Learning recommendation method and device based on education element universe and knowledge tracking

Abstract

The invention discloses a learning recommendation method and device based on education meta universe and knowledge tracking, and relates to the technical field of meta universe learning application. The method comprises the following steps: data collection and arrangement: continuously collecting learning behavior data of students on a platform by using various sensors and interactive interfaces in the education universe platform, collecting basic information of the students at the same time, and carrying out classified storage on the collected data according to a time sequence and a data type. Through comprehensive collection of student learning behavior data, basic information and intelligent interaction feedback, in combination with a knowledge tracking model and a comprehensive evaluation method, student knowledge states can be accurately evaluated, recommendation contents are screened and individually sorted according to evaluation results, students can obtain learning resources most suitable for themselves, and learning efficiency is improved. And the recommended contents are screened and individually sorted according to the evaluation result, so that the students can obtain learning resources most suitable for themselves, and the learning efficiency is remarkably improved.

Description

Technical Field

[0001] This invention belongs to the field of metaverse learning application technology, and in particular relates to a learning recommendation method and device based on the educational metaverse and knowledge tracking. Background Technology

[0002] With the rapid development of information technology, the education sector is undergoing profound changes. Traditional teaching models are gradually shifting towards digitalization and intelligence to meet the increasingly diverse learning needs of students. The Education Metaverse, as an emerging educational form, provides students with an immersive and interactive learning experience by constructing a learning environment that blends virtual and reality, greatly stimulating students' learning interest and participation. Within the Education Metaverse, students can participate in various learning activities in an immersive way, such as virtual experiments, historical scene recreations, and scientific exploration, breaking through the limitations of time and space.

[0003] However, within the rich learning resources of the educational metaverse, how to accurately recommend learning content suitable for each student's learning progress and knowledge mastery has become a critical issue that urgently needs to be addressed. Students' learning foundations, abilities, and interests vary significantly. Using uniform learning content and teaching methods makes it difficult to meet the individualized needs of each student, easily leading to learning difficulties or a lack of motivation for some.

[0004] Knowledge tracking technology, as a method for modeling and predicting students' knowledge status, is gradually being applied in the field of education. It infers students' mastery of various knowledge points and their learning trajectory by analyzing student learning behavior data, such as answer patterns, study time, and interactions with learning resources. However, existing knowledge tracking technologies still have shortcomings in accuracy and timeliness, making it difficult to comprehensively and in real-time reflect students' knowledge changes. Furthermore, the application of knowledge tracking technology in conjunction with the educational metaverse is still in the exploratory stage. How to effectively integrate the two to achieve accurate learning recommendations and improve students' learning outcomes is a current research hotspot and challenge in the field of educational technology.

[0005] Furthermore, current learning recommendation systems lack effective intelligent interaction and comprehensive evaluation mechanisms. During the learning process, the interaction between students and the system is relatively simple, making it impossible to obtain students' learning feelings, questions, and needs in a timely manner. This results in a low degree of matching between recommended content and students' actual needs. Moreover, the evaluation of students' learning outcomes is often limited to single indicators such as exam scores, making it difficult to comprehensively and objectively assess students' knowledge mastery and learning abilities, and thus failing to provide sufficient basis for personalized learning recommendations.

[0006] To address these issues, we provide a learning recommendation method and apparatus based on the educational metaverse and knowledge tracing. Summary of the Invention

[0007] The purpose of this invention is to provide a learning recommendation method and apparatus based on the educational metaverse and knowledge tracking, which solves the problem that existing learning recommendation systems lack effective intelligent interaction and comprehensive evaluation mechanisms.

[0008] To solve the above-mentioned technical problems, the present invention is achieved through the following technical solution.

