Personalized education platform and method based on artificial intelligence

Abstract

The invention relates to the technical field of personalized education platforms, in particular to an artificial intelligence-based personalized education platform and method.The artificial intelligence-based personalized education platform comprises a storage module, an information acquisition module, a target generation module, a scheme making module and a display module, and the scheme making module is used for generating an interested knowledge point sequence according to interested knowledge points set by a user and displaying the interested knowledge point sequence; interested knowledge points in the interested knowledge point sequence are divided into a plurality of preset sub-learning points, and an interest value judgment unit is arranged in the scheme making module; the display module is connected with the target generation module, the display module comprises a plurality of small screens, a plurality of preset sub-learning points are displayed through the small screens, the characteristics of each learner can be accurately grasped, and customized learning content and paths are provided; the learning effect can be evaluated in real time, and targeted feedback and guidance can be given in time; the learning progress and content can be dynamically adjusted according to the performance and state of the learner; and humanized learning experience is provided.

Description

Technical Field

[0001] The present invention relates to the technical field of personalized education platforms, and in particular to a personalized education platform and method based on artificial intelligence. Background Art

[0002] With the rapid development of information technology, the field of education is also undergoing unprecedented changes. In recent years, online education platforms have sprung up like mushrooms after rain, providing learners with rich and diverse learning resources and flexible learning methods. These platforms have, to a certain extent, broken the time and space limitations of traditional education, making learning more convenient. However, although these platforms have made significant progress in content coverage and accessibility, they still have many shortcomings in personalized learning experience and learning effects.

[0003] At present, most mainstream online education platforms on the market adopt a one-size-fits-all teaching model. They usually provide preset course content and fixed learning paths, which are difficult to effectively adjust according to the individual differences of learners. Although this standardized teaching method is more convenient in management and promotion, it ignores the personalized needs of learners. Each learner has a unique knowledge background, learning style and interests, and existing education platforms often cannot respond to these individual differences in a timely and accurate manner.

[0004] In addition, existing education platforms also have obvious deficiencies in learning effect evaluation and feedback mechanisms. Most platforms rely solely on simple test questions to evaluate learning outcomes, and are unable to fully and deeply understand the learners' true level of mastery. At the same time, the feedback from these platforms is usually delayed and general, making it difficult to provide learners with timely and targeted guidance. This rough evaluation and feedback method not only fails to effectively stimulate learners' interest in learning, but may also lead to low learning efficiency.

[0005] Another common problem is that existing platforms are unable to dynamically adjust learning content. They often use a preset learning progress that is difficult to flexibly adjust based on learners' real-time performance and status. This rigid learning progress arrangement may make some learners feel overwhelmed, while others may feel that the progress is too slow, thus affecting the overall learning experience and effect.

[0006] What is more noteworthy is that existing education platforms generally lack attention and analysis of learners' emotional states. The learning process is not only the accumulation of knowledge, but also a complex emotional experience process. Learners' emotional states directly affect the effectiveness and sustainability of learning, but current platforms are almost blank in this regard.

[0007] In the face of these problems, there is an urgent need for an educational platform that can truly realize personalized and intelligent learning. Summary of the invention

[0008] The present invention proposes an AI-based personalized education platform and method to address these deficiencies of existing education platforms. It aims to build a truly intelligent and personalized education platform through advanced AI technology combined with big data analysis and emotional computing. This platform can provide each learner with the most optimized learning experience, greatly improve learning efficiency and effectiveness, and cultivate learners' learning interest and autonomous learning ability.

[0009] The present invention proposes a personalized education platform based on artificial intelligence, including: a storage module, the storage module includes a first database and a second database, wherein: the first database stores user information, user's historical learning data and user identity authentication information; the second database stores learning materials corresponding to each knowledge point, knowledge points corresponding to each learning material and resource classification information corresponding to each learning material; an information acquisition module, the information acquisition module stores a number of test questions and is used to obtain the user's identity authentication information, retrieve the user's historical learning data from the storage module according to the user's identity authentication information and transmit the historical learning data to a target generation module; a target generation module, which is used for users to perform target generation training; a program formulation module, which is used to generate a sequence of knowledge points of interest according to the knowledge points of interest set by the user, and the knowledge points of interest in the sequence of knowledge points of interest are divided into a number of preset sub-learning points, and the program formulation module is provided with an interest value determination unit element; a display module, the display module is connected to the target generation module, the display module includes a plurality of small screens, and the preset plurality of sub-learning points are displayed through the plurality of small screens, the display module also includes an answer input module, a voice evaluation module, a timing module and a feedback unit; the program formulation module also includes a reinforcement learning module, the reinforcement learning module includes a first feedback unit, the first feedback unit generates a first feedback information when the user completes a test question through the answer input module; the voice evaluation module evaluates the user's oral answer and outputs a second feedback information, and the second feedback information is a voice with a score; when the plurality of test questions of the sub-learning point are answered, the program formulation module generates a learning plan for the user, the learning plan sets an exercise question option after the test question through the display module, and the exercise question option is started through the feedback unit when it is triggered; if the exercise question is triggered, the learning effect diagram is output after the exercise question is completed.

