Teaching-learning-evaluation integrated intelligent course teaching method and system

Abstract

The invention discloses a teaching-learning-evaluation integrated intelligent course teaching method and system, and relates to the field of education, and the method comprises the steps: pushing a pre-class knowledge point preview task, an in-class test and practice type exploration task, and an after-class consolidation practice task to student users according to a pre-established knowledge graph, a practice graph and a practice known library, and the completion data of each task is automatically recorded, and the teaching classification task is adjusted in real time according to the completion data of each task. According to the method, a teaching improvement closed loop of a course integrating theory and practice is realized, and the problems of deep separation of theory teaching and practical application, loss of credible practical resources and the like are solved.

Description

Technical Field

[0001] This invention relates to the field of smart teaching technology that combines theoretical and practical courses, and in particular to a smart course teaching method and system that integrates teaching, learning and assessment. Background Technology

[0002] With the rapid development of artificial intelligence (AI) technology, smart education has become an important direction for the education field. AI technologies, such as deep learning, natural language processing, and computer vision, provide data-driven personalized support for teaching, driving the transformation of educational models. Smart education systems can generate learner profiles by analyzing learning behavior data, enabling customized learning paths; intelligent teaching assistants can provide 24 / 7 Q&A support; and classroom behavior analysis technology enhances the objectivity of teaching evaluation.

[0003] However, existing smart education systems still have significant shortcomings in key areas: (1) Deep separation between theoretical teaching and practical application: Especially in engineering courses, although students can complete simulation design, they find it difficult to transform theoretical knowledge into engineering logic; for example, in electronic design courses, a typical problem is that the circuit simulation is successful, but the actual device fails due to real factors such as signal interference and timing conflicts. (2) Lack of credible practical resources: For example, in electronic design courses, when students rely on open source materials on the Internet, they are easily misled by logical loopholes such as ignoring chip driving capabilities and insufficient clock jitter tolerance. Traditional courses lack authoritative practical resource libraries and real-time guidance mechanisms, which leads to repeated design errors. (3) Insufficient systematic teaching evaluation: Traditional evaluation mechanisms are fragmented and result-oriented, lacking the ability to evaluate the entire process of "pre-class diagnosis → in-class formation → post-class performance". They are unable to quantitatively assess knowledge gaps, dynamic adjustments in the teaching process, and engineering practice qualities (such as collaborative innovation), which seriously hinders the realization of the integrated training goal of "knowledge-ability-quality". Summary of the Invention

[0004] To address the technical problems existing in the background art, this invention proposes a teaching method and system for integrated teaching, learning and assessment smart courses.

[0005] Firstly, the present invention proposes an integrated teaching-learning-assessment smart curriculum teaching method, applicable to courses that combine theory and practice, comprising: Before class, based on pre-designed teaching classification tasks, the system intelligently pushes pre-class knowledge point preview tasks to students and automatically records the completion data of students' pre-class knowledge point preview tasks. The teaching classification tasks are adjusted in real time based on the completion data of students' pre-class knowledge point preview tasks; During class, in-class quizzes related to the pre-class knowledge points are released and the completion data of the in-class quizzes is automatically recorded. Based on the completion data of in-class quizzes, students are divided into qualified and unqualified groups. Theoretical learning tasks are pushed to the unqualified group. Based on the pre-established practice map, practical inquiry tasks are pushed to the qualified group. A preset number of related cases for practical inquiry tasks are intelligently pushed from the pre-established practice knowledge base, and the completion status of students in practical inquiry tasks is automatically recorded. After class, based on the students' performance in practical inquiry tasks, reinforcement exercises are pushed to students who have not met the completion goals, and the completion data of the reinforcement exercises is automatically recorded. Based on the completion data of each student user in practical inquiry tasks and consolidation exercise tasks, the course goal achievement rate of student users is analyzed, and the course goal achievement rate analysis results are obtained. Adjust teaching strategies for different teaching categories based on the results of the analysis of course objective achievement.

[0006] Preferably, the process of establishing a knowledge graph includes: Extract core concepts, theorems, formulas, and typical cases from textbooks and / or lecture notes as knowledge points; Design the types of logical relationships between knowledge points; among which, the types of logical relationships include: basic-advanced dependency relationship, containment-included hierarchical relationship, and premise-conclusion derivation relationship; Mark the logical relationships between knowledge points and organize these relationships into a knowledge logic table; Based on the knowledge logic table, the knowledge points and their relationships are visualized as a network topology diagram, forming a structured knowledge graph.

[0007] Preferably, the process of establishing the practice map includes: The sub-modules of practical inquiry tasks are identified, and the knowledge content of each sub-module in practical inquiry tasks is further detailed. The mapping relationship between knowledge content and knowledge graph nodes is marked, and the mapping relationship between knowledge content and knowledge graph nodes is organized into a practical logic table; Based on the time-based logical table, the knowledge points and their relationships are visualized as a network topology diagram, forming a structured practice map.

