Intelligent scene simulation method and system

Abstract

The present invention relates to the field of scene simulation technology, and more specifically, to an intelligent scene simulation method and system. The method comprises the following steps: S1, collecting user demand data for teaching scenes, analyzing learning content based on the demand data, and generating a target simulation scene based on the learning content; S2, collecting real-time learning videos of the user and the target simulation scene, generating a standard character model based on the learning content, and simultaneously generating a user character model based on the real-time learning video; S3, extracting the user's historical learning videos, extracting the user's personality based on the historical learning videos, and adjusting the standard character model. By personalizing the standard character model according to the user, the difference rate between the standard character model and the user character model is more accurate, thereby avoiding errors in calculating the difference rate that may lead to an overestimation and affect the user's normal learning progress.

Description

Technical Field

[0001] The present invention relates to the technical field of scene simulation, and in particular to an intelligent scene simulation method and system. Background Art

[0002] With the continuous development of educational technology, especially the introduction of technologies such as virtual reality (VR) and augmented reality (AR), intelligent scene simulation has become an important direction for improving learning experience and efficiency. The emergence of intelligent scene simulation methods can not only effectively improve learning effects, but also provide users with an immersive learning experience by simulating real scenes.

[0003] However, traditional scenario simulation systems usually adopt preset rules and fixed templates, and lack a real-time feedback mechanism for users' personalized learning needs and learning progress, resulting in an inability to respond promptly to users' actual needs in the learning process. For example, when learning a certain module, users may encounter great difficulties, but the existing system often cannot adjust the scenario according to the user's performance, resulting in low learning efficiency. Therefore, an intelligent scenario simulation method and system are proposed. Summary of the Invention

[0004] The purpose of the present invention is to provide an intelligent scene simulation method and system to solve the problems raised in the above background technology.

[0005] To solve the above technical problems, one of the objectives of the present invention is to provide an intelligent scene simulation method, comprising the following steps:

[0006] S1. Collect user demand data for teaching scenarios, analyze learning content based on the demand data, and generate target simulation scenarios based on the learning content;

[0007] S2. Collect real-time learning videos of users and target simulation scenarios, generate standard persona models based on the learning content, and generate user persona models based on the real-time learning videos;

[0008] S3, extracting the user's historical learning videos, extracting the user's personality based on the historical learning videos, and adjusting the standard character model;

[0009] S4. Determine the learning content stage based on the user persona model, then extract the standard persona model corresponding to the learning content stage and perform standard difference analysis with the user persona model, and obtain a real-time difference rate based on the analysis results;

[0010] S5. Extract historical difference rates based on historical learning videos, and perform scene adaptation threshold analysis on the historical difference rates in combination with the learning difficulty of the learning content stage. Then, perform scene simulation adjustment analysis on the scene adaptation threshold in combination with the real-time difference rate. Based on the analysis results, determine whether to intelligently adjust the target simulation scene.

[0011] As a further improvement of this technical solution, S1 collects demand data on teaching scenarios from users, determines application direction and scenario expectations based on the demand data, and then collects teaching-related data on the network based on the application direction and scenario expectations as basic data for subsequent generation of scenarios.

[0012] As a further improvement of this technical solution, the steps of S1 are as follows:

[0013] S1.1. Extract learning content based on demand data and segment the extracted learning content according to differences, so that the learning content is divided into multiple learning content stages;

[0014] S1.2. Extract data from each learning content stage in combination with teaching-related data, and generate scenario simulations based on the extracted data to obtain the target simulation scenarios corresponding to each learning content stage;

[0015] S1.3. By presenting the target simulation scene to the user, the user can follow the teaching according to the target simulation scene.

[0016] As a further improvement of this technical solution, the steps of S2 are as follows:

[0017] S2.1. Collect real-time learning videos of the user and target simulation scenarios, extract action videos and language text from the collected real-time learning videos, and then generate a user persona model based on the action videos and language text.

[0018] S2.2. Standard learning parameters are extracted according to the teaching content of each learning content stage, and then a standard character model is generated based on the standard learning parameters.