[0010] Preferably, the storage module also includes the preference level corresponding to each small screen in the display module, and each small screen is set as a small screen to be displayed; when the user inputs an answer on the first small screen, the information acquisition module will store the user's preference for the first small screen; a timing module is provided in the information acquisition module, and the timing module calculates the user's question-solving time. When the user is solving the problem, the timing module records the user's question-solving time for each problem in real time, and transmits the user's question-solving time to the first database of the storage module for storage; a preference determination unit is provided in the program formulation module, and when the user completes the exercises of the first small screen through the display module, the exercises corresponding to the first small screen are set to be input through the display module according to the user's preference value for the first small screen; the exercises corresponding to the first small screen are set to include a number of questions to be practiced, and when the user completes all the exercises corresponding to the first small screen, the program formulation module displays the exercises corresponding to the first small screen, and confirms the user's preference for the exercises corresponding to the entire first small screen according to the time taken to answer the exercises of the first small screen completed by the user.

[0011] Preferably, the first feedback information in the reinforcement learning module includes recording the correct or incorrect user operations or the user's irrelevant answers to the questions; when the user makes an error in the operation, the mini program is used to retrieve the relevant error points and relevant test questions for the error situation; if the user makes an irrelevant answer to the question, relevant questions will be automatically asked during the setting, and then the voice evaluation module will be used for testing to reinforce the user's relevant error points.

[0012] Preferably, the display module is also provided with an auxiliary unit, which collects sound information emitted by the user when answering questions, and the sound information is collected through a microphone; if the user answers the questions independently or answers the questions incorrectly, the small screen in the display module displays prompt information, and the first feedback information of the reinforcement learning module also contains the collection of sound information emitted by the user when answering questions.

[0013] Preferably, the voice evaluation module collects the user's voice through a microphone, and evaluates the energy or emotion the user spends on a question when answering the question by evaluating the collected user's voice; when the user completes the exercises corresponding to all sub-learning points, the user's energy and emotion in answering the sub-learning points are calculated to obtain the user's information corresponding to the sub-learning points.

[0014] The method of the personalized education platform based on artificial intelligence includes the following steps:

[0015] Step S1: firstly authenticate the user, obtain the user's authentication information, and then fetch the user's historical learning data from the first database through the user's authentication information, and transmit the user's historical learning data to the target generation module;

[0016] Step S2: The target generation module selects points of interest according to the needs of the user, selects the knowledge points corresponding to the points of interest of the user, and guides the user to generate learning points through the target generation module;

[0017] Step S3: Determine the interest point selected by the user as an interest sub-learning point, and display the sub-learning point through a display module, wherein a plurality of small screens are provided in the display module, and the resource classification corresponding to the sub-learning point is displayed through the small screens;

[0018] Step S4: According to the needs of the user, after the sub-learning point is displayed, the corresponding operation is performed through the answer input module on the display module. If the user answers the question independently, the answer input module outputs the first feedback information according to the user's answer. If the user answers the question incorrectly, the system directly gives the answer and outputs the second feedback information.

[0019] Step S5: After the user answers the question through the answer input module, the answer input module generates a learning plan for the user through the plan formulation module;

[0020] Step S6: Then, according to the learning plan completed by the user, exercises for the user are formed, and the exercises completed by the user are displayed. If the user chooses to trigger, the exercises are displayed through the display module, so that the user can make independent choices. If the user does not choose, the exercises are generated according to the random selection of the system;

[0021] Step S7: After the user completes the exercises in step S6, a learning effect graph is formed, and then the user's overall learning effect graph is formed after the exercises in each sub-learning point are completed.

[0022] Preferably, in step S2, a number of sub-learning points are provided in the storage module, and a keyword screening module is used to select words for each sub-learning point, and then the sub-learning points are summarized according to the set keywords to obtain the learning knowledge points corresponding to the user; in the keyword screening module, mining is performed according to the keywords contained in the sub-learning points, and then the included elements in the sub-learning points are extracted according to the keywords, and organized through the display module.

[0023] Preferably, when the sub-learning points of step S3 are output, the sub-learning points are divided into several preset modules according to the keywords in the keyword screening module, and each module has corresponding sub-learning points for division. Each sub-learning point divides the data in the form of keywords, and then divides the divided data into designated sub-learning points, thereby generating several sub-learning points.

[0024] Preferably, in step S4, the answer input module includes multiple answer input methods: the first method is to play through voice, the user listens to the question and answers it by voice; the second method is to use the small screen in the display module for display, and display the corresponding question, and the user selects the answer to answer the question; the third method is to use the small screen in the display module for display, and display the corresponding question, and the user answers by handwriting; the fourth method is to use the small screen in the display module for display, and display the corresponding question, and display the answer the user wants in the form of a program in the system, so as to input it using the keyboard; when answering questions using any of the four methods, the system will obtain a first feedback message, and will also generate a second feedback message and output it by voice; when the user answers the question incorrectly, the system automatically asks and answers the question through the reinforcement learning module, and outputs the question sentence through voice, and then judges whether the user answers correctly based on the user's voice. When the user keeps answering the question incorrectly, the system will evaluate the user's question based on the user's voice.

