AI learning method based on learning password and application thereof

Abstract

The invention discloses an AI learning method based on learning passwords and application thereof, and relates to the technical field of AI learning methods, the AI learning method comprises the steps of learning password generation analysis, personalized student plan making and learning task execution feedback, the learning password generation analysis generates learning passwords through learner behaviors and academic expressions, and the learning passwords are generated through learning task execution feedback; the evaluation module is used for evaluating the familiarity of learners to different knowledge points and the overall mastering level of a subject; the learning password generation and analysis comprises data collection, learning password generation and learning password analysis. By designing learning password generation analysis, a corresponding generation function of a personalized learning password is realized, the problem that a traditional learning analysis method cannot accurately diagnose strong and weak items of students in a personalized manner is solved, a basis is provided for formulating a targeted learning plan, a comprehensive subject ability index is provided for the students, and the learning efficiency of the students is improved. And the learning experience of students is greatly improved.

Description

Technical Field

[0001] The present invention relates to the technical field of AI learning methods, and specifically to an AI learning method based on learning passwords and its application. Background Art

[0002] In the current education field, with the popularity of online learning and the increasing maturity of digital education tools, the data of students' learning in virtual environments continues to accumulate. However, existing online learning systems often fail to make full use of this data to provide more personalized and effective learning support. Traditional learning analysis methods mainly focus on students' overall performance and lack an in-depth understanding of subject knowledge points. This one-size-fits-all approach cannot accurately identify students' strengths and weaknesses, resulting in a waste of educational resources and reduced learning effects.

[0003] Existing online learning systems often lack effective personalized guidance during the learning process and cannot fully meet students' diverse subject needs. Students may have a good grasp of certain knowledge points, but may have obvious weaknesses in other knowledge areas. The traditional education system is difficult to provide accurate teaching based on students' individual differences, resulting in a lack of timely personalized support when students encounter difficulties in the learning process.

[0004] Therefore, this application proposes an AI learning method based on learning passwords and its application to provide a path that better meets students' learning needs and promotes the overall improvement of learners.

[0005] 1. Patent document CN109545016B discloses a learning interaction method and a learning client based on a learning client. The above patent realizes the ability to promote the matching of students' autonomous learning progress on the learning client with their in-school learning progress, thereby avoiding affecting students' learning efficiency. However, the above patent cannot realize the corresponding generation function of personalized learning passwords.

[0006] 2. Patent document CN116089705B discloses a personalized recommendation method and system for learning topics based on AI. The above patent realizes continuous adjustment of the model to improve the recommendation efficiency of the method of the present invention for learning topics, but the above patent cannot realize targeted task planning functions.

[0007] 3. Patent document CN109636690B discloses a comprehensive scoring method for learning effectiveness based on learners' online learning behavior data. The above patent can motivate learners' learning enthusiasm and improve the utilization rate of MOOC course resources. Finally, the radar chart allows learners to accurately understand their own learning status, but the above patent cannot realize the function of associating knowledge points with feedback on learning task execution.

[0008] 4. Patent document CN115033716B discloses a self-learning method based on a general self-learning system. The above patent realizes the construction of reasoning rings and reasoning cycles about knowledge in the knowledge base, and maintains the self-consistency and activity of knowledge in the knowledge base through dynamic, sustainable and highly active reasoning cycles. However, the above patent cannot realize the planned balance function of learning tasks.

[0009] In summary, the above patent cannot realize the corresponding generation function of personalized learning passwords, the targeted task planning function, the learning task execution feedback knowledge point association function and the learning task planning balance function, resulting in the waste of teaching resources, the inability to formulate targeted learning plans, the inability to comprehensively improve subject understanding ability and low learning efficiency;

[0010] To this end, the present application proposes an AI learning method based on learning passwords and its application, which can realize the corresponding generation function of personalized learning passwords, targeted task planning function, learning task execution feedback knowledge point association function and planning balance function for learning tasks. Summary of the invention

[0011] The purpose of the present invention is to provide an AI learning method based on a learning password and its application, so as to solve the technical problems proposed in the above-mentioned background technology that the corresponding generation function of personalized learning passwords, the targeted task planning function, the learning task execution feedback knowledge point association function and the planning balance function of learning tasks cannot be realized, resulting in waste of teaching resources, inability to formulate targeted learning plans, inability to comprehensively improve subject understanding ability and low learning efficiency.