[0019] As a further improvement of this technical solution, the steps of S3 are as follows:

[0020] S3.1. Establish a user history database and save the user's historical learning videos based on their unique identifier;

[0021] If the user is a new user, then the user's unique identifier is created in the historical database, and step S3.2 is not performed. At the same time, the user's personal data is collected;

[0022] S3.2. Extract historical learning videos from the user history database based on the learning user, combine the historical learning videos to perform user personality data analysis, and then combine the user personality data with the standard character model for personalized adjustment.

[0023] As a further improvement of this technical solution, the step of S4 is as follows:

[0024] S4.1. Determine the user's learning content stage based on the user persona model and monitor the learning progress. When the user persona model has completed the learning content stage, extract the standard persona model for the completed learning content stage.

[0025] S4.2. Perform a standard difference analysis on the standard persona model extracted in S4.1 and the user persona model, and obtain a real-time difference rate of the learning standard between the standard persona model and the user persona model based on the analysis result.

[0026] As a further improvement of this technical solution, the steps of S5 are as follows:

[0027] S5.1. Extract historical difference rates based on historical learning videos and the corresponding learning content stages. Then, combine the historical difference rates with the historical learning content stages to analyze user learning ability. Then, combine the learning ability with the learning difficulty of the current learning content stage to perform scenario adaptation threshold analysis.

[0028] S5.2. The scene adaptation threshold is combined with the real-time difference rate to perform scene simulation adjustment analysis and comparison. When the scene adaptation threshold is greater than the real-time difference rate, monitoring is continued. Conversely, when the scene adaptation threshold is less than the real-time difference rate, the target simulation scene is intelligently adjusted.

[0029] As a further improvement of this technical solution, when S5 performs scene adaptation threshold analysis, if the user has no historical difference rate data, the administrator sets an initial scene adaptation threshold, and then substitutes the initial scene adaptation threshold into S5.2 for comparison with the real-time difference rate.

[0030] A second object of the present invention is to provide an intelligent scene simulation system, comprising any one of the intelligent scene simulation methods described above, including a demand collection unit, a difference rate calculation unit, and a scene adjustment unit;

[0031] The demand collection unit is used to collect user demand data for teaching scenarios, perform learning content analysis based on the demand data, generate a target simulation scenario based on the learning content, collect real-time learning videos of the user and the target simulation scenario, generate a standard character model based on the learning content, and generate a user character model based on the real-time learning video;

[0032] The difference rate calculation unit is used to extract the user's historical learning videos, extract the user's personality based on the historical learning videos, adjust the standard character model, determine the learning content stage based on the user character model, and then extract the standard character model corresponding to the learning content stage and the user character model to perform standard difference analysis, and obtain a real-time difference rate based on the analysis results;

[0033] The scene adjustment unit is used to extract the historical difference rate based on the historical learning video, and at the same time perform scene adaptation threshold analysis on the historical difference rate in combination with the learning difficulty of the learning content stage, and then perform scene simulation adjustment analysis on the scene adaptation threshold in combination with the real-time difference rate, and determine the target simulation scene to be intelligently adjusted based on the analysis results.

[0034] Compared with the prior art, the present invention has the following beneficial effects:

[0035] 1. An intelligent scenario simulation method and system that can tailor a personalized learning path for each user by real-time monitoring and analysis of the user's learning performance. At the same time, by flexibly adjusting the depth and breadth of learning content based on the user's real-time difference rate and the difficulty of the current learning stage, the introduction of scenario adaptation thresholds ensures that the difficulty of each learning stage matches the user's actual ability, avoiding situations where the learning process is too difficult or too easy.

[0036] 2. An intelligent scene simulation method and system, which personalizes the standard character model according to the user, so that the difference rate between the standard character model and the user character model is more accurate, avoiding the difference rate calculation error that may lead to an excessive difference rate and affect the user's normal learning progress. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 It is the overall flow chart of the present invention;

[0038] Figure 2 A flowchart of a target simulation scenario corresponding to each learning content stage is obtained for the present invention;

[0039] Figure 3 A flowchart of the process of generating a user persona model for the present invention;

[0040] Figure 4 This is a flowchart of the present invention combining user personality data with a standard character model for personalized adjustment;

[0041] Figure 5 A flowchart of the standard character model for the learning content stage extracted by the present invention;

[0042] Figure 6 A flowchart of the present invention for intelligently adjusting a target simulation scenario;

[0043] Figure 7 This is a schematic diagram of the structural principle of the demand collection unit of the present invention.