[0009] This invention is a learning recommendation method based on the Education Metaverse and knowledge tracking, comprising the following steps: Data collection and organization: Utilizing various sensors and interactive interfaces within the Education Metaverse platform, continuously collect student learning behavior data on the platform, along with basic student information. The collected data is then categorized and stored according to time sequence and data type. Knowledge status assessment: The organized data is input into a pre-built knowledge tracking model. This model employs deep learning algorithms, using multi-layer neural networks to analyze and process the data. Based on students' answers to exercises in various subjects, the model comprehensively calculates the probability of students mastering each knowledge point and outputs the corresponding knowledge status vector. Recommended content selection: Based on the student's mastery probability of each knowledge point derived from the knowledge status assessment, recommended content is selected from the learning resource library of the Education Metaverse platform. The system filters and recommends content. For knowledge points with a low probability of mastery, it selects related basic knowledge explanations; for knowledge points with a high probability of mastery, it selects extended learning content. This is then personalized by combining students' learning interest tags, preferences for different types of learning content, and learning habits. Machine learning algorithms are used to personalize the recommended content, updating and adjusting it in real time. During the student's learning process, real-time monitoring technology continuously collects new learning behavior data. Once new data is generated, it is immediately input into the knowledge tracking model to update the model's assessment of the student's knowledge status. Based on the updated knowledge status, the system re-filters and re-ranks the recommended content. If a student shows strong interest in a particular extended knowledge point, the system promptly adjusts the recommendation strategy to provide the student with more learning resources related to that extended knowledge point.

[0010] The present invention is further configured such that the learning behavior data includes, but is not limited to, learning paths in virtual scenes, interactive operations with virtual teaching aids, answering questions, learning duration, content posted in discussion forums, and collection and marking of learning content.

[0011] The present invention is further configured such that a comprehensive evaluation method is introduced into the knowledge status assessment, which scores and evaluates the student's knowledge status from the dimensions of accuracy, stability and application ability of knowledge mastery, and forms a comprehensive knowledge status assessment report.

[0012] The present invention is further configured such that, in the data collection and organization process, intelligent voice assistants and intelligent customer service interactive tools are used to proactively inquire about students' feelings, doubts, and evaluations of the learning content during the learning process, thereby collecting student feedback data.

[0013] The present invention is further configured such that big data storage and management technology is used in the data collection and organization process to classify and store the collected data, and to establish an index for quick query and retrieval, thereby ensuring the integrity and traceability of the data.

[0014] The present invention is further configured such that the answer situation in the knowledge state assessment includes the problem-solving approach, analyzing the problem-solving steps and reasoning expressed by the student through natural language processing technology, analyzing the error type to determine whether it is a conceptual misunderstanding, calculation error or improper application of method, the duration and number of times the relevant knowledge explanation video is watched, analyzing the student's learning focus and attention to difficulties in combination with the pause, rewind and fast forward operations during the video viewing process, as well as the comments on knowledge points in the discussion area, assessing the student's mastery level and doubts through sentiment analysis and semantic understanding, comprehensively calculating the probability of the student's mastery of each knowledge point, and outputting the corresponding knowledge state vector.

[0015] The present invention is further configured such that the knowledge tracking model in the knowledge state assessment adopts a recurrent neural network and a long short-term memory network structure to perform in-depth analysis and processing of data, and has a real-time dynamic update mechanism, continuously optimizing the model parameters as new data is continuously input.

[0016] The learning device based on the Education Metaverse and knowledge tracking includes hardware modules, software modules, and a database design. The hardware modules include servers, sensor devices, and intelligent interactive terminals. Equipped with multi-core processors, memory processing devices, and high-speed storage devices, it supports the entire learning recommendation system, handles and stores large amounts of student data, and ensures stable operation under high concurrency. Sensor devices are deployed in the virtual scene of the Education Metaverse platform. High-definition cameras capture students' facial expressions and body movements to analyze their learning emotions and concentration. Motion capture devices accurately record students' interactions with the virtual environment, including operational steps in the virtual laboratory and hand-raising actions in the virtual classroom. Microphones collect students' voice information for voice interaction and recognition, facilitating the collection of student questions and feedback. The intelligent interactive terminals used by students include virtual reality headsets, augmented reality glasses, and smart tablets with high-resolution displays, high-performance processors, and graphics processing units to ensure the smooth operation of the Education Metaverse platform.

