Virtual teaching data interaction method and system based on digital twinning

Abstract

The invention discloses a virtual teaching data interaction method and system based on digital twinning. The method comprises the following steps: acquiring audio and video data; determining a fluctuation range according to the data to obtain a change type; clustering the data according to the change type to obtain a clustered data set and isolated digital information; constructing a digital twin virtual teaching model; simulating the isolated digital information to obtain an output variable quantity, and performing parameter setting; and determining a data anomaly degree and making a virtual teaching data interaction decision in combination with the prediction accuracy of the digital twin virtual teaching model. According to the method, the accuracy of virtual teaching data interaction decision can be improved.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a virtual teaching data interaction method and system based on digital twins. Background Art

[0002] At present, in the implementation of virtual teaching systems, in order to ensure the stable operation of the system, a large amount of teaching data needs to be collected, processed and transmitted. These data interaction processes cover various hardware and software devices and communication protocols, and constitute the key link for the stable operation of virtual teaching systems. However, with the surge in the amount of educational data, the quality of data collection is affected by multiple factors such as the performance of the collection equipment and the communication environment, resulting in a decline in data quality, which in turn affects the overall stability of the system.

[0003] In the prior art, although encryption algorithms are used to ensure data security, this often comes at the expense of data transmission and processing speed. In addition, when processing virtual teaching data, the prior art fails to fully consider the impact of data change types and data anomalies on the system, which limits the accuracy of interactive decision-making of virtual teaching data.

[0004] In summary, the existing technology has shortcomings in data interaction methods and fails to fully consider the impact of data change types and abnormalities on the system, resulting in inaccurate data analysis and lack of accuracy in decision-making. Summary of the invention

[0005] The present invention provides a virtual teaching data interaction method based on digital twins to improve the accuracy of virtual teaching data interaction decision-making.

[0006] In the first aspect, in order to solve the above technical problems, the present invention provides a virtual teaching data interaction method based on digital twins, comprising: Acquire audio data and video data from teaching data; determining a fluctuation range according to the audio data and the video data; Making a judgment based on the fluctuation range and a preset standard fluctuation range, and determining the change type based on the judgment result; Inputting the digital information into a pre-trained isolation forest model according to the change type to obtain a clustering data set and isolated digital information; Constructing an initial digital twin virtual teaching model according to the clustering data set, and training the initial digital twin virtual teaching model to obtain a digital twin virtual teaching model; Performing virtual teaching simulation on the isolated digital information to obtain output variation, and setting parameters of the isolated digital information according to the output variation; The data abnormality is determined according to the isolated digital information after the parameter setting, and a virtual teaching data interaction decision is made in combination with the prediction accuracy of the digital twin virtual teaching model.

[0007] In an optional implementation manner, determining the fluctuation range according to the audio data and the video data includes: Preprocessing the audio data and video data; Convert the preprocessed audio data into a waveform signal, and obtain the maximum and minimum values ​​of the waveform signal; Subtracting the maximum value and the minimum value of the waveform signal to obtain the audio data fluctuation range; Extracting fluctuation characteristics of the preprocessed video data; Analyzing the fluctuation characteristics to obtain a fluctuation range of the video data; The fluctuation range includes an audio data fluctuation range and a video data fluctuation range.

[0008] In an optional implementation, judging based on the fluctuation range and a preset standard fluctuation range, and determining the change type based on the judgment result, includes: Determining whether the fluctuation range of the audio data exceeds a preset audio standard fluctuation range; If the audio data exceeds the preset standard fluctuation range, the change type of the audio data is determined to be a sharp increase; If the audio data does not exceed the preset standard fluctuation range, the change type of the audio data is determined to be stable; Determining whether the fluctuation range of the video data exceeds a preset video standard fluctuation range; If the fluctuation range exceeds the preset video standard, the change type of the video data is determined to be a sudden change; If the preset video standard fluctuation range is not exceeded, the change type of the video data is determined to be non-mutation.

[0009] In an optional implementation, the audio data and the video data are input into a pre-trained isolation forest model according to the change type to obtain a clustering data set and isolated digital information, including: Selecting the corresponding digital information as input data of the isolation forest model according to the change type; The isolation forest model clusters the input data by constructing multiple isolated trees for identification; According to the clustering processing result, the input data is divided into the cluster data set and the isolated digital information.

[0010] In an optional implementation, the digital information is input into a pre-trained isolation forest model according to the change type to obtain a clustering data set and isolated digital information, wherein the training process of the isolation forest model includes: Extracting waveform features from the digital information as input data for the model; Recursively selecting optimal features and segmentation points for the input data, and continuously segmenting the input data into two subsets, and when each subset contains only one data sample, the initial construction of the isolation forest model is completed; Randomly select one of the waveform features, and sort the data samples from small to large according to the value of the waveform feature to form training set data; Calculate the average value of the training set data, and randomly select two data samples as reference samples, and divide the training set data into three categories: a sample set between the two reference samples, a sample set smaller than the smaller reference sample, and a sample set larger than the larger reference sample; The isolation forest model is trained to determine whether the average path lengths of the three types of sample sets in the isolation tree are all less than a preset threshold. If so, the isolation forest model training is completed; if not, the isolation forest model continues to be trained.