[0008] Preferably, the process of establishing a practical knowledge base includes: Based on industry case studies and widely used AI models in the field, a practical knowledge base including error attribution analysis is generated.

[0009] Preferably, after automatically recording the completion data of students' pre-class knowledge point preview tasks, the method further includes: Based on the completion data of students' pre-class knowledge point preview tasks, personalized learning path recommendations are made to students, and the learning trajectory records of the knowledge graph are updated in real time based on the students' choices or self-adjustments.

[0010] Preferably, personalized learning path recommendations are made to students based on their completion data of pre-class knowledge point preview tasks, including: Based on the completion data of students' pre-class knowledge point preview tasks and knowledge graphs, a dynamic learner profile of students is constructed. Based on the dynamic learner profiles of student users, candidate paths that match the current ability boundaries of student users are extracted from the knowledge graph. The expected benefits of different paths are calculated through a reinforcement learning model, generating 3 to 5 differentiated personalized learning path recommendation schemes, which are then recommended to student users.

[0011] Preferably, in practical inquiry tasks, AI digital tutors are used to interact with student users, and the difficult questions marked by student users during the interaction are automatically categorized into questions that require teachers to focus on explaining. The teaching categories and tasks are adjusted in real time based on the issues that teachers need to focus on explaining and / or the students' performance in practical inquiry tasks.

[0012] Preferably, in practical inquiry tasks, when a student user's timeout occurs on the task interface, such as when the single-step dwell time is greater than or equal to the preset time, or when repeated modification behavior is detected, the AI ​​digital tutor automatically pushes a pop-up explanation of the relevant knowledge graph nodes.

[0013] Preferably, in the consolidation exercise task, when the student user has not yet reached the completion goal in the consolidation exercise, the AI ​​digital tutor automatically pushes the micro-lesson video of the knowledge point, and then pushes the consolidation exercise task again until the goal is achieved.

[0014] Preferably, in the course objective achievement analysis process, assuming a course has n course objectives and m evaluation stages, the formula for calculating the achievement evaluation value of the i-th course objective is: ; Preferably, the evaluation process includes diagnostic evaluation, formative evaluation, and performance evaluation; Preferably, diagnostic assessment includes online learning before class and online pre-class tests; Preferably, formative assessment includes in-class quizzes, post-class reinforcement exercises, midterm exams, and final exams. Preferably, the performance evaluation calculation process includes: performing AI plagiarism checks on the completion data of student users in practical inquiry tasks and consolidation exercise tasks to obtain AI plagiarism scores, obtaining teacher scores for student users and scores among student users, and obtaining performance evaluation scores based on AI plagiarism scores, teacher scores, and scores among student users.

[0015] Preferably, the teaching strategy for adjusting the teaching categories of tasks based on the analysis results of the achievement of course objectives includes: teachers manually or by using AI large models to adjust the teaching strategies for the teaching categories of tasks based on the analysis results of the achievement of course objectives.

[0016] Secondly, this invention also proposes an integrated teaching-learning-assessment smart curriculum teaching system, which is applied to courses that combine theory and practice, including: a knowledge learning system module, an intelligent practice tool module, and a multi-dimensional assessment module; The knowledge learning system module includes a knowledge graph intelligent recommendation learning unit and an assignment tracking and monitoring unit. The knowledge graph-based intelligent recommendation learning unit is used to intelligently push pre-class knowledge point preview tasks to student users based on pre-established teaching classification tasks before class; The homework tracking and monitoring unit is used to automatically record the completion data of students' pre-class knowledge point preview tasks; The knowledge graph-based intelligent recommendation learning unit is used to adjust the teaching classification tasks in real time based on the completion data of students' pre-class knowledge point preview tasks; The knowledge graph intelligent recommendation learning unit is also used to release in-class quizzes related to pre-class knowledge point preparation tasks during class, and the homework tracking and monitoring unit is used to automatically record the completion data of in-class quizzes. The knowledge graph-based intelligent recommendation learning unit also uses the completion data of in-class quizzes to divide student users into groups that meet the standards and groups that do not, and pushes theoretical learning tasks to the groups that do not meet the standards. The intelligent tool module is used to push practical exploration tasks to the target group based on the pre-established practice map, and intelligently push a preset number of related cases of practical exploration tasks from the pre-established practice knowledge base. The assignment tracking and monitoring unit is also used to automatically record the completion status of student users in practical inquiry tasks; The knowledge graph-based intelligent recommendation learning unit is also used to adjust the teaching classification tasks in real time based on the students' performance in practical inquiry tasks; The knowledge graph-based intelligent recommendation learning unit is also used to push reinforcement exercises to students who have not met the completion goals based on their performance in practical inquiry tasks. The homework tracking and monitoring unit is also used to automatically record and track the completion data of reinforcement exercise tasks; The multidimensional evaluation module is used to analyze the achievement of course objectives by each student user based on their performance in practical inquiry tasks and consolidation exercises, and to obtain the results of the course objective achievement analysis.