[0012] To achieve the above-mentioned purpose, the present invention provides the following technical solutions: an AI learning method based on learning passwords, including learning password generation analysis, personalized student plan formulation and learning task execution feedback, wherein the learning password generation analysis generates learning passwords through learner behavior and academic performance, and is used to evaluate learners' familiarity with different knowledge points and their overall mastery of the subject;

[0013] The learning password generation and analysis includes: data collection, learning password generation and learning password analysis;

[0014] Data collection involves using data mining and learning analytics algorithms to collect learning history, test scores, and subject interest data;

[0015] Learning password generation is based on association rule mining and clustering algorithm operation, using machine learning algorithm to consider the relevance of knowledge points, difficulty and learning style factors to generate corresponding learning passwords;

[0016] Learning code analysis uses data visualization and intelligence analysis to classify learners' strengths and weaknesses by analyzing the generated learning codes.

[0017] Preferably, the personalized learning plan formulation is to formulate a personalized learning plan according to the learning password, so as to ensure that the learner can be effectively consolidated and improved in both strengths and weaknesses;

[0018] Personalized learning plan formulation includes: difficulty weight allocation, daily task setting and knowledge point association;

[0019] The difficulty weight allocation is to use the weight adjustment algorithm to assign weights to each knowledge point according to the difficulty information in the learning password;

[0020] Daily task setting is to use the task planning algorithm to formulate daily learning tasks, covering the consolidation of strengths, training of weaknesses and learning of new knowledge;

[0021] Knowledge point association means designing tasks using association analysis algorithms to encourage learners to establish associations between different knowledge points.

[0022] Preferably, the learning task execution feedback is used to study according to the learning plan formulated according to the personalized learning plan, and obtain real-time feedback to adjust the plan;

[0023] Learning task execution feedback includes: daily task execution, real-time feedback and learning plan adjustment;

[0024] Daily task execution means using the learning task tracking system to execute tasks according to the daily learning plan, including: review, training and new knowledge learning;

[0025] Real-time feedback means AI uses real-time data analysis and machine learning algorithms to monitor learners’ learning progress and performance in daily tasks in real time and provide instant feedback;

[0026] Learning plan adjustment means dynamically adjusting the learning plan and optimizing the learning route based on real-time feedback using an adaptive learning path algorithm.

[0027] Preferably, the learning method is also designed with learning resource recommendation, which is used to intelligently recommend learning resources, including texts, videos and exercises, according to the learner's learning password and real-time needs;

[0028] Learning resource recommendations include: knowledge point matching, subject diversity, and real-time updates;

[0029] Knowledge point matching is to use the recommendation system and content filtering algorithm to match the learner's knowledge needs according to the learning password and recommend learning resources for related knowledge points;

[0030] Disciplinary diversity means providing learning resources of different forms and sources through the integration of multi-source disciplinary resources to meet the diverse disciplinary needs of learners;

[0031] Real-time updating means using scheduled crawlers and resource dynamic update algorithms to regularly update subject content and resources to ensure that learners obtain the latest and most relevant information.

[0032] Preferably, the learning task execution feedback is designed with NLP technology application, and the NLP technology application is used to provide a natural and highly interactive learning experience;

[0033] NLP technology applications include: speech recognition interaction, question answering and semantic recognition;

[0034] Speech recognition interaction means using speech recognition engine and interactive design, introducing speech recognition, supporting learners to ask and answer questions through voice, and improving learning interactivity;

[0035] Question answering is to use natural language understanding and question answering models to understand learners’ questions using NLP technology and provide detailed and personalized answers;

[0036] Semantic recognition uses semantic analysis algorithms and deep learning models to understand learners' language expressions through speech analysis, parse learning materials, accurately locate learning needs, provide personalized suggestions, and establish standardized models.

[0037] Preferably, the NLP technology application also designs a knowledge graph construction application, which uses the knowledge graph to strengthen the association between knowledge points and improve learners' understanding of the overall subject system;

[0038] Knowledge graph construction applications include: knowledge association graphs, practical application scenarios, and knowledge system construction;

[0039] The knowledge association graph uses the graph database and knowledge graph construction algorithm to build a knowledge graph, showing the associations between different knowledge points and helping learners understand the hierarchical structure of knowledge;

[0040] Practical application scenarios use scenario modeling and application scenario design, combined with knowledge graphs to display practical application scenarios, allowing learners to apply what they have learned in practical problems;

[0041] The knowledge system construction integrates knowledge system modeling and graph analysis algorithms, and uses knowledge graphs to help learners build a complete knowledge system, promoting in-depth understanding and subject integration.