[0044] The meaning of each number in the figure is:

[0045] 10. Requirements collection unit; 20. Difference rate calculation unit; 30. Scenario adjustment unit. DETAILED DESCRIPTION

[0046] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0047] like Figure 1 - Figure 7 As shown, one of the purposes of the present invention is to provide an intelligent scene simulation method, comprising the following steps:

[0048] S1. Collect user demand data for teaching scenarios, analyze learning content based on the demand data, and generate target simulation scenarios based on the learning content;

[0049] S1 collects user demand data for teaching scenarios, determines the application direction and scenario expectations based on the demand data, and then collects teaching-related data on the network based on the application direction and scenario expectations as the basic data for subsequent scenario generation. The specific steps are as follows:

[0050] User demand collection: clarify users' needs for teaching scenarios, determine their expectations for learning content, methods, interactions, etc., and then use data analysis to collect needs;

[0051] Demand data analysis: Extract valuable information from the collected data, analyze the core needs of users, determine the direction of teaching applications based on the analysis results, and clarify scenario expectations;

[0052] Network data collection and construction of basic data: According to the application direction and scenario expectations, relevant teaching data is collected from the network to support subsequent scenario generation.

[0053] The steps of S1 are as follows:

[0054] S1.1. Extract learning content based on demand data and segment the extracted learning content according to differences, so that the learning content is divided into multiple learning content stages;

[0055] Obtain all learning-related demand data and use text analysis techniques (such as named entity recognition and sentiment analysis) to identify the main aspects of learning content, such as topics, objectives, difficulty levels, etc. Based on the user's current knowledge level and learning goals, divide the learning content into basic, intermediate, advanced and other stages according to difficulty. At the same time, based on the dependencies between learning content, divide it into multiple logical learning stages.

[0056] S1.2. Extract data from each learning content stage in combination with teaching-related data, and generate scenario simulations based on the extracted data to obtain the target simulation scenarios corresponding to each learning content stage. The specific steps are as follows:

[0057] Extract data from teaching data for each learning stage: Based on the characteristics of each learning stage, further data extraction is carried out in combination with teaching data to optimize the content of each learning stage to ensure that it can improve learning effects in a targeted manner;

[0058] Generate scenario simulations based on extracted data: Use virtual reality (VR), augmented reality (AR), or other simulation tools (such as computer-based learning platforms) to design corresponding learning scenarios based on the different stages and goals of the learning content. At the same time, customize the simulations to ensure that the goals of each learning stage are reflected in the simulation. Goal-oriented modeling can be used to ensure that the goals of each stage can be gradually achieved through simulation scenarios.

[0059] S1.3. Displaying the target simulation scene to the user enables the user to follow the instruction according to the target simulation scene;

[0060] The generated simulation scenarios are presented to users in an appropriate manner, such as online education platforms, VR devices, animated videos, etc. At the same time, user interaction methods are designed to enable users to operate or make decisions in the scenarios, and then guide users to follow the learning process of the scenarios.

[0061] S2. Collect real-time learning videos of users and target simulation scenarios, generate standard persona models based on the learning content, and generate user persona models based on the real-time learning videos;

[0062] The steps of S2 are as follows:

[0063] S2.1. Collect real-time learning videos of the user and target simulation scenarios, extract action videos and language text from the collected real-time learning videos, and then generate a user persona model based on the action videos and language text. The specific steps are as follows:

[0064] Collect real-time learning videos of users and target simulation scenarios: Use cameras, sensors, or other real-time video acquisition devices to capture the user's movements, language, and facial expressions during the learning process. This video data includes the user's interaction with the learning scenario, helping to better simulate user behavior.