[0017] Preferably, the knowledge learning system module also includes an AI digital tutor, which interacts with student users in practical inquiry tasks and automatically categorizes the difficult questions marked by students during the interaction into questions that require teachers to focus on explaining; the knowledge graph intelligent recommendation learning unit is also used to adjust the teaching classification tasks in real time according to the questions that require teachers to focus on explaining.

[0018] The proposed integrated teaching-learning-assessment smart curriculum teaching method and system, based on pre-established knowledge graphs, practice graphs, and practical knowledge bases, pushes pre-class knowledge point preview tasks, in-class quizzes and practical inquiry tasks, and post-class consolidation exercises to student users. It automatically records the completion data of each task and adjusts the teaching categories based on the completion data in real time, thus forming a closed loop for curriculum improvement that integrates theory and practice. This solves problems such as the deep separation between theoretical teaching and practical application, and the lack of reliable practical resources. Furthermore, by analyzing the completion status of each student in practical inquiry tasks and consolidation exercises, the system analyzes the achievement of course objectives. This allows teachers to adjust teaching strategies for different categories of tasks manually or using AI models based on the analysis results, thus forming a closed loop for teaching improvement. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the structure of an integrated teaching-learning-assessment smart curriculum teaching system in one embodiment of the present invention. Detailed Implementation

[0020] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0021] Firstly, the present invention proposes an integrated teaching-learning-assessment smart curriculum teaching method, applicable to courses that combine theory and practice, comprising: Before class, the system intelligently pushes pre-class knowledge point preview tasks to students based on pre-designed teaching classification tasks and automatically records the completion data of students' pre-class knowledge point preview tasks. The teaching classification tasks are adjusted in real time based on the completion data of students' pre-class knowledge point preview tasks; During class, in-class quizzes related to the pre-class knowledge points are released and the completion data of the in-class quizzes is automatically recorded. Based on the completion data of in-class quizzes, students are divided into qualified and unqualified groups. Theoretical learning tasks are pushed to the unqualified group; practical inquiry tasks are pushed to the qualified group based on the pre-established practice map. A preset number of related cases for practical inquiry tasks are intelligently pushed from the pre-established practice knowledge base, and the completion status of students in practical inquiry tasks is automatically recorded. After class, based on the students' performance in practical inquiry tasks, reinforcement exercises are pushed to students who have not achieved the completion goals, and the completion data of the reinforcement exercises is automatically recorded. Based on the completion data of each student user in practical inquiry tasks and consolidation exercises, an analysis of the student user's achievement of course objectives was conducted to obtain the results of the course objective achievement analysis.

[0022] This invention pushes pre-class knowledge point preview tasks, in-class quizzes and practical inquiry tasks, and post-class consolidation exercises to student users based on pre-established knowledge graphs, practice graphs, and practical knowledge bases. It automatically records the completion data of each task and adjusts the teaching categories based on the completion data in real time, thus forming a closed loop for curriculum improvement that integrates theory and practice. This solves problems such as the deep disconnect between theoretical teaching and practical application, and the lack of reliable practical resources. Furthermore, it analyzes the achievement of course objectives by analyzing the completion data of each student user in practical inquiry tasks and consolidation exercises. This allows teachers to adjust teaching strategies for the teaching categories based on the analysis results, either manually or using a large AI model, thus forming a closed loop for teaching improvement.

[0023] The design process of the teaching classification task in this embodiment includes: Pre-establish a knowledge graph; Teachers can annotate key teaching nodes in the knowledge graph; Teachers can design categorized teaching tasks based on the knowledge graph and the labeled key nodes.

[0024] The teaching tasks in this embodiment include: basic consolidation, ability enhancement, and extended exploration.

[0025] It is important to understand that a knowledge graph is a structured form of knowledge representation used to organize and visualize core concepts, theorems, formulas, typical cases, and other knowledge points in a course, as well as the logical relationships between them.

[0026] A practice graph establishes a mapping relationship between knowledge points and practical engineering examples. It is formed by decomposing practical design tasks, sorting out the sub-modules contained in each practical inquiry task, marking the mapping relationship between knowledge content and knowledge graph nodes, and finally creating a structured graph.

[0027] The Practical Knowledge Base is a self-built AI agent that primarily provides trusted practice resources and error analysis.

[0028] The knowledge graph creation process in this embodiment includes: Extract core concepts, theorems, formulas, and typical cases from textbooks and / or lecture notes as knowledge points; Then, design the types of logical relationships between knowledge points; among which, the types of logical relationships include, but are not limited to, basic-advanced dependency relationships, hierarchical relationships of inclusion-included relationships, and deductive relationships of premise-conclusion relationships; Next, mark the logical relationships between the knowledge points, and organize the logical relationships between the knowledge points into a knowledge logic table of knowledge point A → relationship → knowledge point B; Finally, based on the knowledge logic table, professional drawing tools are used to visualize the knowledge points and their relationships as a network topology diagram, forming a structured knowledge graph.