[0042] Preferably, the personalized learning plan formulation is also designed with extension analysis and practical problem solving, which is used to emphasize the extension of knowledge and encourage learners to apply knowledge to solving practical problems;

[0043] Extensional analysis of practical problem solving includes: practical problem solving tasks, case analysis and contextual connections;

[0044] Practical problem-solving tasks are tasks designed using task design and case analysis, requiring learners to solve practical problems with the knowledge they have learned and deepen their understanding of the knowledge;

[0045] Case analysis is to use case library construction and case analysis tools to provide actual cases and guide learners to deepen the application and understanding of knowledge through case analysis;

[0046] Contextual association means using timeline modeling and knowledge relationship visualization to connect knowledge points in a contextual manner, showing the development process of knowledge and helping learners understand the context of knowledge.

[0047] Preferably, the real-time feedback is also designed with a learning effect evaluation adjustment, and the learning effect evaluation adjustment adjusts the learning plan through regular learning effect evaluation;

[0048] Learning effect evaluation and adjustment include: regular evaluation, learning effect analysis and learning plan adjustment;

[0049] Regular assessment means designing regular assessment tasks to evaluate learners’ mastery of knowledge points through assessment design and learner performance monitoring;

[0050] Learning effect analysis is to use learning passwords, learning feedback and evaluation results through data analysis and learning effect evaluation algorithms to comprehensively analyze learners' subject levels and discover potential problems;

[0051] Learning plan adjustment means adjusting the learning plan according to the evaluation results through adaptive learning path algorithm and feedback mechanism, further personalizing the learning path and maintaining learning efficiency.

[0052] Preferably, the application includes a connotation comprehension ability application:

[0053] S1. Self-study new knowledge and test the ability to understand the connotation based on deep learning algorithms;

[0054] S2. When the ability to understand the connotation is lower than the threshold of 80%, the test is stopped. The school learning materials are used as the learning objects, and the natural language processing technology NLP is used to parse the learning materials, including text classification and subject modeling, and this is used as a reference standard;

[0055] S3, train students' semantic comprehension ability, compare students' understanding with reference standards, and use a one-to-one interactive model for one week;

[0056] S4. Take another test on the new knowledge content one week later. If you reach 80% of the threshold, you will pass. If not, repeat the above training until you reach the threshold.

[0057] Preferably, the application also includes extensional comprehension capability application:

[0058] S5. Self-study new knowledge and test the ability to understand connotations based on deep learning algorithms;

[0059] S6. When the extensional comprehension ability is below the threshold of 80%, the test is stopped, and the knowledge graph is used to associate the knowledge with the themes of nature, technology, society, politics, economy, life, and emotional scenarios, and this is used as the reference standard range;

[0060] S7. Combine the subjects associated with the knowledge graph and conduct subjective communication with students. The communication results are very flexible and there is no standard answer, but it is based on the scope of knowledge extension.

[0061] S8. After training 5 to 10 topics, conduct an extension test on the new knowledge until the extension understanding value reaches the threshold of 80%, otherwise repeat the cycle training.

[0062] Compared with the prior art, the present invention has the following beneficial effects:

[0063] 1. The present invention realizes the corresponding generation function of personalized learning passwords by designing learning password generation analysis, solves the problem that traditional learning analysis methods cannot accurately and individually diagnose students' strengths and weaknesses, provides a basis for formulating targeted learning plans, and provides students with a comprehensive subject ability indicator, which greatly improves students' learning experience;

[0064] 2. The present invention realizes the targeted task planning function by designing personalized student planning, solves the problem of waste of traditional teaching resources, and improves learning efficiency;

[0065] 3. The present invention realizes the function of associating knowledge points with learning task execution feedback by designing learning task execution feedback, improves weak points in a targeted manner, helps learners understand the correlation between different knowledge fields, and improves overall subject understanding ability;

[0066] 4. The present invention realizes the planning balance function of learning tasks by designing learning effect evaluation and adjustment, provides personalized support and teaching suggestions for students, solves the problem of insufficient personalized support in traditional teaching, and makes subject knowledge more comprehensive and in-depth. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] Figure 1 It is a learning process diagram of the present invention;