[0017] The present invention is further configured such that the software module includes a data processing module, a knowledge tracking module, a recommendation strategy generation module, and a recommendation result display module. The data processing module receives data transmitted from various sensors, cleans the data, removes noise, fills in missing values, and performs preprocessing, including data standardization and normalization, and extracts key features from the data. The processed data is then transmitted to the knowledge tracking module, which uses natural language processing and sentiment analysis tools to process the text data collected by intelligent interaction. The knowledge tracking module has a built-in knowledge tracking model based on a deep learning algorithm. It receives the processed data transmitted from the data processing module, learns and trains the data through a multi-layer neural network, calculates the probability of students mastering each knowledge point and the knowledge state vector, and transmits the calculation results to the recommendation strategy generation module. This, combined with intelligent... The system enables interactive feedback and optimizes the knowledge status assessment process. The recommendation strategy generation module, based on the student's knowledge status output by the knowledge tracking module and combined with the learning resource library of the Education Metaverse platform, formulates personalized learning recommendation strategies. For knowledge points where students have a lower level of mastery, relevant basic knowledge explanations, case studies, and practice questions are selected from the resource library. For knowledge points where students have a higher level of mastery, extended learning content is selected. Combining students' learning interests and habits, machine learning algorithms are used to optimize and rank the recommended content. Based on intelligent interactive feedback and comprehensive evaluation results, the recommendation strategy is dynamically adjusted. The recommendation result display module presents the recommended content generated by the recommendation strategy generation module in an intuitive way on the Education Metaverse platform, including 3D pop-ups and immersive task lists, facilitating students' access to and learning of the recommended content.

[0018] The invention is further configured such that the database design includes a student information database, a learning behavior database, a learning resource database, and a feedback database. The student information database stores basic student information, including age, learning stage, subject foundation, learning interest tags, and learning habits, providing basic data for personalized learning recommendations. The learning behavior database records all learning behavior data of students on the Education Metaverse platform, including learning paths, interactions with virtual teaching aids, answering questions, learning duration, discussion forum posts, favorites, and tagging information, used for knowledge status assessment and the formulation of learning recommendation strategies. The learning resource database stores various learning resources in the Education Metaverse platform, including basic knowledge explanation materials, case analysis materials, practice questions, advanced application cases, and cutting-edge academic research materials, categorized and stored according to subject, knowledge point, and difficulty level to facilitate the selection of recommended content. The feedback database stores feedback information provided by students through intelligent interactive tools, including learning feelings, questions, and evaluations of recommended content, used to optimize recommendation strategies and knowledge tracking models.

[0019] The present invention has the following beneficial effects.

[0020] 1. This invention, by comprehensively collecting student learning behavior data, basic information, and intelligent interactive feedback, combined with a knowledge tracking model and comprehensive evaluation method, can accurately assess students' knowledge status. Based on the assessment results, it filters and personalizes recommended content, enabling students to obtain the most suitable learning resources and avoiding learning content that is too difficult or too easy, thereby significantly improving learning efficiency. For students with weak foundations, the system can accurately recommend reinforcement practice materials for basic knowledge points, helping them quickly make up for knowledge gaps and improve academic performance. The application of intelligent interactive tools enables natural interaction between students and the system. Students can provide feedback on their learning experience and ask questions at any time. The system adjusts its recommendation strategy in real time based on this feedback, providing learning resources that better meet students' needs, enhancing students' initiative and participation in learning, and optimizing the learning experience.