[0025] Preferably, in step S5, after the user finishes answering the questions, the system will automatically retrieve the learning materials for the next sub-learning point based on the user's answers, and then display the questions on a small screen, so that the user can make independent choices and then display the corresponding questions. If there is no independent choice, the system will display the planned questions. If the user makes an error in operation, the system will correct the user's error by reinforcing the test questions in the learning process.

[0026] The beneficial technical effects brought by the technical solution of the present invention are:

[0027] The present invention can accurately grasp the characteristics of each learner and provide tailor-made learning content and paths; it can evaluate learning effects in real time and give timely and targeted feedback and guidance; it can dynamically adjust learning progress and content according to the learner's performance and status; more importantly, it can pay attention to and analyze the learner's emotional state and provide a humanized learning experience.

[0028] The implementation of this invention can not only solve the deficiencies of existing educational platforms in terms of personalization, real-time feedback, dynamic adjustment, and emotional care, but also open up a new path for the intelligent development of the education field. Its successful implementation will greatly improve learning efficiency, enhance learning experience, and ultimately realize the educational ideal of teaching students in accordance with their aptitude, providing strong support for the all-round development of each learner. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 It is the overall logic block diagram of the system of the present invention.

[0030] Figure 2 It is a logic block diagram of the storage module of the present invention.

[0031] Figure 3 It is a logic block diagram of the information acquisition module of the present invention.

[0032] Figure 4 A logic block diagram of the target generation module of the present invention.

[0033] Figure 5 A logical block diagram of the modules for formulating the scheme of the present invention.

[0034] Figure 6 It is a logic block diagram of the display module of the present invention.

[0035] Figure 7 It is a logical block diagram of the reinforcement learning module of the present invention.

[0036] Figure 8 It is a logic block diagram of the speech evaluation module of the present invention. DETAILED DESCRIPTION

[0037] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation methods, structures, features and effects thereof are described in detail below in conjunction with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0038] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0039] See also Figure 1-8 The present invention provides a personalized education platform and method based on artificial intelligence, aiming to solve the problem that traditional education methods are difficult to meet the personalized needs of students. By combining artificial intelligence technology and personalized learning concepts, the present invention can provide each student with a tailored learning experience.

[0040] First, let us understand the specific content of the present invention in detail. According to the first aspect of the present invention, a personalized education platform based on artificial intelligence is provided, which includes the following key modules:

[0041] The storage module 1 includes a first database 11 and a second database 12. The first database 11 stores user information, historical learning data of users, and user identity authentication information; the second database 12 stores learning materials corresponding to each knowledge point, knowledge points corresponding to each learning material, and resource classification information corresponding to each learning material. This design enables the platform to fully grasp the user's learning situation and available teaching resources, providing a solid data foundation for subsequent personalized learning.

[0042] Information acquisition module 2 stores several test questions and is used to obtain the user's identity authentication information. The function of this module is to retrieve the user's historical learning data from storage module 1 according to the user's identity authentication information, and transmit this data to target generation module 3. In this way, the system can quickly understand the user's learning background and current status, laying the foundation for formulating a personalized learning plan.

[0043] Goal Generation Module 3 is used for users to conduct goal generation training. The design concept of this module is to help users clarify their learning goals, thereby improving the pertinence and efficiency of learning. For example, if a user's historical data shows that he performs well in mathematics but is weak in language skills, then Goal Generation Module 3 may suggest that the user devote more energy to language learning to achieve all-round development.

[0044] The program formulation module 4 generates a sequence of knowledge points of interest based on the knowledge points of interest set by the user. It is worth noting that each knowledge point of interest in this sequence of knowledge points of interest is further subdivided into a number of preset sub-learning points. This subdivision strategy makes the learning plan more refined and specific, which is conducive to the user to gradually overcome the difficulties. In addition, the program formulation module 4 is also provided with an interest value determination unit 41, which is used to evaluate the user's interest in each knowledge point, so as to further optimize the learning plan.

[0045] The display module 5 is connected to the target generation module 3 and includes a plurality of small screens 51. These small screens 51 are used to display a plurality of preset sub-learning points, so that the user can intuitively understand his or her learning progress. In addition, the display module 5 also includes an answer input module 52, a voice evaluation module 53, a timing module 54 and a feedback unit 55. These components work together to provide users with a variety of learning interaction methods.

[0046] A prominent feature of the present invention is that the program formulation module 4 also includes a reinforcement learning module 42. The reinforcement learning module 42 includes a first feedback unit, and when the user completes a test question through the answer input module 52, the first feedback unit generates first feedback information. This instant feedback mechanism can help users understand their learning effects in a timely manner, thereby adjusting their learning strategies.

[0047] Another innovative point is the design of the voice evaluation module 53. It can not only evaluate the user's oral answer, but also output a second feedback information, and this feedback information is presented in the form of voice with a score. This design is particularly suitable for language learning and can effectively improve the user's oral ability.