[0068] Figure 2 This is a schematic diagram of the application of the connotation understanding ability value improvement of the present invention;

[0069] Figure 3 This is a schematic diagram of an application of the extension comprehension capability value enhancement of the present invention. DETAILED DESCRIPTION

[0070] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0071] Example 1

[0072] See also Figure 1 , Figure 2 and Figure 3 , an embodiment provided by the present invention: an AI learning method based on learning passwords, including learning password generation analysis, personalized student plan formulation and learning task execution feedback, wherein the learning password generation analysis generates learning passwords through learner behavior and academic performance, and is used to evaluate learners' familiarity with different knowledge points and their overall mastery of the subject;

[0073] The learning password generation and analysis includes: data collection, learning password generation and learning password analysis;

[0074] Data collection involves using data mining and learning analytics algorithms to collect learning history, test scores, and subject interest data;

[0075] Learning password generation is based on association rule mining and clustering algorithm operation, using machine learning algorithm to consider the relevance of knowledge points, difficulty and learning style factors to generate corresponding learning passwords;

[0076] Learning code analysis uses data visualization and intelligence analysis to analyze the generated learning codes and divide the learners' strengths and weaknesses;

[0077] Furthermore, log analysis and database query are used to collect information such as learners' learning history, test scores, and subject interests. Every interaction of learners on the learning platform, including watching videos, doing exercises, and participating in discussions, is recorded. In addition, students' test scores, extracurricular reading records, and subject assessment results are collected. Machine learning libraries in Python, such as Scikit-learn or TensorFlow, are used for model training. With the help of association rule mining and cluster analysis machine learning algorithms, students' performance on different knowledge points, subject difficulty and other factors are considered, and this information is combined to generate a learning password. Matplotlib or seaborn data visualization tools are used to present the distribution of learners' strengths and weaknesses in the form of charts. In addition, the intelligence analysis technology SWOT analysis is used to interpret the generated learning password, and the strengths and weaknesses in the subject are identified by statistically analyzing the learners' performance on various knowledge points.

[0078] Example 2

[0079] See also Figure 1 , Figure 2 and Figure 3 , an embodiment provided by the present invention: an AI learning method based on a learning password, wherein the personalized learning plan formulation is to formulate a personalized learning plan according to the learning password, so as to ensure that the learner can be effectively consolidated and improved in both strengths and weaknesses;

[0080] Personalized learning plan formulation includes: difficulty weight allocation, daily task setting and knowledge point association;

[0081] The difficulty weight allocation is to use the weight adjustment algorithm to assign weights to each knowledge point according to the difficulty information in the learning password;

[0082] Daily task setting is to use the task planning algorithm to formulate daily learning tasks, covering the consolidation of strengths, training of weaknesses and learning of new knowledge;

[0083] Knowledge point association means designing tasks using association analysis algorithms to encourage learners to establish associations between different knowledge points;

[0084] Furthermore, statistical methods, such as the calculation of mean and standard deviation and the modeling of expert systems, are used to evaluate the difficulty of each knowledge point based on the previously generated learning passwords. The opinions of subject experts, the performance of historical learners and other factors are used to adjust the difficulty level. Weights are assigned to each knowledge point. Task planning algorithms, such as greedy algorithms and genetic algorithms, are used to combine the difficulty weights and the learners' time availability to assign different priorities to the daily tasks, ensuring that the tasks not only include the consolidation of strengths and the training of weaknesses, but also cover the learning of new subject knowledge. Association analysis algorithms, such as association rule mining, are used to encourage learners to link related knowledge points together through cross-review and related tasks to improve overall subject understanding.

[0085] Example 3

[0086] See also Figure 1 , Figure 2 and Figure 3 , an embodiment provided by the present invention: an AI learning method based on a learning password, wherein the learning task execution feedback is used to learn according to a learning plan formulated according to a personalized learning plan, and obtain real-time feedback to adjust the plan;

[0087] Learning task execution feedback includes: daily task execution, real-time feedback and learning plan adjustment;

[0088] Daily task execution means using the learning task tracking system to execute tasks according to the daily learning plan, including: review, training and new knowledge learning;

[0089] Real-time feedback means AI uses real-time data analysis and machine learning algorithms to monitor learners’ learning progress and performance in daily tasks in real time and provide instant feedback;