[0021] 2. This invention scores and evaluates students' knowledge status from multiple dimensions, including accuracy, stability, and application ability, generating a comprehensive knowledge status assessment report. This comprehensive evaluation method can more objectively reflect students' learning situation, providing a more sufficient basis for personalized learning recommendations. It also guides students to focus on the all-round development of their knowledge and abilities, avoiding the sole pursuit of exam scores. The system monitors students' learning behavior in real time, updating the knowledge tracking model and recommendation strategy immediately upon the generation of new data. The system can promptly capture changes in students' learning interests and knowledge acquisition progress, providing students with the latest and most suitable learning resources, adapting to the dynamic changes in student learning. The Education Metaverse Platform breaks through the limitations of time and space. Combined with the learning recommendation method and device of this invention, students in different regions and with different learning conditions can access personalized learning resources. This helps to narrow the gap in educational resources, promotes educational equity, and allows more students to benefit from high-quality educational services. Attached Figure Description

[0022] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below.

[0023] Figure 1 This is a schematic diagram of a learning recommendation method and device based on the educational metaverse and knowledge tracing.

[0024] Figure 2 This diagram shows the hardware modules of a learning recommendation method and device based on the educational metaverse and knowledge tracing.

[0025] Figure 3 This diagram illustrates the server composition in a learning recommendation method and device based on the educational metaverse and knowledge tracing.

[0026] Figure 4 This diagram illustrates the composition of software modules in a learning recommendation method and device based on the educational metaverse and knowledge tracing.

[0027] Figure 5 This is a diagram illustrating the composition of the database design in a learning recommendation method and device based on the educational metaverse and knowledge tracing. Detailed Implementation

[0028] The technical solutions of the present invention will be described below with reference to the accompanying drawings. The described embodiments are only some embodiments of the present invention, and not all embodiments.

[0029] Example 1

[0030] Please see Figure 1-5 This invention is a learning recommendation method based on the Education Metaverse and knowledge tracking, comprising the following steps: data collection and organization, continuously collecting student learning behavior data on the platform using various sensors and interactive interfaces in the Education Metaverse platform, while also collecting students' basic information, and classifying and storing the collected data according to time sequence and data type; knowledge status assessment, inputting the organized data into a pre-built knowledge tracking model, which uses deep learning algorithms to analyze and process the data through multi-layer neural networks, comprehensively calculating the probability of students mastering each knowledge point based on their answers to exercises in various subjects, and outputting the corresponding knowledge status vector; and content recommendation selection, based on the probability of students mastering each knowledge point obtained from the knowledge status assessment, selecting recommended content from the learning resource library of the Education Metaverse platform. The system filters and recommends content, selecting basic explanations for knowledge points with low probabilities of mastery and extended learning content for knowledge points with high probabilities of mastery. This is then personalized by combining students' learning interest tags, preferences for different types of learning content, and learning habits. Machine learning algorithms are used to personalize the recommended content, updating and adjusting it in real time. During the student's learning process, real-time monitoring technology continuously collects new learning behavior data. Once new data is generated, it is immediately input into the knowledge tracking model to update the model's assessment of the student's knowledge status. Based on the updated knowledge status, the system re-filters and re-ranks the recommended content. If a student shows strong interest in a particular extended knowledge point, the system promptly adjusts the recommendation strategy to provide the student with more learning resources related to that extended knowledge point.

[0031] Example 2

[0032] Please see Figure 1-5Building upon Example 1, learning behavior data includes, but is not limited to, learning paths in virtual scenarios, interactions with virtual teaching aids, quiz responses, learning duration, discussions in the discussion forum, and the collection and marking of learning content. A comprehensive evaluation method is introduced in the knowledge status assessment, scoring and evaluating students' knowledge status from the dimensions of accuracy, stability, and application ability, generating a comprehensive knowledge status assessment report. In data collection and organization, intelligent voice assistants and intelligent customer service tools are used to proactively inquire about students' feelings, questions, and evaluations of the learning content, collecting student feedback data. Big data storage and management technologies are employed in data collection and organization to classify and store the collected data, creating indexes for quick retrieval and access, ensuring data integrity and traceability. The knowledge state assessment includes the student's problem-solving approach, which is analyzed using natural language processing to interpret the steps and thought process. Error types are analyzed to determine whether they stem from misunderstanding of concepts, calculation errors, or improper application of methods. The assessment also includes the duration and number of times the student watched relevant knowledge explanation videos, analyzing their focus and areas of difficulty based on pause, rewind, and fast-forward actions during video viewing. Furthermore, the assessment includes comments made in the discussion forum regarding knowledge points. Sentiment analysis and semantic understanding are used to assess the student's mastery and areas of confusion. The overall probability of mastery for each knowledge point is calculated, and a corresponding knowledge state vector is output. The knowledge tracking model in the knowledge state assessment employs recurrent neural networks and long short-term memory network structures for deep data analysis and processing, with a real-time dynamic update mechanism that continuously optimizes model parameters as new data is input.