[0048] When the user completes several test questions of the sub-learning point, the program formulation module 4 will generate a learning program for the user. This learning program sets the exercise options after the test questions through the display module 5. The user can choose whether to trigger these exercises, and the triggering is initiated through the feedback unit 55. If the user chooses to trigger the exercise, after completing the exercise, the system will output a learning effect diagram to intuitively display the user's learning results.

[0049] The second aspect of the present invention further optimizes the function of the storage module 1. Specifically, the storage module 1 also includes the preference level corresponding to each small screen 51 in the display module 5. The system sets each small screen 51 as a small screen to be displayed. When the user enters an answer on the first small screen, the information acquisition module 2 will store the user's preference for this small screen. This design can help the system understand the user's learning preferences, thereby providing a more personalized learning experience.

[0050] It is worth mentioning that a timing module 54 is also provided in the information acquisition module 2. Its function is to calculate the time taken by the user to answer questions and record the time taken by the user to answer each question in real time. These time data will be transmitted to the first database 11 of the storage module 1 for storage. By analyzing these time data, the system can understand the user's performance on different types of questions, so as to formulate a learning plan more accurately.

[0051] The program formulation module 4 is also provided with a preference determination unit 43. When the user completes the exercises on the first small screen, the system will set corresponding exercises according to the user's preference value for this small screen. The exercises corresponding to each small screen include a number of questions to be practiced. After the user completes all the exercises, the program formulation module 4 will confirm the user's preference for the entire practice process based on the time the user takes to answer the questions. This design enables the system to continuously optimize the presentation of learning content and improve the user's learning interest and efficiency.

[0052] The third aspect of the present invention focuses on the specific content of the first feedback information in the reinforcement learning module 42. This feedback information not only includes a record of whether the user's operation is correct, but also includes whether the user's answer is incorrect. When the user makes an error in operation, the system will use the applet to retrieve the relevant error points again and find the corresponding test questions. If the user answers the wrong question, the system will automatically ask relevant questions after a set time, and then test through the voice evaluation module 53 to reinforce the user's error points. This design can help users better understand and master knowledge points and improve learning effects.

[0053] In the fourth aspect of the present invention, the display module 5 is also provided with an auxiliary unit 56. The main function of this auxiliary unit 56 is to collect the sound information emitted by the user when answering questions, and this sound information is collected through a microphone. When the user answers questions independently or answers questions incorrectly, the small screen 51 in the display module 5 will display corresponding prompt information. It is worth noting that the first feedback information of the reinforcement learning module 42 also includes the collection of sound information emitted by the user when answering questions. This design can not only help the system understand the user's learning status more comprehensively, but also provide data support for advanced functions such as speech recognition and sentiment analysis.

[0054] The fifth aspect of the present invention focuses on the function of the voice evaluation module 53. This module collects the user's voice through a microphone, analyzes the voice, and evaluates the energy or emotion the user puts into answering each question. When the user completes the exercises corresponding to all sub-learning points, the system calculates the energy and emotion the user puts into each sub-learning point, thereby obtaining the user's attitude information towards each sub-learning point. This design can help the system better understand the user's learning tendencies and difficulties, thereby providing more accurate learning suggestions.

[0055] Next, let us learn more about the method aspect of the present invention. The present invention provides a method for a personalized education platform based on artificial intelligence, which includes the following key steps:

[0056] Step S1: First, authenticate the user. After obtaining the user's authentication information, the system retrieves the user's historical learning data from the first database 11 based on the information and transmits the data to the target generation module 3. This step lays the data foundation for subsequent personalized learning.

[0057] Step S2: The goal generation module 3 selects points of interest according to the user's needs and determines the knowledge points corresponding to these points of interest. Then, the goal generation module 3 guides the user to generate specific learning points. The purpose of this step is to ensure that the learning plan not only meets the user's interests, but also meets the requirements of the knowledge system.

[0058] Step S3: The system determines the interest point selected by the user as an interest sub-learning point, and displays it through the plurality of small screens 51 of the display module 5. The displayed content will be organized according to the resource classification corresponding to the sub-learning point to ensure the clarity and order of the information.

[0059] Step S4: After the sub-learning point is displayed, the user can operate through the answer input module 52 on the display module 5. If the user chooses to answer the question independently, the answer input module 52 will generate first feedback information based on the user's answer. If the user answers the question incorrectly, the system will directly give the correct answer and generate second feedback information. This instant feedback mechanism can help users understand their learning effects in a timely manner and correct mistakes quickly.

[0060] Step S5: After the user completes answering the questions, the answer input module 52 generates a learning plan for the user through the plan formulation module 4. This learning plan is formulated based on the user's answering performance and learning needs, and can provide the user with more personalized learning guidance.

[0061] Step S6: The system generates exercises according to the learning plan completed by the user and displays these exercises. The user can choose whether to trigger these exercises. If the user chooses to trigger, the exercises will be displayed through the display module 5, and the user can choose the order of doing the exercises. If the user does not make a choice, the system will generate exercises in a preset random order. This flexible design respects the user's learning autonomy and ensures the systematic nature of learning.

[0062] Step S7: After the user completes the exercises in step S6, the system will generate a learning effect graph. When the user completes all the exercises in each sub-learning point, the system will generate a total learning effect graph. This visual display of learning effects can help users intuitively understand their learning progress and results and stimulate learning motivation.