[0011] In an optional implementation, an initial digital twin virtual teaching model is constructed according to the clustering data set, and the initial digital twin virtual teaching model is trained to obtain the digital twin virtual teaching model, including: Extracting teaching-related feature information from the clustering data set; Construct an initial digital twin virtual teaching model based on feature information; Classifying the digital information according to the change type and iteratively adjusting the digital twin virtual teaching model according to the classification result; Obtaining the error between the output result of the digital twin virtual teaching model and the preset teaching result; Performing training of the digital twin virtual teaching model through error feedback according to the error; When the preset error standard is reached, the training is determined to be completed and the digital twin virtual teaching model is obtained.

[0012] In an optional implementation, performing a virtual teaching simulation on the isolated digital information to obtain an output variation, and setting parameters of the isolated digital information according to the output variation, includes: Inputting the isolated digital information into the digital twin virtual teaching model to simulate a virtual teaching scene; Obtaining the actual output of the digital twin virtual teaching model and performing a subtraction calculation with the preset output to obtain an output change; According to the output variation, relevant parameters in the isolated digital information are set.

[0013] In an optional implementation, the data abnormality is determined according to the isolated digital information after the parameter setting, and combined with the prediction accuracy of the digital twin virtual teaching model, a virtual teaching data interaction decision is made, including: Calculate the data anomaly degree of isolated digital information after parameter adjustment; Obtaining the prediction accuracy of the digital twin virtual teaching model; Combining the data abnormality and the prediction accuracy, a virtual teaching data interaction decision is made.

[0014] In a second aspect, the present invention provides a virtual teaching data interaction system based on digital twins, comprising: A data acquisition module is used to acquire audio data and video data in teaching data; A fluctuation range determination module, used to determine the fluctuation range according to the audio data and the video data; A change type judgment module, used to judge the change type according to the fluctuation range and a preset standard fluctuation range; A clustering module, used for clustering the audio data and the video data according to the change type to obtain a clustered data set and isolated digital information; A model construction and training module, used to construct an initial digital twin virtual teaching model according to the feature information, and classify the digital information according to the change type, iteratively adjust the digital twin virtual teaching model according to the classification result, obtain the error between the output result of the digital twin virtual teaching model and the preset teaching result, and train the digital twin virtual teaching model through error feedback according to the error. When the preset error standard is reached, it is determined that the training is completed, and the digital twin virtual teaching model is obtained; A simulation and parameter setting module, used for performing virtual teaching simulation on the isolated digital information, obtaining an output variation, and performing parameter setting on the isolated digital information according to the output variation; The decision-making module is used to determine the data abnormality according to the isolated digital information after the parameter setting, and to make a virtual teaching data interaction decision in combination with the prediction accuracy of the digital twin virtual teaching model.

[0015] In the third aspect, the present invention also provides a computer-readable storage medium, which includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute any one of the above-mentioned virtual teaching data interaction methods based on digital twins.

[0016] Compared with the prior art, the present invention has the following beneficial effects: the present invention improves the accuracy of virtual teaching data interaction decision-making and enhances the stability and security of the system through a virtual teaching data interaction method based on digital twins. First, the data acquisition module is used to obtain audio data and video data in real time, and then the feature extraction module is used to extract and analyze the features of these data, determine the fluctuation range, and identify the data change type by comparing with the preset standard fluctuation range. The data is clustered and analyzed with the help of the isolation forest model to accurately distinguish the clustered data set and isolated digital information. The model construction and training module trains the digital twin virtual teaching model to simulate the real teaching environment, performs virtual teaching simulation on the isolated digital information, obtains the output change, and optimizes the parameters of the isolated digital information according to the output change. The decision-making module comprehensively considers the data abnormality of the isolated digital information after parameter setting and the prediction accuracy of the digital twin virtual teaching model, and formulates a reasonable virtual teaching data interaction decision.

[0017] To sum up, the present invention provides a virtual teaching data interaction method based on digital twins, which realizes the precise management of virtual teaching operations, saves educational resources, and improves teaching efficiency. In the long-term operation, it can significantly reduce the decision-making errors caused by inaccurate data interaction and reduce operating costs. At the same time, it improves the efficiency and effectiveness of teaching resource management, and has important social value and broad application prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a flowchart of a virtual teaching data interaction method based on digital twins provided by the first embodiment of the present invention; Figure 2 It is a structural diagram of a virtual teaching data interaction system based on digital twins provided in the second embodiment of the present invention. DETAILED DESCRIPTION

