehensively depicting the student's behavioral performance during the course learning process.

[0110] Step S23: Based on student learning behavior status data and course teaching plan data, conduct learning attention deviation assessment during the teaching of each course, and generate learning attention deviation data corresponding to each course teaching.

[0111] In this embodiment of the invention, firstly, dependency parsing in Natural Language Processing (NLP) is used to analyze the course syllabus and teaching content descriptions in the course teaching planning data, clarifying the logical relationships between various knowledge points and generating course content sequence data, clearly presenting the sequential order and hierarchical structure of the course teaching content. Next, a complexity assessment method based on knowledge graphs is used to match the knowledge points in the course content with nodes in the knowledge graph. Based on factors such as the number of associations and the depth of concepts involved, the complexity of each knowledge point is calculated, generating course content complexity data. For example, in a computer programming course, the "Data Structures and Algorithms" chapter involves multiple algorithm principles and data storage methods, and has strong associations with multiple nodes in the knowledge graph, thus being judged as high-complexity content. Then, based on the course content complexity data and the course content sequence data, logical reasoning algorithms are used to analyze the logical connection relationships between adjacent knowledge points, generating course content logical connection derivation data to determine whether the arrangement of the teaching content conforms to students' cognitive patterns. Then, through text analysis of online discussions and Q&A interactions recorded in the course teaching logs, combined with data derived from the logical connection of course content, the interaction in course teaching is identified, generating course teaching interaction data, such as frequently interacted knowledge points and interaction forms (questioning, answering, debate), etc. Simultaneously, based on the time points of changes in student learning behavior status in the student learning behavior status data, the learning cycle for each student in different course teaching stages is calculated, generating course learning cycle data, including the start and end times of the learning focus period, fatigue period, and adjustment period. Finally, using a comprehensive evaluation model, the interaction frequency and depth in the course teaching interaction data are combined with the learning stages in the course learning cycle data. The relationship between interaction and learning cycle under normal learning conditions is compared to assess the degree of student attention deviation during each course teaching period, generating learning attention deviation data corresponding to each course.

[0112] Step S24: Based on the learning attention deviation data corresponding to each course, conduct a course response evaluation on the corresponding course teaching within the course teaching plan data, and generate response evaluation data for each course teaching.

[0113] In this embodiment of the invention, a sliding window algorithm from time series analysis is used to process the learning attention deviation data corresponding to each course. Using a fixed time interval as a window, the algorithm monitors fluctuations in student learning behavior, generating behavioral fluctuation data and marking the time points and amplitudes of significant changes in student learning status. For example, if a student's attention deviation increases sharply within a short period during a specific time period of a course, exceeding a set fluctuation threshold, that period is determined to be a behavioral fluctuation period. Then, combining the course teaching log data with frequently accessed teaching resources and discussion topics during the behavioral fluctuation period, a text clustering algorithm is used to locate course content of interest to students, generating interest-based course content data. For example, students show strong interest in content related to "artificial intelligence application cases." Next, using an inverted index algorithm from information retrieval technology, based on the interest-based course content data, related content is retrieved from the course teaching plan data and teaching resource database, generating course-related content retrieval data, including relevant knowledge points, extended materials, and similar cases. Finally, using an evaluation index system, the data retrieved from course-related content was compared and analyzed with the course teaching plan data from multiple dimensions, including the relevance of teaching content, the adaptability of teaching methods, and the rationality of teaching progress. The course response evaluation was conducted on the corresponding course teaching within the course teaching plan data, generating response evaluation data for each course teaching, such as the matching score between teaching content and student interests, and the score of the effect of teaching methods on improving student attention. This comprehensively evaluated the response effect and optimization potential of the course teaching plan in actual teaching.

[0114] Furthermore, step S23 includes the following steps:

[0115] Step S231: Analyze the content sequence of the course based on the course teaching plan data to generate course content sequence data;

[0116] In this embodiment of the invention, dependency parsing techniques from the field of Natural Language Processing (NLP) are used to perform in-depth analysis of the course syllabus text and teaching content descriptions in the course teaching planning data. The course text is decomposed into word units, and the grammatical role of each word in the sentence and the dependency relationships between words are determined through dependency parsing algorithms, such as subject-verb, verb-object, and modifier-headword structural relationships. Taking the description "explaining the layered structure of the TCP / IP protocol stack and analyzing the functions of each layer and the data transmission process" in the computer network course syllabus as an example, dependency parsing clarifies that "explaining" and "the layered structure of the TCP / IP protocol stack" form a verb-object relationship, and "analyzing" and "the functions of each layer and the data transmission process" also form a verb-object relationship. At the same time, it is identified that "the layered structure of the TCP / IP protocol stack" is the prerequisite content for the analysis of "the functions of each layer and the data transmission process". By systematically analyzing the entire course teaching plan text, the sequential relationship between course knowledge points is identified, generating course content sequence data that includes information such as knowledge point name, chapter, prerequisite knowledge points, and subsequent knowledge points, clearly presenting the development path and logical sequence of the course teaching content.

[0117] Step S232: Analyze the complexity of the course content based on the order data of the course content, and generate course content complexity data;

[0118] In this embodiment of the invention, a complexity assessment method based on knowledge graphs is employed to match and map knowledge points in the sequential data of course content with a pre-constructed professional knowledge graph. Each node in the knowledge graph represents a knowledge point, and the edges between nodes represent the relationships between knowledge points, such as inclusion, causal, and parallel relationships. Taking the knowledge point "Multivariable Calculus" in the Advanced Mathematics course as an example, this knowledge point is associated with multiple nodes in the knowledge graph, such as "Single-Variable Calculus," "Partial Derivatives," and "Multiple Integrals," and involves multi-dimensional conceptual expansion and complex operational rules. Based on factors such as the number of associated nodes in the knowledge graph (e.g., associated with more than 5 other knowledge points), the weight of the associated edges (e.g., higher weight associated with core foundational knowledge points), and the conceptual depth involved in the knowledge point itself (e.g., including multiple abstract concepts such as limits, continuity, and differentiability), a weighted calculation model is used to calculate the corresponding complexity value for each knowledge point. The complexity values ​​of all knowledge points are summarized and organized to generate course content complexity data that includes knowledge point name, complexity score, and complexity dimension description (such as conceptual complexity, computational complexity, and application complexity), thereby quantitatively evaluating the distribution of difficulty levels of the course content.