[0063] In one embodiment of the present invention, step S2 also includes a keyword screening process. Several sub-learning points are preset in the storage module 1, and the system will use the keyword screening module to select words for each sub-learning point. Then, the system will summarize the sub-learning points according to the set keywords to obtain the learning knowledge points corresponding to the user. In the keyword screening module, the system will conduct in-depth mining based on the keywords contained in the sub-learning points, extract the core elements in the sub-learning points, and organize and display them through the display module 5. This keyword-based knowledge point extraction and organization method can help users understand the knowledge structure more clearly and improve learning efficiency.

[0064] In another embodiment, the sub-learning point output process of step S3 is more refined. The system will divide the sub-learning points into several preset modules based on the results of the keyword screening module. Each module has a corresponding sub-learning point, and each sub-learning point divides the data in the form of keywords. These divided data will be further allocated to the designated sub-learning points, and finally form a knowledge system with clear hierarchy and structure. This multi-level knowledge organization method can help users better understand the connection between knowledge and form a systematic cognitive structure.

[0065] In a preferred embodiment of the present invention, the answer input module 52 of step S4 provides multiple answer input modes to adapt to the learning habits of different users and different types of learning content:

[0066] The first method is to play the questions by voice, and the user will answer them by voice after listening to them. This method is particularly suitable for language learning and listening training.

[0067] The second method is to display the questions on the small screen 51 in the display module 5, and the user answers by selecting the answers. This method is suitable for multiple-choice questions and quick tests.

[0068] The third method is also to display the question on the small screen 51, but the user needs to write the answer by hand. This method is suitable for questions that require detailed derivation or expression, such as mathematical calculations or composition questions.

[0069] The fourth method also uses the small screen 51 to display the question, but the user needs to input the answer through the keyboard. This method is suitable for programming questions or questions that require accurate input.

[0070] Regardless of which method the user chooses to answer the question, the system will generate the first feedback information and the second feedback information and output them through voice. When the user answers the question incorrectly, the reinforcement learning module 42 will automatically start the question-and-answer session and output the question through voice. The system will judge whether the user's voice answer is correct based on the user's voice answer. If the user continues to answer incorrectly, the system will evaluate the user's answer based on the user's voice to determine the user's specific difficulties in understanding and mastering the knowledge points. This diversified answering method and intelligent feedback mechanism can comprehensively enhance the user's learning experience and adapt to different types of learning content and personal learning preferences.

[0071] In another embodiment of the present invention, step S5 involves a more complex learning material retrieval and question display mechanism. Specifically, when the user completes a round of answering questions, the system will automatically retrieve the learning materials of the next sub-learning point according to the user's answering situation. These materials will be displayed on the small screen 51, and the user can independently select the questions of interest to study. If the user does not make an independent choice, the system will display the questions in a pre-set order. This design not only ensures the continuity of learning, but also gives users a certain degree of autonomy in choice, which can effectively improve the pertinence and interest of learning.

[0072] It is worth noting that if the user makes an error during the operation, the system will not simply give the correct answer, but will correct the user's error in a targeted manner through the question bank in the reinforcement learning module 42. For example, if the user makes a calculation error when solving a math problem, the system may provide a series of similar calculation problems with increasing difficulty to help the user consolidate basic calculation skills. This error correction mechanism can not only help users overcome specific difficulties in learning, but also cultivate users' self-correction ability.

[0073] An important feature of the present invention is its adaptive learning algorithm. The algorithm continuously optimizes the learning path and content based on the user's learning data. Specifically, the system calculates the user's mastery of each knowledge point according to the following formula:

[0074] M=(C*W1+T*W2+A*W3) / (W1+W2+W3)

[0075] Among them, M represents the degree of mastery, C represents the correct rate, T represents the average answering time, A represents the accuracy (taking into account the situation where some answers are correct), and W1, W2, and W3 are the corresponding weights respectively.

[0076] Preferably, the values ​​of W1, W2, and W3 can be adjusted according to different subjects and learning stages. For example, for basic math questions that require quick responses, the value of W2 may be larger; while for science questions that require in-depth thinking, the values ​​of W1 and W3 may be larger. This flexible weight setting enables the system to more accurately evaluate the user's learning status.

[0077] In actual application, the system will compare the M value with the preset threshold. If the M value is lower than the threshold, the system will increase the number of exercises for the relevant knowledge points; if the M value is higher than the threshold, the system will reduce the number of exercises for the relevant knowledge points and introduce more difficult questions. This dynamic adjustment mechanism ensures that the learning content always stays within the user's zone of proximal development, and will neither lose its challenge because it is too simple, nor dampen the enthusiasm for learning because it is too difficult.

[0078] The present invention also introduces sentiment analysis technology to further optimize the learning experience. The speech evaluation module 53 can not only recognize the user's speech content, but also analyze the user's tone and emotions. For example, the system may detect the following emotional indicators:

[0079] E=(V*0.3+P*0.3+I*0.4)*100%

[0080] Among them, E represents the emotion index, V represents the volume change, P represents the pitch change, and I represents the speech speed change. The weights of these indicators can be adjusted according to actual conditions.