[0029] In this embodiment, the process of establishing the practice map includes: First, we sort out the sub-modules included in each practical inquiry task, and further refine the knowledge content included in each sub-module of the practical inquiry task. Then, the mapping relationship between knowledge content and knowledge graph nodes is marked, and the mapping relationship between knowledge content and knowledge graph nodes is initially organized into a practical logic table of practical inquiry tasks → sub-modules → knowledge content → knowledge points. Finally, based on the time-logic table, the knowledge points and their relationships are visualized as a network topology diagram, forming a structured practice map.

[0030] This embodiment addresses the core disconnect in engineering education—the failure of physical prototypes despite successful simulations—through a two-way mapping mechanism between practical graphs and knowledge graphs, significantly enhancing students' ability to translate theory into engineering practice.

[0031] This embodiment generates a practical knowledge base containing error attribution analysis based on an industry case library and widely used AI models in the field, such as ChatGPT and DeepSeek. For example, it provides PCB layout optimization solutions and spectrum analyzer usage guidelines for signal interference problems.

[0032] The practical knowledge base generated in this embodiment can provide reliable resources such as industry-verified standardized cases and standardized operating procedures, blocking the misleading logic loopholes in open source materials from the source, and significantly reducing the recurrence rate of design errors by students.

[0033] In this embodiment, after automatically recording the completion data of students' pre-class knowledge point preview tasks, the method further includes: Personalized learning path recommendations are made to students based on their completion data of pre-class knowledge point preview tasks. The resource sequence is automatically adjusted based on real-time completion data of pre-class knowledge point preview tasks to achieve personalized resource recommendations and accurately match micro-lesson videos, practice question sets and supplementary materials. Based on student users' choices or self-adjustments, the learning trajectory records of the knowledge graph are updated in real time, so that students can achieve self-directed learning according to their selected or self-adjusted personalized learning paths.

[0034] The completion data for the pre-class knowledge point preview task includes accuracy, time taken, and number of retries.

[0035] This includes recommending personalized learning paths to students based on their completion data of pre-class knowledge review tasks, including: Based on the completion data of students' pre-class knowledge point preview tasks and knowledge graphs, a dynamic learner profile of students is constructed to quantitatively label their cognitive level, learning preferences and potential obstacles. Based on the dynamic learner profiles of student users, a large AI model is used to extract candidate paths from the knowledge graph that match the current ability boundaries of student users. The expected benefits of different paths are calculated through a reinforcement learning model, generating 3 to 5 differentiated personalized learning path recommendation schemes and pushing them to student users.

[0036] In this embodiment, in practical inquiry tasks, an AI digital tutor interacts with student users, and the difficult questions marked by student users during the interaction are automatically categorized into questions that require the teacher to explain in detail. The teaching categories and tasks are adjusted in real time based on the issues that teachers need to focus on explaining.

[0037] In practical inquiry tasks, when a student user's timeout occurs on the task interface, such as when the single-step dwell time is greater than or equal to the preset time, or when repeated modification behavior is detected, an explanation of the relevant knowledge graph node will be automatically pushed out.

[0038] In practical inquiry tasks, when a student user's error rate for a certain knowledge point is greater than 30%, meaning they have not reached the completion goal, the system automatically extracts consolidation exercises that conform to Bloom's Taxonomy levels L1-L3 from the pre-set question bank for that knowledge point and pushes them to the student user in a targeted manner.

[0039] When a student fails to complete the reinforcement exercise task, a micro-lesson video for that knowledge point will be automatically pushed to them, followed by another reinforcement exercise task, until the goal is achieved.

[0040] In the analysis of course objective achievement, assuming a course has n course objectives and m evaluation stages, the formula for calculating the achievement evaluation value of the i-th course objective is: .

[0041] The evaluation process includes diagnostic evaluation, formative evaluation, and performance evaluation.

[0042] Diagnostic assessment includes online learning before class and online pre-class tests; Formative assessment includes in-class quizzes, post-class reinforcement exercises, midterm exams, and final exams. The performance evaluation calculation process includes: performing AI plagiarism checks on students' completion of practical inquiry tasks and consolidation exercise tasks to obtain AI plagiarism scores, obtaining teacher ratings for students and ratings among students, and obtaining performance evaluation scores based on AI plagiarism scores, teacher ratings, and ratings among students.

[0043] Therefore, diagnostic evaluation is used to track the completion data of each student user's pre-class knowledge point preview tasks for each knowledge point, and to generate class learning statistics and student profiles based on the completion data of each student user's pre-class knowledge point preview tasks for each knowledge point, so as to accurately locate knowledge gaps. Formative assessment is used to generate error analysis reports based on in-class quizzes and post-class reinforcement exercises, and to analyze common weaknesses and knowledge mastery abilities in the class based on mid-term and final exam results, so as to trigger adjustments in teaching strategies.