[0119] Step S233: Based on the complexity data of the course content, deduce the logical connection of the course content sequence data to generate logical connection deduction data of the course content;

[0120] In this embodiment of the invention, logical reasoning algorithms are used to deduce the logical connection between course content based on the sequence data and complexity data of the course content. First, the sequential relationship between adjacent knowledge points is determined according to the sequence data of the course content. Then, the transition in difficulty between adjacent knowledge points is judged by combining the complexity data of the course content. For example, in a programming language course, if the knowledge point of "variable declaration and data type" (low complexity) is directly followed by the knowledge point of "recursive algorithm implementation" (high complexity), the logical reasoning algorithm analysis reveals that there is a lack of basic knowledge points such as "function definition and call" as a transition, indicating a logical gap in the connection. By analyzing the sequence of the entire course content segment by segment, and using the logical rules set by the rule engine (such as the increment of complexity not exceeding a threshold, and the prerequisite knowledge points must cover the foundation required by the subsequent knowledge points), logical connection deduction data of the course content is generated, which includes the knowledge point name, the current logical state (reasonable / unreasonable), the description of the logical gap, and the suggested connection method. This clarifies the rationality of the logical relationship in the arrangement of course teaching content and the direction for optimization.

[0121] Step S234: Based on the logical connection of the course content, derive data to identify the interaction in the course teaching and generate course teaching interaction data;

[0122] In this embodiment of the invention, natural language processing is performed on online discussion records and Q&A interaction texts in the course teaching log data. Combined with the logical connection derivation data of the course content, the interaction situation of course teaching is identified. NLP technologies such as text segmentation, part-of-speech tagging, and named entity recognition are used to extract keywords and key information from the discussion and interaction texts, such as the names of the knowledge points involved, the types of questions raised by students (conceptual understanding, application practice), and the answers provided by teachers or other students. The extracted key information is matched with the knowledge points in the logical connection derivation data of the course content. If a large number of questions about logical gaps in a certain knowledge point appear in the discussion, such as students frequently asking "how to combine recursive algorithms with previously learned function calls," then the logical connection part corresponding to that knowledge point is determined to be a high-frequency area of ​​teaching interaction. By analyzing all teaching log interaction texts, the number of interactions, interaction types (questioning, answering, debating, etc.), and number of participants for each knowledge point are statistically analyzed. This generates course teaching interaction data containing knowledge point names, interaction frequencies, high-frequency interaction periods, and interaction type distribution, intuitively presenting the hot spots and focal points of teacher-student interaction during the course teaching process.

[0123] Step S235: Calculate the course learning cycle based on student learning behavior status data and generate course learning cycle data;

[0124] In this embodiment of the invention, student learning behavior state data is used to monitor and analyze changes in student behavior state during the course learning process in chronological order. When a student's learning behavior state changes from "focused learning" to "superficial browsing," then to "adjusting learning strategies," and finally back to "focused learning," this series of state changes is considered a complete learning cycle. Using time series analysis, the start and end times of each learning behavior state are recorded, and the time interval between adjacent states is calculated. For example, if a student focuses on watching an instructional video for 30 minutes (entering the "focused learning" state), then frequently switches videos (entering the "superficial browsing" state), stops switching after 10 minutes and begins organizing notes (entering the "adjusting learning strategies" state), and then focuses on watching videos again after 15 minutes (returning to the "focused learning" state), then the duration of this learning cycle is 55 minutes. By analyzing all changes in student learning behavior state throughout the entire course learning process, course learning cycle data is generated for each student, including learning cycle number, cycle start time, cycle end time, duration of each state, and order of state changes, accurately depicting the time distribution and state transition patterns of students during course learning.

[0125] Step S236: Based on the course teaching interaction data and course learning cycle data, conduct an assessment of the learning attention deviation during the teaching of each course, and generate learning attention deviation data corresponding to each course.

[0126] In this embodiment of the invention, a comprehensive evaluation model is used to combine data on course teaching interaction with data on the course learning cycle to assess learning attention deviation. First, the frequency and depth of interaction in the course teaching interaction data are analyzed. If, within a certain learning cycle, students exhibit low interaction frequency and shallow interaction depth at key knowledge points—for example, only asking simple, repetitive questions without delving into the deeper meaning of the knowledge points—it indicates that the student may have insufficient attention to the key content during this cycle. Simultaneously, changes in learning status within the course learning cycle data are considered. For instance, if students spend excessive time in a "shallow browsing" state when learning complex knowledge points, far exceeding the average duration of this state within a normal learning cycle, this further corroborates the attention deviation. By establishing an evaluation index system, including interaction participation indicators (percentage of interaction times, percentage of effective speech words) and learning state stability indicators (percentage of focused time, duration of non-focused time), and using a weighted summation algorithm, quantitative evaluation is conducted for each course teaching period. This generates learning attention deviation data for each course, including information such as course name, teaching period, attention deviation score, and deviation reason analysis (e.g., insufficient interaction, fluctuations in learning state). This accurately determines the specific situation and severity of students' attention deviation during course learning.

[0127] Furthermore, step S236 includes the following steps:

[0128] Based on the data on course teaching interaction, the interaction patterns of each course teaching are identified, and corresponding interaction pattern data for each course teaching is generated.

[0129] In this embodiment of the invention, a sequence pattern mining algorithm is used to perform in-depth processing of course teaching interaction data. This data includes online discussion records, questions and answers during Q&A sessions, and interactive information from collaborative learning. Each interactive behavior is encoded in chronological order to form a behavior sequence; for example, asking a question is encoded as "A", answering a question as "B", and initiating a discussion as "C". Taking the interaction record of a programming course as an example, a student first initiates a discussion about function calls (C), then other students provide answers (B), and the teacher supplements key knowledge points (B), forming the sequence "CBB". The sequence pattern mining algorithm searches for frequently occurring subsequences in a large number of interaction behavior sequences. If the "CBB" sequence appears frequently in the interaction records of multiple lessons, and the frequency exceeds a set threshold, then the sequence is determined to be an interaction pattern, namely the "student initiates discussion - student answers - teacher supplements" pattern. Simultaneously, information such as the roles of participants and the categories of knowledge points involved in the interaction content are analyzed within the interaction pattern. By analyzing the data on all course teaching interactions, interactive mode data corresponding to each course teaching is generated, which includes information such as the interaction mode name, mode sequence, frequency of occurrence, knowledge points involved, and participating roles.