[0033] Example 3

[0034] Please see Figure 1-5This learning device, based on the educational metaverse and knowledge tracking, includes hardware modules, software modules, and a database design. The hardware modules include servers, sensor devices, and intelligent interactive terminals, equipped with multi-core processors, memory computing devices, and high-speed storage devices to support the entire learning recommendation system, handle and store large amounts of student data, and ensure stable operation under high concurrency. The sensor devices are deployed in the virtual scene of the educational metaverse platform. High-definition cameras are used to capture students' facial expressions and body movements to analyze their learning emotions and concentration. Motion capture devices accurately record students' interactions with the virtual environment, including activities in a virtual laboratory. The system guides students through steps and gestures such as raising their hands to speak in the virtual classroom. Microphones are used to collect students' voice information, enabling voice interaction and recognition to gather student questions and feedback. The intelligent interactive terminal for students consists of virtual reality headsets, augmented reality glasses, and smart tablets, equipped with high-resolution displays, high-performance processors, and graphics processing units to ensure smooth operation of the educational metaverse platform. Software modules include a data processing module, a knowledge tracking module, a recommendation strategy generation module, and a recommendation result display module. The data processing module receives data from various sensors, cleans the data, removes noise, fills in missing values, and performs preprocessing, including... After standardization and normalization, and extraction of key features from the data, the processed data is transmitted to the knowledge tracking module. Utilizing natural language processing and sentiment analysis tools, the module processes the text data collected through intelligent interaction. The knowledge tracking module, with its built-in deep learning-based knowledge tracking model, receives the processed data from the data processing module and learns and trains on the data through a multi-layer neural network. It calculates the probability of students mastering each knowledge point and the knowledge state vector, and transmits the calculation results to the recommendation strategy generation module. Combined with intelligent interaction feedback, the module optimizes the knowledge state assessment process. Based on the student knowledge state output by the knowledge tracking module and in conjunction with educational elements... The Metaverse platform's learning resource library develops personalized learning recommendation strategies. For knowledge points where students have a lower level of mastery, it selects relevant basic knowledge explanations, case studies, and practice questions from the resource library. For knowledge points where students have a higher level of mastery, it selects extended learning content. Combining students' learning interests and habits, machine learning algorithms are used to optimize and rank the recommended content. Based on intelligent interactive feedback and comprehensive evaluation results, the recommendation strategy is dynamically adjusted. The recommendation result display module presents the recommended content generated by the recommendation strategy generation module in an intuitive way on the Metaverse platform. Display methods include 3D pop-ups and immersive task lists, making it convenient for students to access and learn the recommended content.

[0035] Example 4

[0036] Please see Figure 1-5The database design includes a student information database, a learning behavior database, a learning resource database, and a feedback database. The student information database stores basic student information, including age, learning stage, subject foundation, learning interest tags, and learning habits, providing basic data for personalized learning recommendations. The learning behavior database records all learning behavior data of students on the Education Metaverse platform, including learning paths, interactions with virtual teaching aids, answering questions, learning time, discussion forum posts, favorites, and tagging information, used for knowledge status assessment and the formulation of learning recommendation strategies. The learning resource database stores various learning resources in the Education Metaverse platform, including basic knowledge explanations, case analysis materials, practice questions, advanced application cases, and cutting-edge academic research materials, categorized and stored according to subject, knowledge point, and difficulty level to facilitate the selection of recommended content. The feedback database stores feedback information provided by students through intelligent interactive tools, including learning feelings, questions, and evaluations of recommended content, used to optimize recommendation strategies and knowledge tracking models.