[0044] Performance-based assessment is used to comprehensively evaluate each student user's knowledge transfer and practical abilities.

[0045] This embodiment establishes a comprehensive evaluation system integrating diagnostic assessment, formative assessment, and performance assessment. It covers the entire process from precise pre-class diagnosis to comprehensive in-class testing and explicit post-class ability assessment. The system features tiered objectives, controllable processes, and traceable results, balancing efficiency and accuracy to achieve a comprehensive and precise evaluation of students' knowledge, abilities, and qualities. Furthermore, based on the evaluation value of course goal achievement, it drives the dynamic optimization of teaching content and strategies, forming a closed loop for teaching improvement.

[0046] In one specific embodiment, diagnostic evaluation (weighted at 10%) focuses on pre-class online learning (weighted at 4%) and online pre-class tests (weighted at 6%). Formative assessment (70% weight): includes in-class quizzes (10% weight), after-class reinforcement exercises (10% weight), and mid-term and final exams (50% weight). Performance evaluation (weighted at 20%): Based on the course project task of colored light circuit, combined with A1 plagiarism check, teacher scoring, and peer evaluation, a three-dimensional performance evaluation of "technology-collaboration-engineering" is formed to comprehensively assess knowledge transfer and practical ability.

[0047] In this embodiment, after obtaining the course objective achievement analysis results, the method further includes: adjusting the teaching strategy for the teaching classification tasks based on the course objective achievement analysis results.

[0048] In one specific embodiment, the teaching strategy for the teaching category tasks is adjusted based on the results of the course objective achievement analysis. Specifically, this includes: teachers manually adjusting the teaching strategy for the teaching category tasks based on the results of the course objective achievement analysis to form a closed loop for teaching improvement.

[0049] In another specific embodiment, the teaching strategy for the teaching category tasks is adjusted based on the results of the course objective achievement analysis. Specifically, this includes using an AI big data model to adjust the teaching strategy for the teaching category tasks based on the results of the course objective achievement analysis, so as to form a closed loop for teaching improvement.

[0050] Secondly, such as Figure 1 As shown, the present invention proposes an integrated teaching-learning-assessment smart curriculum teaching system, which includes: a knowledge learning system module, an intelligent practice tool module, and a multi-dimensional assessment module; The knowledge learning system module includes a knowledge graph intelligent recommendation learning unit and an assignment tracking and monitoring unit. The knowledge graph-based intelligent recommendation learning unit is used to intelligently push pre-class knowledge point preview tasks to student users based on pre-established teaching classification tasks before class; The homework tracking and monitoring unit is used to automatically record the completion data of students' pre-class knowledge point preview tasks; The knowledge graph-based intelligent recommendation learning unit is used to adjust the teaching classification tasks in real time based on the completion data of students' pre-class knowledge point preview tasks; The knowledge graph intelligent recommendation learning unit is also used to release in-class quizzes related to pre-class knowledge point preparation tasks during class, and the homework tracking and monitoring unit is used to automatically record the completion data of in-class quizzes. The knowledge graph-based intelligent recommendation learning unit also uses the completion data of in-class quizzes to divide student users into qualified and unqualified groups, and pushes theoretical learning tasks to the unqualified group to reinforce their knowledge. The intelligent tool module is used to pre-build a practice map and a practice knowledge base, and push practice-related inquiry tasks to the target group based on the practice map, and intelligently push a preset number of related cases of practice-related inquiry tasks from the practice knowledge base; The assignment tracking and monitoring unit is also used to automatically record the completion status of student users in practical inquiry tasks; The knowledge graph-based intelligent recommendation learning unit is also used to adjust the teaching classification tasks in real time based on the students' performance in practical inquiry tasks, so that teachers can focus on explaining the key knowledge points in the next class. The knowledge graph-based intelligent recommendation learning unit is also used to push reinforcement exercises to students who have not met the completion goals based on their performance in practical inquiry tasks. The homework tracking and monitoring unit is also used to automatically record and track the completion data of reinforcement exercise tasks; The multidimensional evaluation module is used to analyze the achievement of course objectives by each student user based on their performance in practical inquiry tasks and consolidation exercises. The results of the course objective achievement analysis are used to adjust teaching strategies accordingly.

[0051] In this embodiment, the knowledge graph intelligent recommendation learning unit is used to pre-build a knowledge graph, obtain the annotations of key teaching nodes of the knowledge graph by teachers and the teaching classification tasks designed by teachers based on the annotations of the knowledge graph, and intelligently push pre-class knowledge point preview tasks to students based on the teaching classification tasks.