[0130] Preferably, the learning depth value-added analysis is performed on each course teaching session according to the course learning cycle to generate learning depth value-added data corresponding to each course teaching session.

[0131] In this embodiment of the invention, a knowledge mastery assessment model is employed to perform a learning depth value-added analysis based on course learning cycle data. Within each course learning cycle, data on student participation in learning activities is extracted, such as the duration of watching instructional videos, the accuracy rate of completing assignments, test scores, and the quality of written learning summaries. This data is mapped to a pre-defined knowledge mastery assessment index system, which covers six levels: knowledge memorization, understanding, application, analysis, evaluation, and creation. For example, watching instructional videos corresponds to the knowledge memorization level, completing assignments reflects the knowledge application level, and participating in tests reflects the knowledge understanding and analysis levels. Taking a one-week course learning cycle as an example, a student's test score at the beginning of the cycle is 60 points. After one week of learning, the student's test score is 80 points, while the assignment accuracy rate increases from 60% to 85%, and the analysis of knowledge points in the written learning summary becomes more in-depth. By calculating the score difference of each assessment index before and after the cycle, and combining the weights of each index (e.g., test score weight 0.4, assignment accuracy weight 0.3, learning summary quality weight 0.3), a weighted summation formula is used to calculate the learning depth value-added score within that learning cycle. The same calculation is performed on all learning cycles of each course to generate learning depth value-added data for each course, which includes information such as course name, learning cycle number, initial knowledge mastery score, final knowledge mastery score, and learning depth value-added score. This quantitatively reflects the improvement of students' learning depth in different course learning cycles.

[0132] Preferably, interactive frequency analysis is performed on the teaching frequency of each course based on the learning depth value-added data and the interactive mode data corresponding to each course, so as to generate interactive frequency data corresponding to each course.

[0133] In this embodiment of the invention, an interactive frequency analysis model is constructed to combine the learning depth enhancement data and interaction mode data corresponding to each course teaching for analysis. First, an association matrix between interaction modes and learning depth enhancement is established. The rows of the matrix represent different interaction modes, and the columns represent different learning depth enhancement intervals (such as enhancement scores of 0-10, 11-20, etc.). Taking a mathematics course as an example, if the "teacher explanation-student question-group discussion" interaction mode appears 15 times in the course teaching period with a learning depth enhancement score of 11-20, and appears 5 times in the 0-10 interval, these data are filled into the corresponding positions in the association matrix. Then, the frequency ratio of each interaction mode in different learning depth enhancement intervals is calculated to analyze the synergistic relationship between different interaction modes and learning depth enhancement. If the frequency of a certain interaction mode in the high learning depth enhancement interval is significantly higher than in other intervals, it indicates that the interaction mode has a strong frequency correlation with the improvement of learning depth; conversely, if the distribution is relatively even across intervals, the frequency correlation is weak. By analyzing and calculating the correlation matrix, interactive frequency data corresponding to each course teaching is generated, including information such as course name, interaction mode name, frequency score, and frequency interval, thus clarifying the degree of synergistic matching between interactive modes and learning depth improvement in different course teaching.

[0134] Preferably, the learning attention deviation during the teaching of each course is evaluated based on the interaction frequency data corresponding to the teaching of each course, and the learning attention deviation data corresponding to the teaching of each course is generated.

[0135] In this embodiment of the invention, an attention deviation assessment model is used to assess learning attention deviation based on the interaction frequency data corresponding to each course. A standard threshold for interaction frequency is set. If the interaction frequency score during a certain course teaching period is lower than the standard threshold, and the corresponding learning depth enhancement score is also lower than the average level, combined with the complexity of the course content, it is determined that the student may have learning attention deviation during this period. For example, in a physics course, the interaction frequency score for the current teaching period is 40 points (the standard threshold is 60 points), and the learning depth enhancement score is only 5 points (the average level is 12 points). Furthermore, this period covers complex electromagnetic theory, therefore, it is determined that the student has moderate learning attention deviation during this period. By comprehensively analyzing the interaction frequency data, learning depth enhancement data, and course content complexity data of each course teaching session, the decision tree algorithm is used to construct evaluation rules and generate learning attention deviation data for each course teaching session, including information such as course name, teaching session, attention deviation level (mild, moderate, severe), and deviation reason analysis (such as interaction mode not matching learning, content difficulty too high). This allows for accurate judgment of students' attention concentration status and deviation reasons at different times during the course learning process.

[0136] Furthermore, step S24 includes the following steps:

[0137] Step S241: Monitor student learning behavior fluctuations based on the learning attention deviation data corresponding to each course, and generate behavior fluctuation data;

[0138] In this embodiment of the invention, the moving average algorithm in time series analysis is used to process the learning attention deviation data corresponding to each course. The learning attention deviation data includes information such as course name, teaching time period, and level of attention deviation. A 10-minute time window is used to perform a sliding calculation on the attention deviation data for each course. For example, in the teaching process of a history course, minutes 1-10, 2-11, 3-12, and so on are successively used as time windows, and the average level of attention deviation within each window is calculated. A normal fluctuation range threshold is set. If the average deviation within a certain time window exceeds the upper limit of normal fluctuation, or if multiple consecutive windows show a continuous upward trend in the deviation, then behavioral fluctuation is determined to have occurred during that period. For example, if the average deviation reaches a severe level within the 30-40 minute time window, and the deviation also increases in the previous two windows, then this period is marked as a behavioral fluctuation period. The system comprehensively analyzes the learning attention deviation data of all courses to generate behavioral fluctuation data containing information such as course name, period of behavioral fluctuation, start time of fluctuation, end time of fluctuation, and degree of fluctuation (mild, moderate, severe), thereby enabling precise monitoring of students' learning behavior fluctuations.