[0090] Learning plan adjustment means dynamically adjusting the learning plan and optimizing the learning path based on the real-time feedback using the adaptive learning path algorithm;

[0091] Furthermore, learners complete designated tasks on the learning platform, which may include watching teaching videos, doing exercises, participating in discussions, etc. The learning behavior will be recorded, including learning time, completion status, etc. The AI ​​system analyzes the learner's behavioral data to identify the learner's current subject level, strengths and weaknesses, and may use machine learning algorithms to evaluate the learner's mastery in real time. In addition, the system can also monitor the learner's learning attitude, including participation and concentration. The AI ​​system adjusts the learning plan based on the learner's real-time performance and preset adjustment rules. If the learner performs well in a certain knowledge point, the learning time of this knowledge point can be reduced, and more energy can be devoted to the training of weaknesses. Conversely, if the learner finds it difficult to master a certain knowledge point, corresponding learning tasks can be added.

[0092] Example 4

[0093] See also Figure 1 , Figure 2 and Figure 3 , an embodiment provided by the present invention: an AI learning method based on learning passwords, wherein the learning task execution feedback is designed with an NLP technology application, and the NLP technology application is used to provide a natural and highly interactive learning experience;

[0094] NLP technology applications include: speech recognition interaction, question answering and semantic recognition;

[0095] Speech recognition interaction means using speech recognition engine and interactive design, introducing speech recognition, supporting learners to ask and answer questions through voice, and improving learning interactivity;

[0096] Question answering is to use natural language understanding and question answering models to understand learners’ questions using NLP technology and provide detailed and personalized answers;

[0097] Semantic recognition is to use semantic analysis algorithms and deep learning models to understand learners' language expressions through speech analysis to parse learning materials, accurately locate learning needs, provide personalized suggestions and establish standardized models;

[0098] Furthermore, we use Google Speech Recognition to integrate the speech recognition engine and design interactive scenarios, allowing learners to ask questions and answer questions raised by the system through voice. When the system responds, the text is converted into speech and provided to the learners. We integrate the natural language understanding NLU model BERT to analyze the questions raised by learners, extract key information, understand the semantics of the questions, retrieve and generate corresponding answers based on the understanding of the questions, ensure that the answers are accurate and personalized, use the semantic analysis algorithm Word Embeddings to analyze the learners' language expressions, understand the intentions and needs behind them, and provide personalized learning suggestions based on the results of semantic analysis, including recommending resources and adjusting learning plans.

[0099] Example 5

[0100] See also Figure 1 , Figure 2 and Figure 3 , an embodiment provided by the present invention: an AI learning method based on learning passwords, wherein the real-time feedback is also designed with a learning effect evaluation adjustment, and the learning effect evaluation adjustment adjusts the learning plan through regular learning effect evaluation;

[0101] Learning effect evaluation and adjustment include: regular evaluation, learning effect analysis and learning plan adjustment;

[0102] Regular assessment means designing regular assessment tasks to evaluate learners’ mastery of knowledge points through assessment design and learner performance monitoring;

[0103] Learning effect analysis is to use learning passwords, learning feedback and evaluation results through data analysis and learning effect evaluation algorithms to comprehensively analyze learners' subject levels and discover potential problems;

[0104] Learning plan adjustment means adjusting the learning plan according to the evaluation results through the adaptive learning path algorithm and feedback mechanism, further personalizing the learning path and maintaining learning efficiency;

[0105] Furthermore, assessment tasks are regularly arranged, including quizzes, exams and project assignments. The assessments cover different knowledge points to comprehensively evaluate learners' subject level. Learners' assessment scores and performance data are collected. Comprehensive reports on subject level are generated by combining learning codes, daily task completion status, and regular assessment results. Data analysis tools are used to count learners' performance on different knowledge points to identify strengths and weaknesses. Factors such as learners' learning attitude and participation are considered to comprehensively evaluate learning outcomes. Based on learners' strengths and weaknesses, the focus of daily learning tasks is reallocated. If a certain knowledge point performs poorly, corresponding training and practice are added. The depth and breadth of the subject are adjusted based on learners' interests and subject needs.

[0106] Working principle: First, the system collects data such as learners’ learning history, test scores, and subject interests, and uses machine learning algorithms to generate learning codes, which reflect students’ performance and subject level in different knowledge points. Through the weight distribution of difficulty, the system assigns weights to each knowledge point to highlight students’ strengths and weaknesses.