[0081] Preferably, when the E value is lower than 30%, the system will judge that the user may be in a depressed or fatigued state, and will automatically adjust the learning content, such as inserting some relaxing and interesting quizzes or encouraging voice prompts. When the E value is higher than 80%, the system will judge that the user is in an excited or highly focused state, and more challenging learning content may be provided. This emotional perception mechanism can make the learning process more humane and improve the user's learning enthusiasm and persistence.

[0082] Another innovative point of the present invention is the introduction of a collaborative learning function. The system will recommend suitable learning partners based on the user's learning progress and interests. For example, for the same knowledge point, the system may pair user A, who has a higher level of mastery, with user B, who has a lower level of mastery, and encourage them to learn interactively through online discussions or question-and-answer sessions. This peer-to-peer learning model can not only promote the consolidation and dissemination of knowledge, but also cultivate users' communication and expression skills.

[0083] In practical applications, the personalized education platform of the present invention can be widely used in various educational scenarios. For example, in K12 education, the platform can help students better understand and master textbook knowledge, while cultivating autonomous learning ability. In higher education, the platform can assist students in in-depth learning and research of professional knowledge. In the field of vocational education, the platform can provide students with targeted skills training and qualification certification preparation.

[0084] In general, the artificial intelligence-based personalized education platform and method provided by the present invention have the following advantages:

[0085] First, by combining data analysis, artificial intelligence and educational psychology, the present invention can provide each user with a truly personalized learning experience. The system not only considers the user's knowledge level and learning progress, but also pays attention to the user's interests, learning style and emotional state, so as to develop the most suitable learning plan for the user.

[0086] Secondly, the reinforcement learning mechanism and real-time feedback system of the present invention can help users master knowledge points more quickly and effectively. By correcting mistakes and strengthening weak links in a timely manner, users can make the greatest learning progress in the shortest time.

[0087] Thirdly, the multimodal learning interface (including text, voice, image, etc.) of the present invention can meet the learning preferences of different users and improve the interest and efficiency of learning. In particular, the voice interaction function is not only convenient for users to operate, but also can effectively train language skills.

[0088] Finally, the collaborative learning function and sentiment analysis technology of the present invention make the learning process more humane and social, which helps to improve the learning enthusiasm and persistence of users. This is not only conducive to the acquisition of knowledge, but also can cultivate users' social skills and emotional intelligence.

[0089] In order to verify the superiority of the present invention, we conducted a series of comprehensive tests on the AI-based personalized education platform. These tests are designed to evaluate the platform's performance in terms of learning effects, user experience, and system performance. The following are our test methods, results, and analysis.

[0090] First, we selected a group of 100 students of different ages and learning backgrounds as test subjects. These students were randomly divided into two groups: the experimental group used our personalized education platform, and the control group used a traditional online learning platform. The test lasted for three months, during which we closely monitored the students' learning progress, performance changes, and learning experience.

[0091] We focused on the following key indicators:

[0092] 1. Learning efficiency: measured by comparing the number of knowledge points mastered by students per unit time.

[0093] 2. Knowledge retention: Assess students’ memory of learned knowledge through regular retesting.

[0094] 3. Learning interest: Assess students’ learning enthusiasm through questionnaires and platform usage frequency.

[0095] 4. Degree of personalization: The degree to which the learning content and paths provided by the assessment system are differentiated for different students.

[0096] 5. Error correction capability: evaluates how well the system identifies and corrects student errors.

[0097] Here are our test results:

[0098] Learning efficiency 8.5 knowledge points / hour 5.2 Knowledge points / hour 63.5％ Knowledge retention rate 85％ 62％ 37.1％ Learning interest 8.7 / 10 6.2 / 10 40.3％ Degree of personalization 92％ 45％ 104.4％ Error correction success rate 89％ 60％ 48.3％

[0099] These test results fully demonstrate the superiority of our personalized education platform. Let’s take a deeper look at what these data mean:

[0100] The significant improvement in learning efficiency (63.5%) reflects the core advantage of our platform. This is due to our intelligent algorithm that can accurately identify each student's knowledge blind spots and provide targeted learning content. For example, when the system finds that a student performs poorly in algebraic operations, it will automatically add relevant exercises and adjust the difficulty gradient to ensure that students overcome the difficulties in the shortest time.

[0101] The improvement in knowledge retention rate (37.1%) reflects the effectiveness of our platform's reinforcement learning mechanism. Through regular review and intelligent reminders, students can better consolidate what they have learned. A typical case is that the system automatically pushes review content at the time when students are most likely to forget (such as the 7th day and the 30th day after learning). This learning method based on the forgetting curve greatly improves the long-term retention rate of knowledge.

[0102] The increase in learning interest (40.3%) is one of the results we are most proud of. This is mainly due to the platform's gamification design and sentiment analysis technology. For example, the system dynamically adjusts the difficulty and form of learning content according to the student's emotional state, inserting some relaxing and interesting quizzes when students feel tired, and providing more challenging questions when students are in high spirits. This humanized design has greatly improved students' enthusiasm for learning.