[0019] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0020] Reference Figure 1 The first embodiment of the present invention provides a virtual teaching data interaction method based on digital twins, comprising the following steps: S11, obtaining audio data and video data in the teaching data; S12, determining a fluctuation range according to the audio data and the video data; S13, judging according to the fluctuation range and a preset standard fluctuation range, and determining the change type according to the judgment result; S14, inputting the digital information into a pre-trained isolation forest model according to the change type to obtain a clustering data set and isolated digital information; S15, constructing an initial digital twin virtual teaching model according to the clustering data set, and training the initial digital twin virtual teaching model to obtain a digital twin virtual teaching model; S16, performing a virtual teaching simulation on the isolated digital information to obtain an output variation, and setting parameters of the isolated digital information according to the output variation; S17, determining the data abnormality according to the isolated digital information after the parameter setting, and making a virtual teaching data interaction decision in combination with the prediction accuracy of the digital twin virtual teaching model.

[0021] It should be noted that S11 to S17 constitute a complete interactive decision-making process for formulating virtual teaching data. Audio data and video data in the teaching process are captured in real time through sensors to provide a data basis for subsequent feature extraction and analysis. The collected audio data and video data are pre-processed to convert the audio signal and video signal into waveform signals, and the maximum and minimum values ​​of the waveform are obtained to calculate the fluctuation range. The fluctuation range is compared with the preset standard fluctuation range, and the type of data change is determined based on the comparison result. This classification helps to identify different dynamics in the teaching process. According to the determined change type, the digital information is input into a pre-trained isolation forest model, which constructs multiple isolated trees to evaluate the degree of isolation of data points, thereby distinguishing clustered data sets from isolated digital information. , the feature information in the clustering data set is used to construct the initial digital twin virtual teaching model. The model simulates the real teaching environment and uses the classification results of the change type for iterative adjustment. In the model training stage, the error between the model output and the expected teaching result is calculated and fed back to optimize the model parameters. When the error drops to the preset error standard, the training is completed, and a digital twin virtual teaching model is obtained to provide simulation and decision support for virtual teaching. Virtual teaching simulation is performed on isolated digital information, and the difference between the actual output and the expected output of the model, that is, the output change, is calculated. The isolated digital information is parameterized according to the output change. The data anomaly is determined based on the isolated digital information after parameter setting, and the virtual teaching data interaction decision is made in combination with the prediction accuracy of the digital twin virtual teaching model.

[0022] It is worth noting that the present invention demonstrates an efficient interactive decision-making process for virtual teaching data. Through intelligent data analysis and model optimization, the accuracy of interactive decision-making in virtual teaching is significantly improved. The method first extracts key feature information related to teaching activities from clustered data sets, such as student interaction frequency, teaching content coverage, and learning progress. Then, these feature information are used to build a preliminary digital twin virtual teaching model, which is intended to simulate the real teaching environment and learning process. Subsequently, the digital information is classified according to the type of change in the teaching data, and the digital twin virtual teaching model is iteratively adjusted according to the classification results. This process involves calculating the error between the model output result and the preset teaching result, and training the model through an error feedback mechanism. When the error of the model is reduced to the preset error standard, the training is completed and an optimized digital twin virtual teaching model is obtained. This model can provide accurate simulation and decision support for virtual teaching, thereby improving the teaching quality and learning experience. At the same time, it also helps to reduce the waste of teaching resources, improve the automation level of teaching management, and bring significant economic and social benefits to the field of education.

[0023] In step S11, audio data and video data in the teaching data are obtained.

[0024] It should be noted that in the virtual teaching environment, smart sensors are deployed on key teaching equipment and interactive platforms to collect audio data and video data in real time. The audio data is obtained by real-time monitoring through microphones installed in the classroom. It reflects the sound dynamics in the teaching process, including teacher explanations, student questions and classroom discussions. It is a key indicator for evaluating teaching interaction and student participation. The video data is captured in real time by cameras in the classroom, recording visual information in the classroom, such as student expressions, movements and teacher demonstrations. These data are used to analyze teaching activities and student reactions. In step S11, the audio data and video data collected in real time by smart sensors can comprehensively and accurately grasp the progress of teaching activities and changes in the classroom environment, and provide reliable data support for subsequent virtual teaching data interaction and decision-making.

[0025] In step S12, determining a fluctuation range according to the audio data and the video data includes: Preprocessing the audio data and video data; Convert the preprocessed audio data into a waveform signal, and obtain the maximum and minimum values ​​of the waveform signal; Subtracting the maximum value and the minimum value of the waveform signal to obtain the audio data fluctuation range; Extracting fluctuation characteristics of the preprocessed video data; Analyzing the fluctuation characteristics to obtain a fluctuation range of the video data; The fluctuation range includes an audio data fluctuation range and a video data fluctuation range.