[0139] Step S242: Locate the content of interest courses based on behavioral fluctuation data and generate interest course content data;

[0140] In this embodiment of the invention, text mining and clustering algorithms are used to locate interest-based course content by combining behavioral fluctuation data and course teaching log data. In the course teaching log data corresponding to the behavioral fluctuation period, student operation records are extracted, such as the names of accessed teaching resources, online discussion content, and homework submissions. These texts are then segmented and stop words are removed. For example, the phrase "I think the tactical application of ancient warfare is very interesting" is segmented into "think," "ancient warfare," "tactical application," and "interesting," and stop words such as "I" and "think" are removed. The TF-IDF (Term Frequency-Inverse Document Frequency) algorithm is used to calculate the weight of each word; a higher weight indicates greater importance of the word in the text. Words with higher weights are clustered. If words such as "ancient warfare," "tactical application," and "unit coordination" frequently appear in the texts of multiple behavioral fluctuation periods and are clustered together, and the total weight of this category of words in the relevant texts exceeds a set threshold, then "ancient warfare tactics" is determined to be course content of interest to students. By analyzing the log text corresponding to all behavioral fluctuation data, interest course content data is generated, which includes information such as course name, keywords of interest course content, number of related texts, and total keyword weight, so as to accurately locate the direction of course content that students are interested in.

[0141] Step S243: Based on the interest course content data, perform course-related content retrieval and identification to generate course-related content retrieval data;

[0142] In this embodiment of the invention, based on interest-based course content data, an inverted index and semantic matching algorithm are used to retrieve and identify course-related content. An inverted index is pre-established for the course syllabus, teaching materials, and extended reading materials in the course teaching planning data, mapping each word to documents containing that word. Taking the interest-based course content keyword "ancient warfare tactics" as an example, the inverted index quickly finds chapters in the syllabus mentioning "ancient warfare tactics," related courseware and video documents in the teaching materials, and related books and papers in the extended reading materials. Simultaneously, a semantic vector model from natural language processing, such as Word2Vec, is used to convert the interest-based course content keywords and the retrieved document content into vector form, and the cosine similarity between the vectors is calculated. If the cosine similarity between a document and the keyword "ancient warfare tactics" exceeds a set threshold, such as 0.7, the document is included in the course-related content retrieval data. Finally, course-related content retrieval data is generated, containing information such as the course name, the name of the retrieved related content document, the document type (syllabus chapter, courseware, literature, etc.), and the similarity score, comprehensively acquiring course content resources related to students' interests.

[0143] Step S244: Based on the data retrieved from the course-related content, conduct a course response evaluation on the corresponding course teaching within the course teaching plan data, and generate response evaluation data for each course teaching.

[0144] In this embodiment of the invention, a course response evaluation index system is constructed to evaluate the corresponding course teaching within the course teaching plan data based on the retrieved data of course-related content. The evaluation index system includes three dimensions: content coverage, resource suitability, and teaching method matching. Content coverage evaluates the extent to which the course teaching plan covers course content that students are interested in. If the course outline only briefly mentions "ancient war tactics" without in-depth explanation, the content coverage score will be low. Resource suitability examines the degree to which the retrieved relevant content resources match the students' learning needs, such as whether the provided courseware clearly demonstrates tactical details. If the courseware is too brief, the resource suitability score will be low. Teaching method matching analyzes whether the current teaching method is suitable for explaining content that students are interested in. If a single lecture method is used to explain complex tactics without combining case analysis or simulation exercises, the teaching method matching score will be low. Weights are assigned to each evaluation index, such as content coverage weight 0.4, resource suitability weight 0.3, and teaching method matching weight 0.3. The response evaluation score of each course teaching is calculated by weighted summation. The evaluation is graded based on scores, such as 90-100 points for excellent and 80-89 points for good. The evaluation data is generated for each course, including course name, evaluation score, evaluation grade, details of scores for each indicator, and suggestions for improvement. This comprehensively assesses how well the course teaching plan responds to students' interests and learning needs, and provides a basis for course optimization.

[0145] Furthermore, step S3 includes the following steps:

[0146] Step S31: Obtain historical teaching information data;

[0147] In this embodiment of the invention, historical teaching information data is obtained from the school's teaching management database via a data interface. This database stores course teaching-related data for multiple academic years and semesters, including course exam score records (covering the total score of each exam, detailed student scores for each question, and score distribution range), student evaluations (including sub-scores and comprehensive scores for teaching content satisfaction, teaching method satisfaction, and teacher attitude satisfaction), teacher teaching summary reports (recording the teaching process, teaching difficulties, and improvement measures in text form), and course completion status (number of graduated students and analysis of reasons for non-graduation). During data acquisition, data filtering technology is employed to precisely filter data according to course name, teaching semester, and other conditions, ensuring that the acquired historical teaching information data corresponds to the course to be analyzed. Simultaneously, a data verification mechanism is used to check the completeness and accuracy of the acquired data, such as verifying whether the total score in the grade record matches the sum of the scores for each question, correcting data items with format errors or logical contradictions, ensuring the authenticity and reliability of the historical teaching information data, and providing a solid data foundation for subsequent analysis.

[0148] Step S32: Plot the course's teaching history stage outcome curves based on historical teaching information data, and generate teaching history stage outcome curve data;

[0149] In this embodiment of the invention, a curve plotting algorithm from data visualization technology is used to process the acquired historical teaching information data. Using the semester as the time dimension, quantitative indicators such as the average exam score and student evaluation score for each semester are used as data points. For example, the average exam score for a computer programming course in the first semester of the 2020-2021 academic year was 75, and the student evaluation score was 80. (First semester of 2020-2021, 75) and (First semester of 2020-2021, 80) are used as two data points. A cubic spline interpolation algorithm is used to generate a smooth curve between adjacent data points, ensuring that the curve accurately reflects the trend of data change while avoiding abrupt changes and sharp transitions. Multiple curves, such as the average exam score curve and the student evaluation score curve, are integrated into the same coordinate system to generate historical teaching achievement curve data that includes a time axis, various indicator curves, and legends. This curve data allows for a direct observation of changes in teaching outcomes at different historical stages, such as whether grades show an upward or downward trend, and whether student satisfaction fluctuates.