[0065] Action video extraction: This process extracts key human body points from videos, typically using existing deep learning models such as OpenPose or MediaPipe. This detects the positions of key human body parts (such as the head, hands, and feet). Based on the extracted key point data, the user's actions in the video are identified. The user's actions are analyzed sequentially to create a precise motion trajectory, ensuring that the character model can smoothly reproduce these actions.

[0066] Language and text extraction: The audio in the video is converted into text using speech recognition technology, extracting the user's spoken content during the learning process. Natural language processing technologies (such as BERT or GPT) are used to analyze the extracted text content to extract emotions, intent, and keywords. The user's emotional state when speaking in the video is then analyzed to generate appropriate emotional responses for the character model.

[0067] Generate a user character model: Map the key points and action information extracted from the video to the digital character's skeleton system to drive the character model's movement. At the same time, based on the extracted text and emotional information, generate the user's facial expressions and voice. Then, combine the action, facial expression, and voice generation modules to generate a complete digital character model that can dynamically interact in the scene based on user input.

[0068] S2.2. Extract standard learning parameters based on the teaching content of each learning content stage, and then generate a standard character model based on the standard learning parameters. The specific steps are as follows:

[0069] Analyze and extract teaching content: Analyze the teaching objectives, teaching materials, learning methods, and learning resources of each learning content stage to determine the relevant learning standards and parameters. Then, convert the extracted teaching content into standardized learning objectives. These learning objectives must be consistent with the course evaluation criteria.

[0070] Extracting standard learning parameters: Extracting standard learning parameters according to different stages of teaching content and converting the extracted learning parameters into a standardized data format so that they can adapt to changes in different learning content and facilitate comparison with other learners or models;

[0071] Build a standard character model: Build a standard character model based on the extracted standard learning parameters.

[0072] S3, extracting the user's historical learning videos, extracting the user's personality based on the historical learning videos, and adjusting the standard character model;

[0073] The steps for S3 are as follows:

[0074] S3.1. Establish a user history database and save the user's historical learning videos based on their unique identifier;

[0075] If the user is a new user, a unique identifier of the user is created in the historical database, and step S3.2 is not performed. At the same time, the user's personal data is collected. The specific steps are as follows:

[0076] Establish a user history database: When a user registers, a unique identifier (UUID) is created for the user and a new record is generated in the database. This includes storing the user's learning videos, learning data, and personal profile information. If the user already exists, skip this step and directly update the data;

[0077] Personalized data collection: Collect users' personalized data, including learning style, learning goals, and users' learning time, frequency, and preferences. By designing user questionnaires or recommendation systems, let users fill in this personalized information.

[0078] S3.2. Extract historical learning videos from the user history database based on the learning user, combine the historical learning videos to analyze the user's personality data, and then combine the user's personality data with the standard character model for personalized adjustment. The specific steps are as follows:

[0079] Historical learning video extraction: extract the user's historical learning video data from the database;

[0080] User Personality Data Analysis: This step analyzes users' historical learning video data to understand their learning preferences and interests. For example, it analyzes the types of learning videos they frequently watch, how long they watch them, and how often they review them. This step uses natural language processing (NLP) technology to extract themes from video content and then uses clustering algorithms to identify users' learning interests.

[0081] Establish a personalized learning model for each user: Combine the user's historical learning video data and personalized data (such as learning style, goals, etc.) to create a personalized learning model for each user;

[0082] Combined with standard personas: Each user's personalized learning model is compared with a pre-set standard persona. The standard persona represents different learning paths and strategies. Through comparative analysis, the differences between the user and the standard learning path can be identified and personalized adjustments can be made accordingly. The specific formula is as follows:

[0083] Mid = CreateModel(Aid, Pid);

[0084] Among them, M id A personalized model is built based on analysis results and personalized data. CreateModel is a model building function. id Personality data obtained for historical learning video data analysis, P id Personalized data collected from users;

[0085] M id，ID =M id +M ID ;

[0086] Among them, M id，ID For the standard character model after personal adjustment, M ID It is a standard character model.