[0026] It should be noted that, first, the audio data and video data collected from the smart sensors are preprocessed. This step includes noise reduction, echo cancellation, volume normalization, image enhancement, and frame rate adjustment, which aims to improve the quality of signals and images. Subsequently, the preprocessed audio data is converted into a waveform signal, and the video data is converted into a series of frames to identify the maximum and minimum values ​​of the fluctuation characteristics within a specific time window. These extreme values ​​reveal the magnitude of change in teaching activities during that time period. By calculating the difference between the maximum and minimum values ​​of these waveform signals and fluctuation characteristics, the fluctuation range is obtained. This range reflects the dynamic changes in teaching activities within a specific time period. In audio data, a larger fluctuation range indicates the intensity of classroom discussions, while in video data, it indicates changes in student participation. These fluctuation ranges are key data for analyzing the types of changes in teaching activities and building digital twin models, providing data support for virtual teaching data interaction, thereby enhancing the accuracy of teaching decisions and the quality of teaching interactions.

[0027] In step S13, a judgment is made based on the fluctuation range and a preset standard fluctuation range, and a change type is determined based on the judgment result, including: Determining whether the fluctuation range of the audio data exceeds a preset audio standard fluctuation range; If the audio data exceeds the preset standard fluctuation range, the change type of the audio data is determined to be a sharp increase; If the audio data does not exceed the preset standard fluctuation range, the change type of the audio data is determined to be stable; Determining whether the fluctuation range of the video data exceeds a preset video standard fluctuation range; If the fluctuation range exceeds the preset video standard, the change type of the video data is determined to be a sudden change; If the preset video standard fluctuation range is not exceeded, the change type of the video data is determined to be non-mutation.

[0028] It should be noted that, for the audio data, a standard fluctuation range is preset, and this range is determined based on the specific conditions of the audio data and the teaching environment, and is used to identify the normal fluctuation of the audio signal. Then, the audio data fluctuation range calculated in step S12 is compared with the preset standard fluctuation range. If the fluctuation range of the audio data exceeds the preset audio standard fluctuation range, it indicates that the audio signal has a significant change, and its change type is determined to be a sudden increase. This sudden increase indicates some important events in the teaching process, such as the teacher suddenly raising the volume and the students discussing in a group. If the fluctuation range of the audio data does not exceed the preset audio standard fluctuation range, the change type of the audio data is determined to be The type is stable, which means that the audio signal is within an acceptable fluctuation range and the teaching environment is relatively stable. For video data, a similar method is adopted. First, a standard fluctuation range is preset, and then the video data fluctuation range obtained in step S12 is compared with the preset standard fluctuation range. If the fluctuation range of the video data exceeds the preset video standard fluctuation range, the change type of the video data is determined to be a mutation, which reflects significant changes in classroom activities, such as students' sudden movements and teachers' dynamic demonstrations. If the fluctuation range of the video data does not exceed the preset video standard fluctuation range, the change type of the video data is determined to be non-mutation, indicating that the video content is within the normal fluctuation range and the teaching activities are proceeding smoothly.

[0029] It is worth noting that through this step, the key dynamic changes in the teaching process can be accurately identified, providing an important basis for building a digital twin virtual teaching model and making virtual teaching data interaction decisions, which helps to optimize the allocation of teaching resources and improve the quality of teaching interaction.

[0030] In step S14, the audio data and the video data are input into a pre-trained isolation forest model according to the change type to obtain a clustering data set and isolated digital information, including: Selecting the corresponding digital information as input data of the isolation forest model according to the change type; The isolation forest model clusters the input data by constructing multiple isolated trees for identification; According to the clustering processing result, the input data is divided into the cluster data set and the isolated digital information.

[0031] According to the change type, the digital information is input into a pre-trained isolation forest model to obtain a clustering data set and isolated digital information, wherein the training process of the isolation forest model includes: Extracting waveform features from the digital information as input data for the model; Recursively selecting optimal features and segmentation points for the input data, and continuously segmenting the input data into two subsets, and when each subset contains only one data sample, the initial construction of the isolation forest model is completed; Randomly select one of the waveform features, and sort the data samples from small to large according to the value of the waveform feature to form training set data; Calculate the average value of the training set data, and randomly select two data samples as reference samples, and divide the training set data into three categories: a sample set between the two reference samples, a sample set smaller than the smaller reference sample, and a sample set larger than the larger reference sample; The isolation forest model is trained to determine whether the average path lengths of the three types of sample sets in the isolation tree are all less than a preset threshold. If so, the isolation forest model training is completed; if not, the isolation forest model continues to be trained.