[0150] Step S33: Assess students' mastery of course knowledge based on the learning attention deviation data corresponding to each course, and generate course knowledge mastery data;

[0151] In this embodiment of the invention, a knowledge mastery assessment model is constructed to assess students' knowledge mastery based on learning attention deviation data corresponding to each course. The learning attention deviation data includes the degree of attention deviation (mild, moderate, severe) and duration of students at different times during course instruction. The model sets different weights for the impact of different degrees of attention deviation on knowledge mastery, such as a weight of 0.8 for severe deviation, 0.5 for moderate deviation, and 0.2 for mild deviation. Simultaneously, the learning weight of each knowledge point is determined by combining the distribution of knowledge points and teaching duration in the course instruction plan data. Taking the knowledge point "limits of functions" in a mathematics course as an example, this knowledge point has a teaching duration of 4 class periods, and its importance in the course knowledge system is set to a learning weight of 0.1. For example, if a student is in a state of severe attention deviation for 2 class periods during the 4 class periods of explaining the knowledge point "limits of functions," according to the formula M = (1-t) / ( ... a ×W a -t b ×W b -t c ×W c )×W dWhere M represents the level of knowledge mastery, and t a t b t c These represent the durations of severe deviation, moderate deviation, and mild deviation, respectively. a W b W c These represent the weights for severe deviation, moderate deviation, and slight deviation, respectively. W d Assign learning weights to knowledge points. Calculate the student's mastery of the "function limits" knowledge point. Summarize and calculate the mastery of each student's knowledge points across all knowledge points in the course to obtain each student's total course knowledge mastery score. Then, through statistical analysis (such as calculating the mean score and standard deviation), generate course knowledge mastery data containing information such as course name, number of students, average knowledge mastery score, and knowledge mastery score distribution intervals to comprehensively assess students' overall mastery of the course knowledge.

[0152] Step S34: Calculate the course teaching outcome deviation based on course knowledge mastery data and course teaching outcome curve data at historical teaching stages, and generate course teaching outcome deviation data;

[0153] In this embodiment of the invention, a deviation calculation model is used to calculate the deviation of course teaching outcomes based on historical teaching outcome curve data and course knowledge mastery data. The average knowledge mastery score in the current course knowledge mastery data is compared with the average exam score of the corresponding semester in the historical teaching outcome curve data. For example, if the average knowledge mastery score for an English course in the current semester is 70 points, while the historical teaching outcome curve shows an average exam score of 78 points for that semester, the deviation value is calculated as -8 points using the deviation value formula. Simultaneously, considering other teaching outcome indicators such as student evaluation scores, a weighted average method is used to calculate the comprehensive deviation value, assigning a weight of 0.6 to exam scores and a weight of 0.4 to student evaluation scores. For example, if the deviation value of the current semester's student evaluation score compared to the same period in history is -5 points, then the comprehensive deviation value = -8 × 0.6 + (-5) × 0.4 = -6.8 points. Generate course teaching outcome deviation data containing information such as course name, deviation value, deviation details of each indicator, and deviation calculation weight, and clarify the degree and direction of the difference between the current course teaching outcome and the historical teaching outcome.

[0154] Step S35: Analyze the course teaching outcomes based on the deviation data of course teaching outcomes and generate teaching outcome data.

[0155] In this embodiment of the invention, based on course teaching outcome deviation data, causal analysis and trend prediction methods are used to analyze course teaching outcomes. For courses with negative deviation values, in-depth analysis of historical teaching outcome curve data and course knowledge mastery data is conducted to identify the causes of the deviation. If it is found that in a certain physics course, when explaining the "Electromagnetism" chapter, students' knowledge mastery is significantly lower than in the same period in previous years, combined with learning attention deviation data, it is found that the duration of moderate and severe attention deviation during the teaching of this chapter has significantly increased. Further analysis of the teaching history summary report reveals that the teaching method for this chapter has not been optimized compared to the past, making it difficult to meet students' needs for understanding complex electromagnetic concepts. Thus, it is determined that the unsuitable teaching method is one of the reasons for the teaching outcome deviation. At the same time, by analyzing the trend of historical teaching outcome curves, the future trend of teaching outcome changes is predicted. If the teaching outcome deviation of a course shows a trend of increasing year by year, and the main reason is the lagging update of teaching content, targeted improvement suggestions are proposed, such as accelerating the update frequency of teaching content and introducing practical cases to enrich the teaching content. Finally, teaching outcome data containing information such as course name, deviation cause analysis, trend prediction, and improvement suggestions are generated, providing a comprehensive analytical basis for course teaching improvement and quality enhancement.

[0156] Furthermore, step S4 includes the following steps:

[0157] Step S41: Analyze the focus of each course based on the response evaluation data corresponding to the teaching of each course, and generate course focus data;

[0158] In this embodiment of the invention, text mining and topic modeling algorithms are used to process the response evaluation data corresponding to each course. The response evaluation data includes textual information such as course name, evaluation score, details of scores for each indicator, and improvement suggestions. Taking the "University Physics" course as an example, the improvement suggestions in its response evaluation data mention "strengthening the mechanics experiment demonstration" and "supplementing cutting-edge physics research cases." Using the Latent Dirichlet Allocation (LDA) topic model, topics are extracted from the text in the response evaluation data of all courses, with five topics set for extraction. After segmenting the text and removing stop words, the model assigns each word to a different topic through probability calculation. For example, words such as "mechanics experiment" and "experiment demonstration" are classified into the "experiment teaching optimization" topic, and words such as "cutting-edge research" and "physics cases" are classified into the "teaching content expansion" topic. The frequency and weight of each topic in the response evaluation data of each course are statistically analyzed to generate course attention content data containing information such as course name, high-frequency attention topics (such as "experiment teaching optimization" and "teaching content expansion"), topic frequency, and topic keywords. This data provides a clear understanding of the key areas of focus for students and teaching assessments in different courses.

[0159] Step S42: Assess students' understanding of course knowledge based on teaching outcome data and generate course knowledge understanding data;

[0160] In this embodiment of the invention, a knowledge comprehension assessment system is constructed to evaluate students' course knowledge comprehension based on teaching outcome data. The teaching outcome data includes information such as course name, analysis of deviation reasons, and students' course knowledge mastery scores. The assessment system sets multiple dimensions, including knowledge memorization (corresponding to simple concept repetition, formula memorization, etc.), knowledge comprehension (corresponding to principle explanation, concept clarification, etc.), and knowledge application (corresponding to problem solving, case analysis, etc.), and assigns weights to each dimension, such as 0.2 for knowledge memorization, 0.3 for knowledge comprehension, and 0.5 for knowledge application. Taking the "Data Structures" course as an example, based on students' scores on short-answer questions (assessing knowledge comprehension), programming questions (assessing knowledge application), and fill-in-the-blank questions (assessing knowledge memorization) in the exam, combined with the percentage of marks for each question type in the exam, each student's score in each dimension is calculated. Assuming a student scores 8 points (out of 10) in the knowledge memorization dimension, 7 points (out of 10) in the knowledge comprehension dimension, and 6 points (out of 10) in the knowledge application dimension, the score is calculated using formula S... 总 =S 记 ×W 记 +S 理 ×W 理 +S 用 ×W 用 S 总 W represents the total score for knowledge comprehension. 记 W 理 W 用 These are the weights for the knowledge memory dimension, the knowledge comprehension dimension, and the knowledge application dimension, respectively. 记 S 理 S 用 The scores are calculated as follows: knowledge memorization score, knowledge comprehension score, and knowledge application score, representing the student's total knowledge comprehension level. Statistical analysis is then performed on all student scores, calculating the mean score, standard deviation, etc., to generate course knowledge comprehension data containing information such as course name, number of students, average knowledge comprehension score, and score distribution range, comprehensively assessing the depth and breadth of students' understanding of the course knowledge.