[0052] It is important to understand that a knowledge graph is a structured form of knowledge representation used to organize and visualize core concepts, theorems, formulas, typical cases, and other knowledge points in a course, as well as the logical relationships between them.

[0053] A practice graph establishes a mapping relationship between knowledge points and practical engineering examples. It is formed by decomposing practical design tasks, sorting out the sub-modules contained in each practical inquiry task, marking the mapping relationship between knowledge content and knowledge graph nodes, and finally creating a structured graph.

[0054] The Practical Knowledge Base is a self-built AI agent that primarily provides trusted practice resources and error analysis.

[0055] The knowledge graph creation process in this embodiment includes: Extract core concepts, theorems, formulas, and typical cases from textbooks and / or lecture notes as knowledge points; Then, design the types of logical relationships between knowledge points; among which, the types of logical relationships include, but are not limited to, basic-advanced dependency relationships, hierarchical relationships of inclusion-included relationships, and deductive relationships of premise-conclusion relationships; Next, mark the logical relationships between the knowledge points, and organize the logical relationships between the knowledge points into a knowledge logic table of knowledge point A → relationship → knowledge point B; Finally, based on the knowledge logic table, professional drawing tools are used to visualize the knowledge points and their relationships as a network topology diagram, forming a structured knowledge graph.

[0056] Among them, the intelligent tools module is used to pre-build practice maps and practice knowledge bases.

[0057] The process of establishing the practice map in this embodiment includes: First, we sort out the sub-modules included in each practical inquiry task, and further refine the knowledge content included in each sub-module of the practical inquiry task. Then, the mapping relationship between knowledge content and knowledge graph nodes is marked, and the mapping relationship between knowledge content and knowledge graph nodes is initially organized into a practical logic table of practical inquiry tasks → sub-modules → knowledge content → knowledge points. Finally, based on the time-logic table, the knowledge points and their relationships are visualized as a network topology diagram, forming a structured practice map.

[0058] This embodiment generates a practical knowledge base with error attribution analysis based on industry case libraries and domain-specific models, such as ChatGPT and DeepSeek. For example, it provides PCB layout optimization solutions and spectrum analyzer usage guidelines for the "signal interference" problem.

[0059] The teaching tasks in this embodiment include: basic consolidation, ability enhancement, and extended exploration.

[0060] In this embodiment, the knowledge graph intelligent recommendation learning unit is also used to recommend personalized learning paths to students based on their completion data of pre-class knowledge point preview tasks, and to update the learning trajectory records of the knowledge graph in real time based on the students' choices or self-adjustments, so as to facilitate personalized learning.

[0061] The completion data for the pre-class knowledge point preview task includes accuracy, time taken, and number of retries.

[0062] This includes recommending personalized learning paths to students based on their completion data of pre-class knowledge review tasks, including: Based on the completion data of students' pre-class knowledge point preview tasks and knowledge graphs, a dynamic learner profile of students is constructed to quantitatively label their cognitive level, learning preferences and potential obstacles. Candidate paths that match the current ability boundaries of student users are extracted from the knowledge graph. The expected returns of different paths are calculated through a reinforcement learning model, and 3 to 5 differentiated personalized learning path recommendations are generated and pushed to student users.

[0063] The knowledge learning system module in this embodiment also includes an AI digital tutor, which interacts with student users in practical inquiry tasks and automatically categorizes the difficult questions marked by student users in the interaction into questions that need to be emphasized by the teacher; the knowledge graph intelligent recommendation learning unit is also used to adjust the teaching classification tasks in real time according to the questions that need to be emphasized by the teacher.

[0064] In practical inquiry tasks, when student users time out on the assignment interface, such as when the time spent on a single step is greater than or equal to the preset time, or when repeated modification behavior is detected, the AI ​​digital tutor is also used to automatically push pop-up knowledge graph related nodes for explanation. In practical inquiry tasks, when a student user's error rate on a certain knowledge point is greater than 30%, meaning they have not achieved the completion goal, the knowledge graph intelligent recommendation learning unit automatically extracts consolidation practice tasks that conform to Bloom's Taxonomy L1-L3 levels from the preset question bank for that knowledge point and pushes them to the student user in a targeted manner.

[0065] When a student fails to complete the reinforcement exercise task, the knowledge graph intelligent recommendation learning unit will automatically push a micro-lesson video for that knowledge point and then push the reinforcement exercise task again until the goal is achieved.

[0066] In this embodiment, the multidimensional evaluation module includes a diagnostic evaluation unit, a formative evaluation unit, and a performance evaluation unit; The diagnostic evaluation unit is used to track the completion data of each student user's pre-class knowledge point preview tasks for each knowledge point, and generate class learning statistics and student profiles based on the completion data of each student user's pre-class knowledge point preview tasks for each knowledge point, so as to accurately locate knowledge gaps. The formative assessment unit is used to generate error analysis reports based on in-class quizzes and post-class reinforcement exercises, and to analyze common weaknesses and knowledge mastery abilities in the class based on mid-term and final exam results, so as to trigger adjustments in teaching strategies.