[0087] S4. Determine the learning content stage based on the user persona model, then extract the standard persona model corresponding to the learning content stage and perform standard difference analysis with the user persona model, and obtain a real-time difference rate based on the analysis results;

[0088] The steps of S4 are as follows;

[0089] S4.1. Determine the user's learning content stage based on the user persona model and monitor the learning progress. When the user persona model has completed the learning content stage, extract the standard persona model for the completed learning content stage.

[0090] S4.2. Perform a standard difference analysis on the standard persona model extracted in S4.1 and the user persona model, and obtain a real-time difference rate of the learning standard between the standard persona model and the user persona model based on the analysis result.

[0091] Determine the user's learning stage: Based on the user's learning progress and personalized model, determine the user's current learning content stage. By analyzing the user's learning history, learning time, task completion status, etc., assess the user's current knowledge and skill level;

[0092] Monitoring learning progress: The process of monitoring learning progress requires real-time tracking of the user's task completion status at each stage. By collecting user behavior data (such as learning time, task scores, video viewing progress, etc.), it can be determined whether the user has successfully passed the current learning stage;

[0093] Extracting standard persona models: When the user completes the current learning stage, the standard persona model corresponding to that stage needs to be extracted from the standard persona model library;

[0094] Conduct standard gap analysis: Once the standard persona and user persona are extracted, a gap analysis is conducted to compare the two personas in terms of knowledge acquisition, learning progress, and behavioral characteristics. This can reveal the gap between the user's learning standards and actual performance.

[0095] Calculate the real-time difference rate: The difference rate measures the difference between the standard persona model and the user persona model. By calculating the real-time difference rate, you can evaluate the deviation of the user's learning progress from the standard progress. The formula is as follows:

[0096]

[0097] Among them, d(M S ,M id，ID) is the difference between the user persona model and the standard persona model, M S is the user persona model, N is the number of dimensions to be compared, and M S，i is the value of the user persona model in the i-th dimension, M id，ID，i is the value of the standard character model in the i-th dimension;

[0098]

[0099] Among them, D(M S ,M id，ID) is the total difference rate between the user persona model and the standard persona model.

[0100] S5. Extract historical difference rates based on historical learning videos, and perform scene adaptation threshold analysis on the historical difference rates in combination with the learning difficulty of the learning content stage. Then, perform scene simulation adjustment analysis on the scene adaptation threshold in combination with the real-time difference rate. Based on the analysis results, determine whether to intelligently adjust the target simulation scene.

[0101] The steps for S5 are as follows:

[0102] S5.1. Extract historical difference rates based on historical learning videos and the corresponding learning content stages. Then, combine the historical difference rates with the historical learning content stages to analyze user learning ability. Then, combine the learning ability with the learning difficulty of the current learning content stage to perform scenario adaptation threshold analysis.

[0103] S5.2. Combine the scene adaptation threshold with the real-time difference rate to perform scene simulation adjustment analysis and comparison. When the scene adaptation threshold is greater than the real-time difference rate, continue monitoring. Conversely, when the scene adaptation threshold is less than the real-time difference rate, intelligently adjust the target simulation scene. The specific steps are as follows:

[0104] Extracting historical difference rate: Extracting historical difference rate from S4. The historical difference rate reflects the gap between the user's performance in the historical learning stage and the standard persona model.

[0105] Extract the learning content stage corresponding to the historical difference rate: Each historical difference rate corresponds to a learning content stage. Based on historical learning data, the difference rate is associated with the learning stage;

[0106] Conduct user learning ability analysis: Evaluate user learning ability by analyzing the correlation between historical discrepancy rates and learning content stages. The learning ability analysis needs to be determined by the long-term trend of the discrepancy rate. If the discrepancy rate gradually decreases, it indicates that the user's learning ability is improving; if the discrepancy rate increases, it indicates that there are problems with learning ability.

[0107] Analyze the scenario adaptation threshold based on the learning difficulty of the current learning content stage: Each learning content stage has different learning difficulty levels. Based on the learning objectives and difficulty of the current learning content stage, determine a scenario adaptation threshold. This threshold will affect whether the user can successfully adapt to the current stage of learning content;

[0108] Compare the scene adaptation threshold with the real-time difference rate to perform scene simulation adjustment analysis: Perform scene simulation adjustment based on the current real-time difference rate. If the real-time difference rate is greater than the scene adaptation threshold, it means that the user is having difficulty adapting to the current learning content and needs to make intelligent adjustments to the learning path. If the real-time difference rate is less than the scene adaptation threshold, the user can continue to monitor and maintain the current learning path.