[0032] It should be noted that the corresponding digital information is selected as the input data of the isolation forest model according to the change type. The isolation forest model clusters the input data by constructing multiple isolated trees for identification. The construction process of the isolation tree involves recursively selecting the optimal features and split points for the input data, and continuously splitting the input data into two subsets until each subset contains only one data sample, thereby completing the initial construction of the isolation forest model. In the training process of the isolation forest model, the waveform features are first extracted from the digital information as the input data of the model. Next, a waveform feature is randomly selected, and the data samples are sorted from small to large according to the feature value to form the training set data. The average value of the training set data is calculated, and two data samples are randomly selected from it as reference samples. The training set data is divided into three categories: sample sets between two reference samples, sample sets smaller than the smaller reference sample, and sample sets larger than the larger reference sample. Finally, the isolation forest model is trained to evaluate whether the average path lengths of the three types of sample sets in the isolation tree are all less than the preset threshold. If the conditions are met, the model training is considered completed. If the conditions are not met, the model is continuously trained until the requirements are met. Based on the clustering processing results, the input data is divided into clustering data sets and isolated digital information. The clustering data sets contain data points in normal patterns, while isolated digital information represents abnormal and special events. In this way, the model can identify anomalies in the data, which indicate special events or problems in the teaching process, such as abnormal behavior of students and failure of teaching equipment.

[0033] It is worth noting that this process not only improves the sensitivity to changes in teaching activities, but also optimizes the response to emergencies in the virtual teaching environment through the efficient clustering ability of the isolation forest model. By analyzing the clustering results and isolated information, the teaching strategy can be adjusted more accurately, thereby improving the teaching quality and learning experience.

[0034] In step S15, an initial digital twin virtual teaching model is constructed according to the clustering data set, and the initial digital twin virtual teaching model is trained to obtain the digital twin virtual teaching model including: Extracting teaching-related feature information from the clustering data set; Construct an initial digital twin virtual teaching model based on feature information; Classifying the digital information according to the change type and iteratively adjusting the digital twin virtual teaching model according to the classification result; Obtaining the error between the output result of the digital twin virtual teaching model and the preset teaching result; Performing training of the digital twin virtual teaching model through error feedback according to the error; When the preset error standard is reached, the training is determined to be completed and the digital twin virtual teaching model is obtained.

[0035] It should be noted that feature information closely related to teaching activities is extracted from the clustering data set. These features include student participation, course content complexity, and teaching interaction frequency. They can comprehensively reflect the teaching environment and learner behavior. Using these feature information, the initial digital twin virtual teaching model is constructed. This model is a preliminary framework that aims to simulate the real teaching environment and learning process, including teachers' teaching behaviors, students' learning behaviors, and the display and interaction of teaching content. Next, the data is classified according to the change type of audio data and video data. Then, the digital twin virtual teaching model is iteratively adjusted according to these classification results to ensure that the model can accurately reflect the behavior and results in different teaching scenarios. After the model is adjusted, the output results of the digital twin virtual teaching model are obtained and compared with the preset teaching results, and the error between the two is calculated. This error assessment is a key step in the model training process. It helps us understand the accuracy of the model prediction and guides the subsequent optimization direction. According to the calculated error, the digital twin virtual teaching model is trained through the error feedback mechanism, which includes adjusting the model parameters and adding more training data to reduce the error. Finally, when the error between the model output result and the preset teaching result drops to the preset error standard, the model training is considered to be completed. At this time, an optimized digital twin virtual teaching model is obtained, which can accurately simulate and predict various situations in the teaching process, providing data support and decision-making basis for virtual teaching.

[0036] In step S16, a virtual teaching simulation is performed on the isolated digital information to obtain an output variation, and parameters of the isolated digital information are set according to the output variation, including: Inputting the isolated digital information into the digital twin virtual teaching model to simulate a virtual teaching scene; Obtaining the actual output of the digital twin virtual teaching model and performing a subtraction calculation with the preset output to obtain an output change; According to the output variation, relevant parameters in the isolated digital information are set.

[0037] It should be noted that isolated digital information is input into the trained digital twin virtual teaching model to simulate the virtual teaching scene. Isolated digital information refers to those data points that are identified as abnormal and significantly different from most data points in the clustering process. They represent special events and situations in the teaching process. In the simulation of the virtual teaching scene, the digital twin virtual teaching model generates actual output based on the input isolated digital information, which includes simulating the students' understanding, participation, and learning outcomes. Then, the actual output of the digital twin virtual teaching model is obtained and compared with the preset output, and the difference between the two is calculated to obtain the output change. The preset output is the expected result set according to the input data and teaching objectives, while the actual output is the predicted result obtained by model simulation. The output change reflects the deviation between the model prediction and the actual teaching objective. Finally, according to the calculated output change, the relevant parameters in the isolated digital information are set, which involves adjusting the difficulty of the teaching content, changing the teaching methods, and adjusting the allocation of teaching resources. Through parameter setting, the isolated digital information can be optimized.

[0038] It is worth noting that by inputting isolated digital information into the digital twin virtual teaching model to simulate the teaching scenario, the actual output of the model is obtained. Subsequently, the actual output is compared with the pre-set expected output, and the output change is calculated. This change shows the deviation between the model prediction and the teaching objective. Based on this output change, the relevant parameters of the isolated digital information are adjusted. The purpose is to optimize these information points to make them more in line with the teaching objectives and improve the teaching effect. This process helps to understand and improve the factors that lead to abnormal teaching results, and then make more targeted teaching decisions in the virtual teaching environment.