[0067] The performance evaluation unit is used to perform AI plagiarism checks on students' completion of practical inquiry tasks and consolidation exercises, obtain AI plagiarism scores, and obtain teacher ratings for students and ratings among students. Based on the AI ​​plagiarism scores, teacher ratings, and ratings among students, each student's knowledge transfer and practical abilities are comprehensively evaluated.

[0068] In the analysis of course objective achievement, assuming a course has n course objectives and m evaluation stages, the formula for calculating the achievement evaluation value of the i-th course objective is: .

[0069] The evaluation process in this embodiment includes diagnostic evaluation, formative evaluation, and performance evaluation; The diagnostic assessment includes online learning before class and online pre-class tests; the formative assessment includes in-class quizzes, post-class reinforcement exercises, mid-term exams, and final exams; the performance assessment calculation process includes: using AI to check the completion data of students' practical inquiry tasks and reinforcement exercise tasks to obtain AI plagiarism scores, and obtaining teacher ratings for students and ratings among students. Based on the AI ​​plagiarism scores, teacher ratings, and ratings among students, the performance assessment score is obtained.

[0070] In this embodiment, after obtaining the course objective achievement analysis results, the knowledge learning system module is also used to adjust the teaching strategies of the teaching classification tasks based on the course objective achievement analysis results using an AI big model.

[0071] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A teach-learn-evaluate integrated smart course teaching method applied to a course combining theory and practice, characterized in that, Comprise: Before class, according to the pre-designed teaching classification task, the pre-class knowledge point preview task is intelligently pushed to the student user, and the completion data of the pre-class knowledge point preview task of the student user is automatically recorded; the completion data of the pre-class knowledge point preview task of the student user is adjusted in real time according to the completion data of the pre-class knowledge point preview task of the student user; during the class, the class test associated with the pre-class knowledge point preview task is issued, and the completion data of the class test is automatically recorded; according to the completion data of the class test, the student user is divided into a standard group and a non-standard group; according to the pre-established practice atlas, the practice inquiry task is pushed to the standard group, a preset number of associated cases of the practice inquiry task are intelligently pushed from the pre-established practice true knowledge library, and the completion of the student user in the practice inquiry task is automatically recorded; the non-standard group is pushed to the theory learning task, and after the non-standard group completes the theory learning task, the practice inquiry task is pushed to it; After class, according to the completion of the student user in the practice inquiry task, the consolidation exercise task is pushed to the student user who does not achieve the completion target, and the completion data of the consolidation exercise task is automatically recorded; according to the completion of each student user in the practice inquiry task and the completion data of the consolidation exercise task, the student user is analyzed for course goal achievement, and the course goal achievement analysis result is obtained; according to the course goal achievement analysis result, the teaching strategy of the teaching classification task is adjusted. The establishment process of the knowledge graph comprises: 2.The teaching-learning-evaluating integrated smart course teaching method according to claim 1, wherein, Extracting core concepts, theorems, formulas and typical cases in the teaching material and / or lecture notes as knowledge points; Design the logical relationship type between knowledge points; wherein, the logical relationship type comprises: the dependent relationship of basis-advanced, the hierarchical relationship of containing-contained, the derivation relationship of premise-conclusion; Annotate the logical relationship between knowledge points, and organize the logical relationship between knowledge points into a knowledge logic table; According to the knowledge logic table, the knowledge points and their relationships are visualized as a network topology graph to form a structured knowledge graph. The establishment process of the practice atlas comprises: 3.The teaching-learning-evaluating integrated smart course teaching method according to claim 1, wherein, Combining the sub-modules contained in the practice inquiry task, and refining the knowledge content contained in each sub-module in the practice inquiry task; Annotate the mapping relationship between knowledge content and knowledge graph nodes, and organize the mapping relationship between knowledge content and knowledge graph nodes into a practice logic table; According to the practice logic table, the knowledge points and their relationships are visualized as a network topology graph to form a structured practice atlas. The establishment process of the practice true knowledge library comprises: 4.The teaching-learning-evaluating integrated smart course teaching method according to claim 1, wherein, Based on the industry case library and the commonly used AI big model in the field, a practice true knowledge library containing error attribution analysis is generated. After automatically recording the completion data of the pre-class knowledge point preview task of the student user, it further comprises: 5.The teaching-learning-evaluating integrated smart course teaching method according to claim 1, wherein, According to the completion data of the pre-class knowledge point preview task of the student user, the student user is recommended for personalized learning path, and the learning trajectory record of the knowledge graph is updated in real time based on the selection or self-adjustment of the student user; Preferably, according to the completion data of the pre-class knowledge point preview task of the student user, the student user is recommended for personalized learning path, comprising: According to the completion data of the pre-class knowledge point preview task of the student user and the knowledge graph, the dynamic learner portrait of the student user is constructed; ​ According to the dynamic learner portrait of the student user, candidate paths conforming to the current ability boundary of the student user are extracted from the knowledge graph, the expected returns of different paths are calculated through a reinforcement learning model, 3-5 differentiated personalized learning path recommendation schemes are generated, and are recommended to the student user. 6.The teaching-learning-evaluating integrated smart course teaching method according to claim 1, wherein, In the practical inquiry task, the AI digital tutor interacts with the student user, and the difficult problems marked by the student user in the interaction are automatically classified as problems that need to be explained by the teacher; According to the problems that need to be explained by the teacher and / or the completion of the student user in the practical inquiry task, the teaching classification task is adjusted in real time; Preferably, in the practical inquiry task, when the student user's work interface in the practical inquiry task times out, such as the single-step stay time being greater than or equal to the preset time length, or repeated modification behavior is monitored, the AI digital tutor automatically pushes the pop-up knowledge graph associated node explanation; Preferably, in the consolidation exercise task, when the student user's consolidation exercise still does not reach the completion goal, the AI digital tutor automatically pushes the micro-lecture video of the knowledge point, and then pushes the consolidation exercise task again until the goal is reached.