[0161] Step S43: Based on the course focus content data and course knowledge comprehension data, perform course teaching matching deviation diagnosis and generate course teaching matching deviation data;

[0162] In this embodiment of the invention, a teaching matching deviation diagnostic model is established, which diagnoses the course teaching matching deviation based on course focus content data and course knowledge comprehension data. The model compares high-frequency focus topics in the course focus content data with the teaching content and methods in the course teaching plan data, while also considering students' knowledge mastery reflected in the course knowledge comprehension data. For example, in the "Marketing" course, the course focus content data shows that "actual marketing case analysis" is a high-frequency focus topic, but the course teaching plan only allocates a small amount of time for case teaching, and students' average score in the knowledge application dimension is low, indicating a mismatch between the teaching content and students' focus and knowledge comprehension needs. The diagnostic model sets matching degree evaluation indicators, including content matching degree (the degree to which the teaching content covers the focus topics), method matching degree (whether the teaching methods are suitable for explaining the focus topics), and effect matching degree (the degree to which students' knowledge comprehension aligns with the learning objectives of the focus topics), and assigns weights to each indicator, such as a weight of 0.4 for content matching degree, 0.3 for method matching degree, and 0.3 for effect matching degree. By calculating the scores of each indicator and the weighted total score, the course teaching matching deviation value is obtained, and course teaching matching deviation data containing information such as course name, matching deviation value, details of matching status of each indicator, and preliminary analysis of the reasons for deviation are generated, so as to clarify the deviation between course teaching content, methods and student needs.

[0163] Step S44: Evaluate the course teaching quality based on the course teaching matching deviation data and generate course teaching quality data.

[0164] In this embodiment of the invention, a course teaching quality evaluation model is constructed, which evaluates course teaching quality based on course teaching matching deviation data. The evaluation model comprehensively considers factors such as teaching matching deviation value, the completeness of course teaching plan (whether elements such as teaching objectives, content, and methods are complete), and historical comparison of teaching outcomes (comparison with historical teaching quality data). An evaluation index system is set, where the teaching matching deviation value has a weight of 0.5, the completeness of teaching plan has a weight of 0.3, and the historical comparison of teaching outcomes has a weight of 0.2. Taking the course "Advanced Mathematics" as an example, if its course teaching matching deviation value is -7 points (a large deviation), the score for the completeness of teaching plan is 8 points (out of 10), which is a decrease compared to historical teaching quality, and the score for historical comparison of teaching outcomes is 6 points (out of 10). This is calculated using the formula S = X × W. X +Y×W Y +Z×W Z Where S is the total score for course teaching quality, X is the teaching matching deviation value, Y is the score for the completeness of the teaching plan, Z is the score for historical comparison of teaching outcomes, and W is the score for the completeness of the teaching plan. X W Y W ZThe total teaching quality score for the course is calculated by weighting the teaching matching deviation value, the completeness of the teaching plan, and the historical comparison of teaching outcomes. Quality levels are then assigned based on pre-defined score ranges: 0-3 is considered satisfactory, -3-0 is considered needing improvement, and below -3 is considered unsatisfactory. Course teaching quality data is generated, including the course name, total teaching quality score, quality level, detailed scores for each indicator, and improvement suggestions. This comprehensive and objective assessment of the quality level of the course teaching plan provides a clear direction for course improvement.

[0165] Furthermore, step S5 includes the following steps:

[0166] Step S51: Analyze the fit between students and teachers' course pace based on the course teaching quality data, and generate course pace fit data;

[0167] In this embodiment of the invention, time series comparative analysis and correlation calculation methods are used to analyze the alignment of students' and teachers' course pace based on course teaching quality data. Course teaching quality data includes information such as course name, total teaching quality score, and details of scores for each indicator. It also incorporates teachers' teaching progress records (such as the start and end times of each knowledge point's explanation and the scheduling of classroom exercises) and students' learning behavior time series data (such as access time to learning resources, assignment submission time, and online test participation time) from course teaching log data. Taking the "English Writing" course as an example, the teacher sets a weekly teaching plan to explain and practice one writing topic. The actual time span of the teacher's explanation of each topic is extracted from the course teaching log data. Assuming the explanation of the "Argumentative Writing" topic starts on Wednesday of week 1 and ends on Monday of week 2, taking 5 days, the distribution of students' assignment submission times for this topic is statistically analyzed from student learning behavior data. If most students submit their assignments after Friday of week 2, it indicates that the students' pace of learning this topic lags behind the teacher's teaching pace. By calculating the Pearson correlation coefficient between the teacher's teaching progress time series and the student's learning behavior time series, the degree of fit between their rhythms is quantified. The closer the coefficient is to 1, the higher the fit; the closer it is to -1, the greater the difference. The teacher's teaching progress and student's learning behavior time series for each course are analyzed one by one, generating course rhythm fit data that includes information such as course name, fit coefficient, description of teacher's teaching rhythm, description of student's learning rhythm, and rhythm difference analysis, clearly showing the rhythm matching between teachers and students in the course learning process.