[0103] The huge advantage of personalization (104.4% improvement) fully demonstrates the core competitiveness of our platform. Traditional platforms often adopt a one-size-fits-all teaching method, while our system can tailor a learning plan for each student. For example, for the same knowledge point of photosynthesis, the system will provide more detailed explanations of chemical reactions for students with a strong science background, and provide more biological application cases for students with a liberal arts background, thereby ensuring that each student can understand and master the knowledge in the way that best suits them.

[0104] The improvement in the error correction success rate (48.3%) highlights the intelligent tutoring capabilities of our platform. The system can not only accurately identify students' mistakes, but also analyze the causes of the errors and provide targeted correction solutions. For example, when a student is found to always ignore special cases when solving geometry problems, the system will automatically generate a series of questions that include special cases to help students develop the habit of comprehensive thinking.

[0105] These test results not only prove that our platform is superior to traditional learning methods in all aspects, but more importantly, they reveal the huge potential of artificial intelligence in the field of education. Our platform uses technologies such as intelligent algorithms, big data analysis, and emotional computing to achieve true teaching in accordance with students’ aptitude and provide each student with the most optimized learning experience.

[0106] In the best implementation, we found that the combination of reinforcement learning algorithms and sentiment analysis technology is the most effective. Specifically, the system analyzes students' learning behaviors and emotional states in real time and dynamically adjusts the learning content and difficulty. For example, when it detects that a student has answered multiple questions correctly in a row and is in high spirits, the system will increase the difficulty appropriately to keep the learning challenging; when it is found that the student's answering speed has slowed down and his mood has dropped, the system will insert some relaxing review content to help the student rebuild his confidence.

[0107] This intelligent + emotional teaching model not only greatly improves learning efficiency, but also significantly improves students' learning interest and self-confidence. In a three-month test, the average score of the student group using this model increased by 32%, far higher than other implementation plans. What's more worth mentioning is that this group of students also has the highest learning enthusiasm, with an average daily active learning time of 45 minutes more than other groups.

[0108] In general, these test results not only verify the technical advantages of our platform, but also confirm the huge potential of personalized and intelligent education. We believe that with the continuous advancement of technology and the continuous accumulation of data, this AI-based education model will bring unprecedented learning experience to more students and truly realize the educational ideal of teaching students in accordance with their aptitude.

[0109] It should be noted that the above embodiments are only preferred implementations of the present invention and are not intended to limit the protection scope of the present invention. Those skilled in the art can make equivalent substitutions or changes based on the technical solutions and inventive concepts of the present invention, and these equivalent substitutions or changes should all be included in the protection scope of the present invention.

Claims

1. Personalized education platform based on artificial intelligence, including: A storage module, the storage module includes a first database and a second database, wherein: the first database stores user information, user's historical learning data and user identity authentication information; the second database stores learning materials corresponding to each knowledge point, knowledge points corresponding to each learning material and resource classification information corresponding to each learning material; an information acquisition module, the information acquisition module stores a number of test questions and is used to obtain the user's identity authentication information, retrieve the user's historical learning data from the storage module according to the user's identity authentication information and transmit the historical learning data to the target generation module; a target generation module, for users to perform target generation training; a program formulation module, for generating a sequence of knowledge points of interest according to the knowledge points of interest set by the user, and the knowledge points of interest in the sequence of knowledge points of interest are divided into a number of preset sub-learning points, and an interest value determination unit is provided in the program formulation module; a display module, the display module and the The target generation module is connected, the display module includes a plurality of small screens, and the preset plurality of sub-learning points are displayed through the plurality of small screens, the display module also includes an answer input module, a voice evaluation module, a timing module and a feedback unit; it is characterized in that the program formulation module also includes a reinforcement learning module, the reinforcement learning module includes a first feedback unit, the first feedback unit generates a first feedback information when the user completes a test question through the answer input module; the voice evaluation module evaluates the user's oral answer and outputs a second feedback information, and the second feedback information is a voice with a score; when several test questions of the sub-learning point are answered, the program formulation module generates a learning plan for the user, the learning plan sets an exercise question option after the test question through the display module, and the exercise question option is started through the feedback unit when it is triggered; if the exercise question is triggered, the learning effect diagram is output after the exercise question is practiced.

2. The personalized education platform based on artificial intelligence according to claim 1, characterized in that: The storage module also includes the preference level corresponding to each small screen in the display module, and each small screen is set as a small screen to be displayed; when the user inputs an answer on the first small screen, the information acquisition module will store the user's preference for the first small screen; a timing module is provided in the information acquisition module, and the timing module calculates the user's question-solving time. When the user is solving the problem, the timing module records the user's question-solving time for each problem in real time, and transmits the user's question-solving time to the first database of the storage module for storage; a preference determination unit is provided in the program formulation module. When the user completes the exercises of the first small screen through the display module, the exercises corresponding to the first small screen are set to be input through the display module according to the user's preference value for the first small screen; the exercises corresponding to the first small screen are set to include a number of questions to be practiced. When the user completes all the exercises corresponding to the first small screen, the program formulation module displays the exercises corresponding to the first small screen, and confirms the user's preference for the exercises corresponding to the entire first small screen according to the time taken to answer the exercises of the first small screen completed by the user.