[0037] The preferred embodiments of the present invention disclosed above are only for the purpose of illustrating the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to specific implementation methods. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can better understand and utilize the present invention.

Claims

1. A learning recommendation method based on an educational meta-universe and knowledge tracking, characterized in that: Comprise the following steps: S1, data collection and arrangement, using various sensors and interactive interfaces in the education metaverse platform, continuously collect students' learning behavior data on the platform, collect students' basic information, and store the collected data in time sequence and data type; S2, knowledge state evaluation, input the arranged data into the pre-constructed knowledge tracking model, the model uses deep learning algorithm, analyzes and processes the data through multi-layer neural network, the model calculates the probability of students mastering each knowledge point according to the students' answer in each subject exercise, and outputs the corresponding knowledge state vector; S3, recommended content screening, according to the probability of students mastering each knowledge point obtained by knowledge state evaluation, screening recommended content from the learning resource library of education metaverse platform, for the knowledge point with low probability of mastering, screening related basic knowledge explanation materials, for the knowledge point with high probability of mastering, screening expansion learning content; S4, personalized sorting, combining students' learning interest labels, preferences for different types of learning content, and learning habits, using machine learning algorithm to sort the recommended content; S5, real-time update and adjustment, through real-time monitoring technology to continuously collect new learning behavior data of students, once new data is generated, input it into the knowledge tracking model immediately, update the evaluation of students' knowledge state by the model, and reselect and sort the recommended content according to the updated knowledge state, if the student shows strong interest in a certain expansion knowledge point, the system adjusts the recommendation strategy in time, and provides more learning resources about the expansion knowledge point for the student. 2.The learning recommendation method based on an educational meta-universe and knowledge tracking according to claim 1, characterized in that: The learning behavior data includes but is not limited to learning path in virtual scene, interaction with virtual teaching aid, answering situation, learning time, speech content in discussion area, collection and marking of learning content. 3.The learning recommendation method based on an educational meta-universe and knowledge tracking according to claim 1, characterized in that: The comprehensive evaluation method is introduced in the knowledge state evaluation, which scores and evaluates the students' knowledge state from the dimensions of accuracy, stability and application ability of knowledge mastery, and forms a comprehensive knowledge state evaluation report. 4.The learning recommendation method based on an educational meta-universe and knowledge tracking according to claim 1, characterized in that: In the data collection and arrangement, the intelligent voice assistant and intelligent customer service intelligent interaction tool are used to actively ask students about their feelings, doubts and evaluation of learning content in the learning process, and collect students' feedback data. 5.The learning recommendation method based on an educational meta-universe and knowledge tracking according to claim 1, characterized in that: In the data collection and arrangement, big data storage and management technology is used to store the collected data, establish index for quick query and call, and ensure the integrity and traceability of the data.

6. The learning recommendation method based on an educational meta-universe and knowledge tracking according to claim 1, characterized in that: The answering situation in the knowledge state evaluation includes problem solving ideas. The problem solving steps and idea elaboration input by the student are analyzed through natural language processing technology. The error type is analyzed. It is judged whether it is a concept understanding error, a calculation mistake or an improper method application. The length and number of times of watching related knowledge explanation videos are watched. The learning concentration and difficulty attention of the student are analyzed by analyzing the pause, playback and fast forward operations in the video watching process. The speech about the knowledge points in the discussion area is analyzed. The mastery degree and doubt points of the student are evaluated through sentiment analysis and semantic understanding. The mastery probability of the student for each knowledge point is calculated, and the corresponding knowledge state vector is output.