[0039] In step S17, the data abnormality is determined according to the isolated digital information after the parameter setting, and a virtual teaching data interaction decision is made in combination with the prediction accuracy of the digital twin virtual teaching model, including: Calculate the data anomaly degree of isolated digital information after parameter adjustment; Obtaining the prediction accuracy of the digital twin virtual teaching model; Combining the data abnormality and the prediction accuracy, a virtual teaching data interaction decision is made.

[0040] It should be noted that the data anomaly of the isolated digital information after parameter adjustment is calculated. This step involves quantitative analysis of the isolated digital information after parameter setting to evaluate the degree of deviation from normal data. Data anomaly is an indicator to measure the difference between a data point and the rest of the data set. It is obtained by calculating the deviation between the data point and the model prediction value. Next, the prediction accuracy of the digital twin virtual teaching model is obtained. The prediction accuracy reflects the degree to which the predicted results of the model in a simulated teaching environment are consistent with the actual results. The higher the accuracy, the more reliable the model's prediction is and the more it can truly reflect the teaching activities. Finally, combined with the data anomaly and the prediction accuracy, a virtual teaching data interaction decision is made. This decision-making process includes deciding whether the teaching activities need to be adjusted, whether additional teaching resource support is needed, and whether the current teaching strategy should continue. When the data anomaly is high and the prediction accuracy is low, corresponding teaching intervention measures need to be taken, such as changing teaching methods, providing additional tutoring, and adjusting the difficulty of teaching content.

[0041] It is worth noting that through this process, it can be ensured that virtual teaching activities can be optimized based on real-time data and model predictions, thereby improving teaching quality and learning efficiency. This decision-making method based on data and model prediction provides an automated and precise teaching management method for virtual teaching, which helps to improve the quality of teaching.

[0042] In order to facilitate the understanding of the present invention, some preferred embodiments of the present invention are further described below.

[0043] In this embodiment, a virtual teaching data interaction decision is made for an online course, in which the teacher teaches through video conferencing software and students participate in class discussions through audio and video.

[0044] The working process is as follows: Step 1: Obtain audio and video data from teaching data. At the beginning of the course, smart sensors and data acquisition interfaces capture the teacher’s audio and students’ video interactions in real time, and these data are transmitted to the data processing center.

[0045] Step 2: Determine the fluctuation range based on the audio data and the video data. Preprocess the raw audio data and video data to extract key features, such as volume changes in the audio and body language of students in the video, and then calculate the maximum and minimum values ​​of these features to determine the fluctuation range.

[0046] Step 3: Determine the type of change based on the fluctuation range and the preset standard fluctuation range. Compare the calculated fluctuation range with the preset standard fluctuation range. When the audio fluctuation range exceeds the preset range, it indicates that the classroom discussion is unusually active. When the video fluctuation range exceeds the preset range, it indicates a sudden change in student behavior.

[0047] Step 4: Input the digital information into the pre-trained isolation forest model according to the change type to obtain a clustered data set and isolated digital information. According to the change type, the data is input into the isolation forest model, which recognizes and distinguishes normal and abnormal teaching interaction patterns, and outputs a clustered data set and isolated digital information.

[0048] Step 5: Construct an initial digital twin virtual teaching model based on the clustering data set, and train the initial digital twin virtual teaching model to obtain a digital twin virtual teaching model. Use the feature information in the clustering data set to construct an initial digital twin virtual teaching model to simulate the real classroom environment, and then use the teaching data to train and optimize the model.

[0049] Step 6: Perform virtual teaching simulation on the isolated digital information to obtain output variation, and set parameters for the isolated digital information according to the output variation. Perform virtual teaching simulation on the isolated digital information, compare the output predicted by the model with the actual output, and calculate the output variation. Adjust the parameters of the isolated digital information according to the variation to better meet the expected teaching effect.

[0050] Step 7: Determine the data abnormality according to the isolated digital information after the parameter setting, and make a virtual teaching data interaction decision in combination with the prediction accuracy of the digital twin virtual teaching model. Evaluate the data abnormality of the isolated digital information after adjusting the parameters, and make a teaching decision in combination with the prediction accuracy of the model. When the data abnormality is low and the prediction accuracy is high, it is decided to continue the current teaching strategy. When the data abnormality is high and the prediction accuracy is low, it is necessary to adjust the teaching method and resource allocation.