3. The personalized education platform based on artificial intelligence according to claim 1, characterized in that: The first feedback information in the reinforcement learning module includes recording the correct or incorrect user operations or the user's irrelevant answers to the questions; when the user makes an error in the operation, the mini program is used again to retrieve the relevant error points and relevant test questions; if the user makes an irrelevant answer to the question, relevant questions will be automatically asked during the setting, and then the voice evaluation module will be used for testing to reinforce the user's relevant error points.

4. The personalized education platform based on artificial intelligence according to claim 1, characterized in that: The display module is also provided with an auxiliary unit, which collects sound information emitted by the user when answering questions, and the sound information is collected through a microphone; if the user answers the questions independently or answers the questions incorrectly, the small screen in the display module displays prompt information, and the first feedback information of the reinforcement learning module also contains the collection of sound information emitted by the user when answering questions.

5. The personalized education platform based on artificial intelligence according to claim 1, characterized in that: The voice evaluation module collects the user's voice through a microphone, and evaluates the energy or emotion the user spends on a question when answering the question based on the collected user's voice; when the user completes the exercises corresponding to all the sub-learning points, the user's energy and emotion in answering the sub-learning points are calculated to obtain the user's information corresponding to the sub-learning points.

6. The method of the personalized education platform based on artificial intelligence according to claim 1, characterized in that: The steps include: Step S1: firstly authenticate the user, obtain the user's authentication information, and then fetch the user's historical learning data from the first database through the user's authentication information, and transmit the user's historical learning data to the target generation module; Step S2: The target generation module selects points of interest according to the needs of the user, selects the knowledge points corresponding to the points of interest of the user, and guides the user to generate learning points through the target generation module; Step S3: Determine the interest point selected by the user as an interest sub-learning point, and display the sub-learning point through a display module, wherein a plurality of small screens are provided in the display module, and the resource classification corresponding to the sub-learning point is displayed through the small screens; Step S4: According to the needs of the user, after the sub-learning point is displayed, the corresponding operation is performed through the answer input module on the display module. If the user answers the question independently, the answer input module outputs the first feedback information according to the user's answer. If the user answers the question incorrectly, the system directly gives the answer and outputs the second feedback information. Step S5: After the user answers the question through the answer input module, the answer input module generates a learning plan for the user through the plan formulation module; Step S6: Then, according to the learning plan completed by the user, exercises for the user are formed, and the exercises completed by the user are displayed. If the user chooses to trigger, the exercises are displayed through the display module, so that the user can make independent choices. If the user does not choose, the exercises are generated according to the random selection of the system; Step S7: After the user completes the exercises in step S6, a learning effect graph is formed, and then the user's overall learning effect graph is formed after the exercises in each sub-learning point are completed.

7. The method of personalized education platform based on artificial intelligence according to claim 6, characterized in that: In the step S2, a number of sub-learning points are provided in the storage module, and a keyword screening module is used to select words for each sub-learning point, and then the sub-learning points are summarized according to the set keywords to obtain the learning knowledge points corresponding to the user; in the keyword screening module, mining is performed according to the keywords contained in the sub-learning points, and then the elements contained in the sub-learning points are extracted according to the keywords, and organized through the display module.

8. The method of personalized education platform based on artificial intelligence according to claim 6, characterized in that: When the sub-learning points of step S3 are output, the sub-learning points are divided into several preset modules according to the keywords in the keyword screening module. At the same time, each module has corresponding sub-learning points for division. Each sub-learning point divides the data in the form of keywords, and then divides the divided data into designated sub-learning points, thereby generating several sub-learning points.

9. The method of personalized education platform based on artificial intelligence according to claim 6, characterized in that: In step S4, the answer input module includes multiple answer input methods: the first method is to play through voice, the user listens to the question and answers it by voice; the second method is to use the small screen in the display module for display, and display the corresponding question, and the user selects the answer to answer the question; the third method is to use the small screen in the display module for display, and display the corresponding question, and the user answers by handwriting; the fourth method is to use the small screen in the display module for display, and display the corresponding question, and display the answer the user wants in the system in the form of a program, so as to input it by keyboard; when answering questions using any of the four methods, the system will obtain a first feedback information, and will also generate a second feedback information and output it by voice; when the user answers the question incorrectly, the system automatically asks and answers questions through the reinforcement learning module, and outputs the question sentence through voice, and then judges whether the user answers correctly according to the user's voice. When the user keeps answering the question incorrectly, the system will evaluate the user's question according to the user's voice.

10. The method for personalized education platform based on artificial intelligence according to claim 6, characterized in that: In step S5, after the user finishes answering the questions, the system will automatically retrieve the learning materials for the next sub-learning point based on the user's answers, and then display the questions on a small screen, so that the user can make independent choices and then display the corresponding questions. If there is no independent choice, the system will display the planned questions. If the user makes an error in operation, the system will correct the user's error by reinforcing the test questions in the learning process.

Images