7. The learning recommendation method based on an educational meta-universe and knowledge tracking according to claim 1, characterized in that: The knowledge tracking model in the knowledge state evaluation adopts a recurrent neural network and a long short-term memory network structure to perform deep analysis and processing on the data. It has a real-time dynamic updating mechanism. With the continuous input of new data, the model parameters are continuously optimized.

8. Learning device based on educational metaverse and knowledge tracking, comprising hardware modules, software modules and database design, characterized by: The hardware module includes a server, a sensor device and an intelligent interaction terminal. It is equipped with a multi-core processor, a memory operation device and a high-speed storage device, which is used to carry the operation of the entire learning recommendation system, supports the processing and storage of a large amount of student data, and ensures the stable operation of the system under high concurrency; The sensor device is deployed in the virtual scene of the education meta-universe platform. A high-definition camera is used to capture the facial expressions and body movements of the student, analyze the learning emotions and concentration of the student, and accurately record the interactive actions of the student with the virtual environment, including the operation steps in the virtual laboratory and the hand-raising speaking actions in the virtual classroom. A microphone is used to collect the voice information of the student, realize voice interaction and voice recognition, and collect the questions and feedback of the student; The intelligent interaction terminal is a virtual reality headset, augmented reality glasses and a smart tablet device used by the student. It has a high-resolution display screen, is equipped with a high-performance processor and a graphics processing unit, and ensures the smooth operation of the education meta-universe platform.

9. The learning apparatus based on educational meta-universe and knowledge tracking according to claim 8, characterized in that: The software module includes a data processing module, a knowledge tracking module, a recommendation strategy generation module and a recommendation result display module; The data processing module receives the data transmitted by each sensor, cleans the data, removes noise data and fills in missing values, performs preprocessing including data standardization and normalization operations, extracts key features from the data, and transmits the processed data to the knowledge tracking module. Natural language processing and sentiment analysis tools are used to process the text data collected by the intelligent interaction; The knowledge tracking module is built-in with a knowledge tracking model based on a deep learning algorithm. It receives the processed data transmitted by the data processing module, learns and trains the data through a multi-layer neural network, calculates the mastery probability and knowledge state vector of the student for each knowledge point, and transmits the calculation result to the recommendation strategy generation module. Combined with the intelligent interaction feedback, the knowledge state evaluation process is optimized. The recommendation strategy generation module formulates personalized learning recommendation strategies based on the student knowledge state output by the knowledge tracking module and the learning resource library of the educational metaverse platform. For knowledge points that students have a low mastery of, relevant basic knowledge explanation materials, case analysis materials, and practice questions are selected from the resource library. For knowledge points that students have a high mastery of, extension learning content is selected. The recommended content is optimized and sorted using machine learning algorithms, taking into account students' learning interests and learning habits. The recommendation strategy is dynamically adjusted based on intelligent interaction feedback and comprehensive evaluation results. The recommendation result display module displays the recommended content generated by the recommendation strategy generation module in an intuitive way on the educational metaverse platform. The display methods include 3D pop-up windows and immersive task lists, making it easy for students to access and learn the recommended content.

10. The learning apparatus based on educational meta-universe and knowledge tracking according to claim 9, characterized in that: The database design includes a student information database, a learning behavior database, a learning resource database, and a feedback database. The student information database stores students' basic information, including age, learning stage, subject foundation, learning interest tags, and learning habits, providing basic data for personalized learning recommendations. The learning behavior database records all learning behavior data of students on the educational metaverse platform, including learning paths, interactive operations with virtual teaching aids, answering situations, learning durations, discussion area speeches, collection and marking information, and is used for knowledge state assessment and learning recommendation strategy formulation. The learning resource database stores various learning resources in the educational metaverse platform, including basic knowledge explanation materials, case analysis materials, practice questions, advanced application cases, and academic frontier research materials, which are classified and stored according to subjects, knowledge points, and difficulty levels, facilitating the selection of recommended content. The feedback database saves feedback information provided by students through intelligent interaction tools, including learning feelings, doubts, and evaluations of recommended content, which is used to optimize the recommendation strategy and the knowledge tracking model.

Images