7. The teaching-learning-evaluating integrated intelligent course teaching method of claim 1, wherein, In the course goal achievement degree analysis process, assuming that a course has n course goals and m evaluation links, the achievement evaluation value calculation formula of the i-th course goal is: ； Preferably, the evaluation links include diagnostic evaluation, formative evaluation, and performance evaluation; Preferably, the diagnostic evaluation includes pre-class online learning and online preview test; Preferably, the formative evaluation includes in-class quizzes, post-class consolidation exercises, midterm exams, and final exams; Preferably, the performance evaluation calculation process includes: AI plagiarism checking on the completion data of the student user in the practical inquiry task and the consolidation exercise task to obtain an AI plagiarism checking score; obtaining the teacher's score for the student user and the scores between each student user; and obtaining the performance evaluation score according to the AI plagiarism checking score, the teacher's score, and the scores between each student user. 8.The teaching-learning-evaluating integrated smart course teaching method according to claim 1, wherein, According to the course goal achievement degree analysis result, the teaching strategy of the teaching classification task is adjusted, specifically including: the teacher manually or using an AI large model to adjust the teaching strategy of the teaching classification task according to the course goal achievement degree analysis result.

9. A teach-learn-evaluate integrated intelligent course teaching system applied to a course combining theory and practice, characterized in that, It includes: a knowledge learning system module, an intelligent practice tool module, and a multi-dimensional evaluation module; The knowledge learning system module includes a knowledge graph intelligent recommendation learning unit, an AI digital tutor, and a homework tracking monitoring unit; The knowledge graph intelligent recommendation learning unit is used to intelligently push pre-class knowledge point preview tasks to student users according to pre-established teaching classification tasks before class; The homework tracking monitoring unit is used to automatically record the completion data of the student user's pre-class knowledge point preview tasks; The knowledge graph intelligent recommendation learning unit is used to adjust the teaching classification task in real time according to the completion data of the student user's pre-class knowledge point preview tasks; The knowledge graph intelligent recommendation learning unit is further configured to publish a class test associated with the pre-class knowledge point preview task in class, and the homework tracking monitoring unit is configured to automatically record completion data of the class test. The knowledge graph intelligent recommendation learning unit is further configured to divide student users into a target group and a non-target group according to the completion data of the class test, and push a theoretical learning task to the non-target group; The intelligent tool module is configured to push an exploration task of a practice type to the target group according to a pre-established practice graph, and intelligently push a preset number of associated cases of the exploration task of the practice type from a pre-established practice true knowledge library; The homework tracking monitoring unit is further configured to automatically record completion of the student users in the exploration task of the practice type; The knowledge graph intelligent recommendation learning unit is further configured to adjust the teaching classification task in real time according to the completion of the student users in the exploration task of the practice type; The knowledge graph intelligent recommendation learning unit is further configured to push a consolidation exercise task to a student user who does not achieve a completion target according to the completion of the student user in the exploration task of the practice type; The homework tracking monitoring unit is further configured to automatically record completion data of the consolidation exercise task; The multi-dimensional evaluation module is configured to analyze a course target achievement degree of the student users according to the completion of the student users in the exploration task of the practice type and the completion data of the consolidation exercise task, and obtain a course target achievement degree analysis result. 10.The integrated teaching-learning smart course teaching system according to claim 9, characterized in that, The knowledge learning system module further includes an AI digital tutor, the AI digital tutor is configured to interact with the student users in the exploration task of the practice type, and automatically classify a difficult problem marked by the student users in the interaction as a problem that needs to be explained by a teacher; and the knowledge graph intelligent recommendation learning unit is further configured to adjust the teaching classification task in real time according to the problem that needs to be explained by the teacher.

Images