[0109] Intelligently adjust the target simulation scenario: When the real-time difference rate is greater than the scenario adaptation threshold, intelligent adjustments need to be made to the user's learning scenario. Adjustments can include rearranging learning content, increasing learning time, or providing additional resources to help users overcome the difficulties of the current stage.

[0110] Continue monitoring and adjustment: Whether continuing monitoring or making scenario adjustments, you need to continue monitoring the user's learning progress. The adjusted learning path needs to continuously track the real-time difference rate to ensure the effectiveness of the scenario adjustment. The formula is as follows:

[0111] T s =f(D s ,L A );

[0112] Among them, Ts is the scene adaptation threshold, D s Set a difficulty level for each learning content stage, L A is the user's learning ability, and f is a function that calculates the adaptation threshold based on the learning difficulty and learning ability.

[0113] When S5 performs scene adaptation threshold analysis, if the user has no historical difference rate data, the administrator sets the initial scene adaptation threshold, and then substitutes the initial scene adaptation threshold into S5.2 for comparison with the real-time difference rate.

[0114] The second object of the present invention is to provide an intelligent scene simulation system, including any one of the above-mentioned intelligent scene simulation methods, including a demand collection unit 10, a difference rate calculation unit 20 and a scene adjustment unit 30;

[0115] The demand collection unit 10 is used to collect user demand data for teaching scenarios, analyze learning content based on the demand data, generate a target simulation scenario based on the learning content, collect real-time learning videos of the user and the target simulation scenario, generate a standard character model based on the learning content, and generate a user character model based on the real-time learning video;

[0116] The difference rate calculation unit 20 is used to extract the user's historical learning videos, extract the user's personality based on the historical learning videos, adjust the standard character model, determine the learning content stage based on the user character model, and then extract the standard character model corresponding to the learning content stage and perform standard difference analysis with the user character model, and obtain a real-time difference rate based on the analysis results;

[0117] The scene adjustment unit 30 is used to extract the historical difference rate based on the historical learning video, and at the same time perform scene adaptation threshold analysis on the historical difference rate in combination with the learning difficulty of the learning content stage, and then perform scene simulation adjustment analysis on the scene adaptation threshold in combination with the real-time difference rate, and determine the target simulation scene to be intelligently adjusted according to the analysis results.

[0118] The above shows and describes the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions are merely preferred examples of the present invention and are not intended to limit the present invention. Various changes and improvements may be made to the present invention without departing from the spirit and scope of the present invention. Such changes and improvements fall within the scope of the present invention. The scope of protection claimed in the present invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent scene simulation method, characterized by: The steps include: S1. Collect user demand data for teaching scenarios, analyze learning content based on the demand data, and generate target simulation scenarios based on the learning content; S2. Collect real-time learning videos of users and target simulation scenarios, generate standard persona models based on the learning content, and generate user persona models based on the real-time learning videos; The steps of S2 are as follows: S2.

1. Collect real-time learning videos of the user and target simulation scenarios, extract action videos and language text from the collected real-time learning videos, and then generate a user persona model based on the action videos and language text. S2.2, extracting standard learning parameters based on the teaching content of each learning content stage, and then generating a standard character model based on the standard learning parameters; S3, extracting the user's historical learning videos, extracting the user's personality based on the historical learning videos, and performing personalized adjustments to the standard character model; S4. Determine the learning content stage based on the user persona model, then extract the standard persona model that has been personalized for the corresponding learning content stage and perform standard difference analysis with the user persona model, and obtain a real-time difference rate based on the analysis results; S5. Extract historical difference rates based on historical learning videos, and perform scene adaptation threshold analysis based on the historical difference rates and the learning difficulty of the learning content stage. Then, perform scene simulation adjustment analysis based on the scene adaptation threshold and the real-time difference rates. Based on the analysis results, determine whether to intelligently adjust the target simulation scene. The steps of S5 are as follows: S5.