[0051] In summary, the present invention provides a virtual teaching data interaction method based on digital twins, which optimizes the configuration of teaching resources and improves the accuracy and efficiency of teaching interaction by intelligently managing data interaction in a virtual teaching environment. First, intelligent sensors are used to capture audio data and video teaching data in real time. These data reflect the actual situation of teaching activities. Then, the collected data is preprocessed and its fluctuation range is calculated to provide a basis for analyzing teaching dynamics. By comparing the data fluctuation range with the preset standard, the change type of teaching activities, such as stability and mutation, is identified. Next, the isolation forest model is used to cluster the data of the change type to distinguish between normal and abnormal teaching interaction modes. A digital twin virtual teaching model is constructed based on the clustered data set, and the model is iteratively optimized by inputting data to improve the accuracy of prediction. Virtual teaching simulation is performed on isolated digital information, and parameters are adjusted according to the output change amount to meet the expected teaching objectives. Finally, accurate virtual teaching data interaction decisions are made based on the data anomaly degree of the isolated digital information after parameter adjustment and the prediction accuracy of the model. The present invention improves the decision-making accuracy of virtual teaching, makes teaching management more scientific and accurate, reduces the waste of teaching resources through intelligent means, improves resource utilization efficiency, and enhances the real-time nature of teaching interaction, thereby optimizing learning experience and teaching effect.

[0052] Reference Figure 2 The present invention provides a virtual teaching data interaction system based on digital twins, comprising: A data acquisition module is used to acquire audio data and video data in teaching data; A fluctuation range determination module, used to determine the fluctuation range according to the audio data and the video data; A change type judgment module, used to judge the change type according to the fluctuation range and a preset standard fluctuation range; A clustering module, used for clustering the audio data and the video data according to the change type to obtain a clustered data set and isolated digital information; A model construction and training module, used to construct an initial digital twin virtual teaching model according to the feature information, and classify the digital information according to the change type, iteratively adjust the digital twin virtual teaching model according to the classification result, obtain the error between the output result of the digital twin virtual teaching model and the preset teaching result, and train the digital twin virtual teaching model through error feedback according to the error. When the preset error standard is reached, it is determined that the training is completed, and the digital twin virtual teaching model is obtained; A simulation and parameter setting module, used for performing virtual teaching simulation on the isolated digital information, obtaining an output variation, and performing parameter setting on the isolated digital information according to the output variation; The decision-making module is used to determine the data abnormality according to the isolated digital information after the parameter setting, and to make a virtual teaching data interaction decision in combination with the prediction accuracy of the digital twin virtual teaching model.

[0053] In summary, the present invention provides a virtual teaching data interaction system based on digital twins, which optimizes the configuration of teaching resources and improves the accuracy and efficiency of teaching interaction by intelligently managing data interaction in a virtual teaching environment. First, intelligent sensors are used to capture audio data and video teaching data in real time. These data reflect the actual situation of teaching activities. Then, the collected data is preprocessed and its fluctuation range is calculated to provide a basis for analyzing teaching dynamics. By comparing the data fluctuation range with the preset standard, the change type of teaching activities, such as stability and mutation, is identified. Next, the isolation forest model is used to cluster the data of the change type to distinguish between normal and abnormal teaching interaction modes. A digital twin virtual teaching model is constructed based on the clustered data set, and the model is iteratively optimized by inputting data to improve the accuracy of prediction. Virtual teaching simulation is performed on isolated digital information, and parameters are adjusted according to the output change amount to meet the expected teaching objectives. Finally, accurate virtual teaching data interaction decisions are made based on the data anomaly degree of the isolated digital information after parameter adjustment and the prediction accuracy of the model. The present invention improves the decision-making accuracy of virtual teaching, makes teaching management more scientific and accurate, reduces the waste of teaching resources through intelligent means, improves resource utilization efficiency, and enhances the real-time nature of teaching interaction, thereby optimizing learning experience and teaching effect.

[0054] It should be noted that a virtual teaching data interaction system based on digital twins provided in an embodiment of the present invention is used to execute all the process steps of a virtual teaching data interaction method based on digital twins in the above-mentioned embodiment. The working principles and beneficial effects of the two correspond one to one, so they will not be repeated here.

[0055] The embodiment of the present invention further provides an electronic device. The electronic device includes: a processor, a memory, and a computer program stored in the memory and executable on the processor, such as a virtual teaching data interaction program based on digital twins. When the processor executes the computer program, the steps in the above-mentioned embodiments of the virtual teaching data interaction method based on digital twins are implemented, such as Figure 1 Alternatively, when the processor executes the computer program, the functions of each module / unit in the above-mentioned device embodiments are implemented, such as the decision-making module.

[0056] Exemplarily, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to implement the present invention. The one or more modules / units may be a series of computer program instruction segments capable of implementing specific functions, which are used to describe the execution process of the computer program in the electronic device.

[0057] The electronic device may be a computing device such as a desktop computer, a notebook, a PDA, and a smart tablet. The electronic device may include, but is not limited to, a processor and a memory. Those skilled in the art will appreciate that the above components are merely examples of electronic devices and do not constitute a limitation on the electronic device. The electronic device may include more or fewer components than the above components, or may combine certain components, or different components. For example, the electronic device may also include input and output devices, network access devices, buses, etc.

[0058] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the electronic device, and uses various interfaces and lines to connect various parts of the entire electronic device.