[0168] Step S52: Analyze the relevant interest courses based on the learning attention deviation data corresponding to each course, and generate relevant interest course data;

[0169] In this embodiment of the invention, an association rule mining algorithm is used to analyze related interest courses based on the learning attention deviation data corresponding to each course. The learning attention deviation data records students' attention states at different teaching periods of the course, combined with students' cross-course learning behavior data on the course teaching platform (such as records of accessing other course resources and participation in cross-course discussions). Taking student A as an example, during the learning process of the "Computer Networks" course, the degree of learning attention deviation was relatively high in weeks 3-5, but during this period, they frequently accessed teaching videos and materials for the "Cybersecurity" course and actively participated in related discussions. By mining student learning behavior data using the Apriori algorithm, setting a support threshold of 0.2 and a confidence threshold of 0.7, the analysis found that when students exhibit attention deviation in a certain course, there is a high probability that they will learn other courses with knowledge relevance. For example, 75% of students exhibiting attention deviation in the "Computer Networks" course will access resources for the "Cybersecurity" course, thus determining that the "Cybersecurity" course is a related interest course for the "Computer Networks" course. A comprehensive analysis of learning attention deviation data and cross-course learning behavior data for all courses was conducted to generate related interest course data, which includes course name, a list of related interest courses, association rules (e.g., "Computer Networks - Network Security" with a support of 0.25 and a confidence of 0.8), and association strength, thus clarifying the relationship between courses based on students' interests and learning behaviors.

[0170] Step S53: Based on the course rhythm fit data and related interest course data, perform course teaching collaboration matching to generate course teaching collaboration data;

[0171] In this embodiment of the invention, a collaborative matching model is constructed to perform collaborative matching of course teaching based on course rhythm fit data and related interest course data. The collaborative matching model considers factors such as course rhythm fit, course content relevance, and student interest preferences, and sets weights for each factor, such as a weight of 0.4 for course rhythm fit, a weight of 0.3 for course content relevance, and a weight of 0.3 for student interest preferences. Taking the "Marketing" course and the "Consumer Behavior" course as examples, the course rhythm fit data shows that the teacher-student rhythm fit coefficients for both courses are relatively high, at 0.8 and 0.7 respectively; the related interest course data indicates that 60% of students learning the "Marketing" course will also pay attention to the "Consumer Behavior" course; at the same time, through content analysis of the course teaching planning data, it was found that there is a 30% overlap between the two courses in terms of consumer demand analysis, market research methods, etc. According to the collaborative matching formula S=C×w1+N×w2+P×w3, where S is the collaborative matching degree, C is the course rhythm fit coefficient, N is the content overlap degree, P is the student interest association ratio, and w1, w2, and w3 are the weights of course rhythm fit, content overlap, and student interest association ratio, respectively. The collaborative matching degree between the "Marketing" course and the "Consumer Behavior" course is calculated to be 0.8×0.4+0.3×0.3+0.6×0.3=0.65. Collaborative matching degree calculations are performed for each pair of courses, generating course teaching collaboration data containing information such as course A name, course B name, collaborative matching degree, and details of matching factors (rhythm fit score, content relevance score, and interest association score). This provides a basis for course teaching resource integration and collaborative teaching arrangements.

[0172] Step S54: Based on the course teaching collaboration data, perform structural adjustment processing on the course teaching planning data to generate structurally adjusted course teaching planning data and update the course data on the teaching interaction platform.

[0173] In this embodiment of the invention, a structural optimization algorithm is used to adjust the structure of course teaching planning data based on collaborative course teaching data. The algorithm, grounded in collaborative course teaching data and incorporating course teaching quality data and student learning needs, adjusts elements such as teaching objectives, content, sequence, and methods within the course teaching plan. For example, for courses with high collaborative matching, such as "Data Structures" and "Algorithm Design," the overlapping teaching content of "sorting algorithms" is integrated to avoid repetitive explanations. Simultaneously, based on students' learning pace, the teaching time for linked list structures in "Data Structures" is moved forward to align with the teaching time for linked list-based algorithm implementations in "Algorithm Design." After adjustment, a structurally adjusted course teaching plan is generated, containing information such as the course name, adjusted teaching objectives, adjusted teaching content arrangement, adjusted teaching schedule, and adjusted teaching method descriptions. Through a data synchronization interface, the structurally adjusted course teaching plan data is transmitted to the teaching interaction platform, overwriting the original course data and updating the platform's course data. This ensures that subsequent teaching is conducted according to the optimized course structure, improving the systematic nature of course teaching and student learning outcomes.

[0174] Furthermore, the present invention also provides a big data-based course quality supervision and analysis system for executing the big data-based course quality supervision and analysis method described above. This big data-based course quality supervision and analysis system includes:

[0175] The course planning module is used to acquire course data from the course teaching platform; design course teaching plans based on the course data; and generate course teaching plan data.

[0176] The course teaching attention response analysis module is used to acquire course teaching log data from the course teaching platform; based on the course teaching log data and course teaching plan data, it evaluates the learning attention deviation during each course teaching and generates learning attention deviation data corresponding to each course teaching; based on the learning attention deviation data corresponding to each course teaching, it evaluates the course teaching response within the course teaching plan data and generates response evaluation data corresponding to each course teaching.

[0177] The course teaching outcome analysis module is used to acquire historical teaching information data; based on the historical teaching information data and the learning attention deviation data corresponding to each course, the teaching outcome of the course is analyzed and teaching outcome data is generated.

[0178] The course quality assessment module is used to assess the course teaching quality based on teaching outcome data and response assessment data corresponding to each course, and to generate course teaching quality data.

[0179] The course structure adjustment module is used to adjust the course structure based on course teaching quality data and learning attention deviation data corresponding to each course, generate adjusted course teaching plan data, and update the course data on the course teaching platform.

[0180] Therefore, the embodiments should be considered as exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of the equivalents of the application are intended to be included within the invention.

[0181] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features of the invention herein.

Claims

1. A method for course quality supervision and analysis based on big data, characterized in that, Includes the following steps: Step S1: Obtain course data from the course teaching platform; design course teaching plans based on the course data, and generate course teaching plan data; Step S2: Obtain course teaching log data from the course teaching platform; Based on course teaching log data and course teaching plan data, we conduct learning attention deviation assessments for each course and generate learning attention deviation data for each course. Based on the learning attention deviation data for each course, we conduct course response assessments for the corresponding courses within the course teaching plan data and generate response assessment data for each course. Step S3: Obtain historical teaching information data; based on the historical teaching information data and the learning attention deviation data corresponding to each course, analyze the teaching outcomes of the courses and generate teaching outcome data; Step S4: Based on the teaching outcome data and the response evaluation data corresponding to each course, evaluate the course teaching planning data to generate course teaching quality data; Step S5: Based on the course teaching quality data and the learning attention deviation data corresponding to each course, adjust the course teaching plan data to generate the adjusted course teaching plan data and update the course data on the course teaching platform.