1. Extract historical difference rates based on historical learning videos and the corresponding learning content stages. Then, analyze the user's learning ability based on the historical difference rates and the historical learning content stages. Then, analyze the learning ability based on the learning difficulty of the current learning content stage to perform a scenario adaptation threshold analysis. Calculate the scenario adaptation threshold based on the learning ability and learning difficulty. S5.

2. The scene adaptation threshold is combined with the real-time difference rate to perform scene simulation adjustment analysis and comparison. When the scene adaptation threshold is greater than the real-time difference rate, monitoring is continued. Conversely, when the scene adaptation threshold is less than the real-time difference rate, the target simulation scene is intelligently adjusted.

2. The intelligent scene simulation method according to claim 1, characterized in that: The S1 collects demand data on teaching scenarios from users, determines the application direction and scenario expectations based on the demand data, and then collects teaching-related data on the network based on the application direction and scenario expectations as basic data for subsequent generation of scenarios.

3. The intelligent scene simulation method according to claim 2, characterized in that: The steps of S1 are as follows: S1.

1. Extract learning content based on demand data and segment the extracted learning content according to differences, so that the learning content is divided into multiple learning content stages; S1.

2. Extract data from each learning content stage in combination with teaching-related data, and generate scenario simulations based on the extracted data to obtain the target simulation scenarios corresponding to each learning content stage; S1.

3. By presenting the target simulation scene to the user, the user can follow the teaching according to the target simulation scene.

4. The intelligent scene simulation method according to claim 1, characterized in that: The steps of S3 are as follows: S3.

1. Establish a user history database and save the user's historical learning videos based on their unique identifier; If the user is a new user, then the user's unique identifier is created in the historical database, and step S3.2 is not performed. At the same time, the user's personal data is collected; S3.

2. Extract historical learning videos from the user history database based on the learning user, combine the historical learning videos to perform user personality data analysis, and then combine the user personality data with the standard character model for personalized adjustment.

5. The intelligent scene simulation method according to claim 1, characterized in that: The steps of S4 are as follows: S4.

1. Determine the user's learning content stage based on the user persona model and monitor the learning progress. When the user persona model has completed the learning content stage, extract the personalized adjusted standard persona for the completed learning content stage. S4.

2. Perform standard difference analysis on the personalized adjusted standard persona model extracted in S4.1 and the user persona model, and obtain the real-time difference rate of the learning standard between the personalized adjusted standard persona model and the user persona model based on the analysis results.

6. The intelligent scene simulation method according to claim 1, characterized in that: When the S5 performs scene adaptation threshold analysis, if the user has no historical difference rate data, the administrator sets an initial scene adaptation threshold, and then substitutes the initial scene adaptation threshold into S5.2 for comparison with the real-time difference rate.

7. An intelligent scene simulation system for implementing an intelligent scene simulation method according to any one of claims 1 to 6, characterized in that: It includes a demand collection unit (10), a difference rate calculation unit (20), and a scene adjustment unit (30); The demand collection unit (10) is used to collect user demand data for teaching scenarios, and perform learning content analysis based on the demand data, generate a target simulation scenario based on the learning content, collect real-time learning videos of the user and the target simulation scenario, generate a standard character model based on the learning content, and generate a user character model based on the real-time learning video; The difference rate calculation unit (20) is used to extract the user's historical learning videos, extract the user's personality according to the historical learning videos, perform personalized adjustment on the standard character model, determine the learning content stage according to the user character model, and then extract the standard character model corresponding to the learning content stage that has undergone personalized adjustment and the user character model to perform standard difference analysis, and obtain a real-time difference rate according to the analysis result; The scene adjustment unit (30) is used to extract the historical difference rate based on the historical learning video, and simultaneously perform scene adaptation threshold analysis based on the historical difference rate and the learning difficulty of the learning content stage, and then perform scene simulation adjustment analysis based on the scene adaptation threshold and the real-time difference rate, and determine the target simulation scene to be intelligently adjusted according to the analysis result.

Images