[0107] Based on the learning code, the system designs a personalized learning plan and assigns daily learning tasks, covering the consolidation of strengths, training of weaknesses and learning of new knowledge. Through the setting of daily tasks, the system not only focuses on the students' subject strengths, but also arranges training and learning in a targeted manner to improve the subject level. The tasks include difficulty weight adjustment to ensure that students maintain their strengths while strengthening their weaknesses in a targeted manner. This process also focuses on the connection between knowledge points, encouraging students to establish connections between different knowledge fields and promote in-depth understanding;

[0108] When students perform daily tasks, the system monitors their learning progress in real time through NLP technologies such as voice recognition, question answering, and semantic recognition, and provides immediate feedback, including subject understanding and task completion. These data are combined with learning passwords and regular assessment results to analyze learning effects. Based on the analysis results, the system adjusts the learning plan and reallocates the focus of tasks to ensure that students' subject levels are comprehensively improved. Through continuous evaluation and adjustment, learning methods can adapt to students' learning needs and achieve a personalized and efficient learning process.

[0109] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and range of equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. An AI learning method based on learning passwords, characterized in that: It includes learning password generation analysis, personalized student plan formulation and learning task execution feedback. The learning password generation analysis generates learning passwords based on learner behavior and academic performance, which are used to evaluate learners' familiarity with different knowledge points and their overall mastery of the subject; The learning password generation and analysis includes: data collection, learning password generation and learning password analysis; Data collection involves using data mining and learning analytics algorithms to collect learning history, test scores, and subject interest data; Learning password generation is based on association rule mining and clustering algorithm operation, using machine learning algorithm to consider the relevance of knowledge points, difficulty and learning style factors to generate corresponding learning passwords; Learning code analysis uses data visualization and intelligence analysis to classify learners' strengths and weaknesses by analyzing the generated learning codes.

2. The AI ​​learning method based on learning passwords according to claim 1, characterized in that: The personalized learning plan formulation is to formulate a personalized learning plan based on the learning code to ensure that the learner can be effectively consolidated and improved in both strengths and weaknesses; Personalized learning plan formulation includes: difficulty weight allocation, daily task setting and knowledge point association; The difficulty weight allocation is to use the weight adjustment algorithm to assign weights to each knowledge point according to the difficulty information in the learning password; Daily task setting is to use the task planning algorithm to formulate daily learning tasks, covering the consolidation of strengths, training of weaknesses and learning of new knowledge; Knowledge point association means designing tasks using association analysis algorithms to encourage learners to establish associations between different knowledge points.

3. The AI ​​learning method based on learning passwords according to claim 1, characterized in that: The learning task execution feedback is used to study according to the learning plan formulated by the personalized learning plan, and obtain real-time feedback to adjust the plan; Learning task execution feedback includes: daily task execution, real-time feedback and learning plan adjustment; Daily task execution means using the learning task tracking system to execute tasks according to the daily learning plan, including: review, training and new knowledge learning; Real-time feedback means AI uses real-time data analysis and machine learning algorithms to monitor learners’ learning progress and performance in daily tasks in real time and provide instant feedback; Learning plan adjustment means dynamically adjusting the learning plan and optimizing the learning route based on real-time feedback using an adaptive learning path algorithm.

4. The AI ​​learning method based on learning passwords according to claim 1, characterized in that: The learning method is also designed with learning resource recommendation, which is used to intelligently recommend learning resources, including texts, videos and exercises, based on the learner's learning password and real-time needs; Learning resource recommendations include: knowledge point matching, subject diversity, and real-time updates; Knowledge point matching is to use the recommendation system and content filtering algorithm to match the learner's knowledge needs according to the learning password and recommend learning resources for related knowledge points; Disciplinary diversity means providing learning resources of different forms and sources through the integration of multi-source disciplinary resources to meet the diverse disciplinary needs of learners; Real-time updating means using scheduled crawlers and resource dynamic update algorithms to regularly update subject content and resources to ensure that learners obtain the latest and most relevant information.