2. The course quality supervision and analysis method based on big data according to claim 1, characterized in that, Step S1 includes the following steps: Step S11: Obtain course data from the course teaching platform; Step S12: Analyze the course adaptation groups based on the course data and generate course adaptation group data; Step S13: Design the total course volume based on the target audience of the course and generate total course volume data; Step S14: Based on the data of the target group for the course and the total number of courses, design the course teaching plan and generate course teaching plan data.

3. The course quality supervision and analysis method based on big data according to claim 1, characterized in that, Step S2 includes the following steps: Step S21: Obtain course teaching log data from the course teaching platform; Step S22: Identify student learning behavior status during course instruction based on course teaching log data, and generate student learning behavior status data; Step S23: Based on student learning behavior status data and course teaching plan data, conduct learning attention deviation assessment during the teaching of each course, and generate learning attention deviation data corresponding to each course teaching. Step S24: Based on the learning attention deviation data corresponding to each course, conduct a course response evaluation on the corresponding course teaching within the course teaching plan data, and generate response evaluation data for each course teaching.

4. The course quality supervision and analysis method based on big data according to claim 3, characterized in that, Step S23 includes the following steps: Step S231: Analyze the content sequence of the course based on the course teaching plan data to generate course content sequence data; Step S232: Analyze the complexity of the course content based on the order data of the course content, and generate course content complexity data; Step S233: Based on the complexity data of the course content, deduce the logical connection of the course content sequence data to generate logical connection deduction data of the course content; Step S234: Based on the logical connection of the course content, derive data to identify the interaction in the course teaching and generate course teaching interaction data; Step S235: Calculate the course learning cycle based on student learning behavior status data and generate course learning cycle data; Step S236: Based on the course teaching interaction data and course learning cycle data, conduct an assessment of the learning attention deviation during the teaching of each course, and generate learning attention deviation data corresponding to each course.

5. The course quality supervision and analysis method based on big data according to claim 4, characterized in that, Step S236 includes the following steps: Based on the data on course teaching interaction, the interaction patterns of each course teaching are identified, and corresponding interaction pattern data for each course teaching is generated. Based on the course learning cycle, conduct learning depth value-added analysis for each course teaching, and generate learning depth value-added data corresponding to each course teaching. Based on the learning depth value-added data and interaction mode data corresponding to each course, the interaction frequency analysis of each course teaching is carried out to generate interaction frequency data corresponding to each course teaching. Based on the interaction frequency data corresponding to each course, the learning attention deviation during each course teaching is assessed, and the learning attention deviation data corresponding to each course teaching is generated.

6. The course quality supervision and analysis method based on big data according to claim 3, characterized in that, Step S24 includes the following steps: Step S241: Monitor student learning behavior fluctuations based on the learning attention deviation data corresponding to each course, and generate behavior fluctuation data; Step S242: Locate the content of interest courses based on behavioral fluctuation data and generate interest course content data; Step S243: Based on the interest course content data, perform course-related content retrieval and identification to generate course-related content retrieval data; Step S244: Based on the data retrieved from the course-related content, conduct a course response evaluation on the corresponding course teaching within the course teaching plan data, and generate response evaluation data for each course teaching.

7. The course quality supervision and analysis method based on big data according to claim 1, characterized in that, Step S3 includes the following steps: Step S31: Obtain historical teaching information data; Step S32: Plot the course's teaching history stage outcome curves based on historical teaching information data, and generate teaching history stage outcome curve data; Step S33: Assess students' mastery of course knowledge based on the learning attention deviation data corresponding to each course, and generate course knowledge mastery data; Step S34: Calculate the course teaching outcome deviation based on course knowledge mastery data and course teaching outcome curve data at historical teaching stages, and generate course teaching outcome deviation data; Step S35: Analyze the course teaching outcomes based on the deviation data of course teaching outcomes and generate teaching outcome data.

8. The course quality supervision and analysis method based on big data according to claim 1, characterized in that, Step S4 includes the following steps: Step S41: Analyze the focus of each course based on the response evaluation data corresponding to the teaching of each course, and generate course focus data; Step S42: Assess students' understanding of course knowledge based on teaching outcome data and generate course knowledge understanding data; Step S43: Based on the course focus content data and course knowledge comprehension data, perform course teaching matching deviation diagnosis and generate course teaching matching deviation data; Step S44: Evaluate the course teaching quality based on the course teaching matching deviation data and generate course teaching quality data.

9. The course quality supervision and analysis method based on big data according to claim 1, characterized in that, Step S5 includes the following steps: Step S51: Analyze the fit between students and teachers' course pace based on the course teaching quality data, and generate course pace fit data; Step S52: Analyze the relevant interest courses based on the learning attention deviation data corresponding to each course, and generate relevant interest course data; Step S53: Based on the course rhythm fit data and related interest course data, perform course teaching collaboration matching to generate course teaching collaboration data; Step S54: Based on the course teaching collaboration data, perform structural adjustment processing on the course teaching planning data to generate structurally adjusted course teaching planning data and update the course data on the teaching interaction platform.

10. A course quality supervision and analysis system based on big data, characterized in that, For executing the big data-based course quality supervision and analysis method as described in claim 1, the big data-based course quality supervision and analysis system includes: The course planning module is used to acquire course data from the course teaching platform; design course teaching plans based on the course data; and generate course teaching plan data. The course teaching attention response analysis module is used to acquire course teaching log data from the course teaching platform; based on the course teaching log data and course teaching plan data, it evaluates the learning attention deviation during each course teaching and generates learning attention deviation data corresponding to each course teaching; based on the learning attention deviation data corresponding to each course teaching, it evaluates the course teaching response within the course teaching plan data and generates response evaluation data corresponding to each course teaching. The course teaching outcome analysis module is used to acquire historical teaching information data; based on the historical teaching information data and the learning attention deviation data corresponding to each course, the teaching outcome of the course is analyzed and teaching outcome data is generated. The course quality assessment module is used to assess the course teaching quality based on teaching outcome data and response assessment data corresponding to each course, and to generate course teaching quality data. The course structure adjustment module is used to adjust the course structure based on course teaching quality data and learning attention deviation data corresponding to each course, generate adjusted course teaching plan data, and update the course data on the course teaching platform.

Images