5. The AI ​​learning method based on learning passwords according to claim 3, characterized in that: The learning task execution feedback is designed with NLP technology application, which is used to provide a natural and interactive learning experience; NLP technology applications include: speech recognition interaction, question answering and semantic recognition; Speech recognition interaction means using speech recognition engine and interactive design, introducing speech recognition, supporting learners to ask and answer questions through voice, and improving learning interactivity; Question answering is to use natural language understanding and question answering models to understand learners’ questions using NLP technology and provide detailed and personalized answers; Semantic recognition is the use of semantic analysis algorithms and deep learning models to understand learners' language expressions through speech analysis, parse learning materials, accurately locate learning needs, provide personalized suggestions and establish standardized models.

6. The AI ​​learning method based on learning passwords according to claim 5, characterized in that: The NLP technology application also includes a knowledge graph construction application, which uses the knowledge graph to strengthen the relationship between knowledge points and improve learners' understanding of the overall subject system; Knowledge graph construction applications include: knowledge association graphs, practical application scenarios, and knowledge system construction; The knowledge association graph uses the graph database and knowledge graph construction algorithm to build a knowledge graph, showing the associations between different knowledge points and helping learners understand the hierarchical structure of knowledge; Practical application scenarios use scenario modeling and application scenario design, combined with knowledge graphs to display practical application scenarios, allowing learners to apply what they have learned in practical problems; The knowledge system construction integrates knowledge system modeling and graph analysis algorithms, and uses knowledge graphs to help learners build a complete knowledge system, promoting in-depth understanding and subject integration.

7. The AI ​​learning method based on learning passwords according to claim 2, characterized in that: The personalized learning plan formulation also includes extension analysis and practical problem solving, which is used to emphasize the extension of knowledge and encourage learners to apply knowledge to solving practical problems; Extensional analysis of practical problem solving includes: practical problem solving tasks, case analysis and contextual connections; Practical problem-solving tasks are tasks designed using task design and case analysis, requiring learners to solve practical problems with the knowledge they have learned and deepen their understanding of the knowledge; Case analysis is to use case library construction and case analysis tools to provide actual cases and guide learners to deepen the application and understanding of knowledge through case analysis; Contextual association means using timeline modeling and knowledge relationship visualization to connect knowledge points in a contextual manner, showing the development process of knowledge and helping learners understand the context of knowledge.

8. The AI ​​learning method based on learning passwords according to claim 3, characterized in that: The real-time feedback is also designed with learning effect evaluation adjustment, which adjusts the learning plan through regular learning effect evaluation; Learning effect evaluation and adjustment include: regular evaluation, learning effect analysis and learning plan adjustment; Regular assessment means designing regular assessment tasks to evaluate learners’ mastery of knowledge points through assessment design and learner performance monitoring; Learning effect analysis is to use learning passwords, learning feedback and evaluation results through data analysis and learning effect evaluation algorithms to comprehensively analyze learners' subject levels and discover potential problems; Learning plan adjustment means adjusting the learning plan according to the evaluation results through adaptive learning path algorithm and feedback mechanism, further personalizing the learning path and maintaining learning efficiency.

9. An application of an AI learning method based on learning passwords, used in an AI learning method based on learning passwords according to any one of claims 1 to 8, characterized in that: The applications include the application of connotation comprehension ability: S1. Self-study new knowledge and test the ability to understand the connotation based on deep learning algorithms; S2. When the ability to understand the connotation is lower than the threshold of 80%, the test is stopped. The school learning materials are used as the learning objects, and the natural language processing technology NLP is used to parse the learning materials, including text classification and subject modeling, and this is used as a reference standard; S3, train students' semantic comprehension ability, compare students' understanding with reference standards, and use a one-to-one interactive model for one week; S4. Take another test on the new knowledge content one week later. If you reach 80% of the threshold, you will pass. If not, repeat the above training until you reach the threshold.

10. The application of an AI learning method based on learning passwords according to claim 9, characterized in that: The application also includes the extensional understanding ability application: S5. Self-study new knowledge and test the ability to understand connotations based on deep learning algorithms; S6. When the extensional comprehension ability is below the threshold of 80%, the test is stopped, and the knowledge graph is used to associate the knowledge with the themes of nature, technology, society, politics, economy, life, and emotional scenarios, and this is used as the reference standard range; S7. Combine the subjects associated with the knowledge graph and conduct subjective communication with students. The communication results are very flexible and there is no standard answer, but it is based on the scope of knowledge extension. S8. After training 5 to 10 topics, conduct an extension test on the new knowledge until the extension understanding value reaches the threshold of 80%, otherwise repeat the cycle training.

Images