[0059] The memory can be used to store the computer program and / or module, and the processor realizes various functions of the electronic device by running or executing the computer program and / or module stored in the memory, and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area can store data created according to the use of the mobile phone (such as audio data, a phone book, etc.), etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0060] Wherein, if the module / unit integrated in the electronic device is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Wherein, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0061] It should be noted that the device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the accompanying drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines. A person of ordinary skill in the art may understand and implement it without paying any creative effort.

[0062] The specific embodiments described above further illustrate the purpose, technical solutions and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. It is particularly pointed out that for those skilled in the art, any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of protection of the present invention.

Claims

1. A virtual teaching data interaction method based on digital twins, characterized in that: Executed by a computer, including: Acquire audio data and video data from teaching data; determining a fluctuation range according to the audio data and the video data; Making a judgment based on the fluctuation range and a preset standard fluctuation range, and determining the change type based on the judgment result; Inputting the digital information into a pre-trained isolation forest model according to the change type to obtain a clustering data set and isolated digital information; Constructing an initial digital twin virtual teaching model according to the clustering data set, and training the initial digital twin virtual teaching model to obtain a digital twin virtual teaching model; Performing virtual teaching simulation on the isolated digital information to obtain output variation, and setting parameters of the isolated digital information according to the output variation; The data abnormality is determined according to the isolated digital information after the parameter setting, and a virtual teaching data interaction decision is made in combination with the prediction accuracy of the digital twin virtual teaching model.

2. The virtual teaching data interaction method based on digital twin according to claim 1 is characterized in that: Determining a fluctuation range according to the audio data and the video data includes: Preprocessing the audio data and video data; Convert the preprocessed audio data into a waveform signal, and obtain the maximum and minimum values ​​of the waveform signal; Subtracting the maximum value and the minimum value of the waveform signal to obtain the audio data fluctuation range; Extracting fluctuation characteristics of the preprocessed video data; Analyzing the fluctuation characteristics to obtain a fluctuation range of the video data; The fluctuation range includes an audio data fluctuation range and a video data fluctuation range.

3. The virtual teaching data interaction method based on digital twin according to claim 2 is characterized in that: A judgment is made based on the fluctuation range and a preset standard fluctuation range, and a change type is determined based on the judgment result, including: Determining whether the fluctuation range of the audio data exceeds a preset audio standard fluctuation range; If the audio data exceeds the preset standard fluctuation range, the change type of the audio data is determined to be a sharp increase; If the audio data does not exceed the preset standard fluctuation range, the change type of the audio data is determined to be stable; Determining whether the fluctuation range of the video data exceeds a preset video standard fluctuation range; If it exceeds the preset video standard fluctuation range, the change type of the video data is determined to be a mutation; If the preset video standard fluctuation range is not exceeded, the change type of the video data is determined to be non-mutation.

4. The virtual teaching data interaction method based on digital twin according to claim 3 is characterized in that: Inputting the audio data and the video data into a pre-trained isolation forest model according to the change type to obtain a clustering data set and isolated digital information, including: Selecting the corresponding digital information as input data of the isolation forest model according to the change type; The isolation forest model performs clustering processing on the input data by constructing multiple isolated trees for identification; According to the clustering processing result, the input data is divided into the cluster data set and the isolated digital information.

5. The virtual teaching data interaction method based on digital twin according to claim 4 is characterized in that: According to the change type, the digital information is input into a pre-trained isolation forest model to obtain a clustering data set and isolated digital information, wherein the training process of the isolation forest model includes: Extracting waveform features from the digital information as input data for the model; Recursively selecting optimal features and segmentation points for the input data, and continuously segmenting the input data into two subsets, and when each subset contains only one data sample, the initial construction of the isolation forest model is completed; Randomly select one of the waveform features, and sort the data samples from small to large according to the value of the waveform feature to form training set data; Calculate the average value of the training set data, and randomly select two data samples as reference samples, and divide the training set data into three categories: a sample set between the two reference samples, a sample set smaller than the smaller reference sample, and a sample set larger than the larger reference sample; The isolation forest model is trained to determine whether the average path lengths of the three types of sample sets in the isolation tree are all less than a preset threshold. If so, the isolation forest model training is completed; if not, the isolation forest model continues to be trained.

6. The virtual teaching data interaction method based on digital twin according to claim 3 is characterized in that An initial digital twin virtual teaching model is constructed according to the clustering data set, and the initial digital twin virtual teaching model is trained to obtain a digital twin virtual teaching model including: Extracting teaching-related feature information from the clustering data set; Construct an initial digital twin virtual teaching model based on feature information; Classifying the digital information according to the change type and iteratively adjusting the digital twin virtual teaching model according to the classification result; Obtaining the error between the output result of the digital twin virtual teaching model and the preset teaching result; Performing training of the digital twin virtual teaching model through error feedback according to the error; When the preset error standard is reached, the training is determined to be completed and the digital twin virtual teaching model is obtained.

7. The virtual teaching data interaction method based on digital twin according to claim 1 is characterized in that Performing a virtual teaching simulation on the isolated digital information to obtain an output variation, and setting parameters of the isolated digital information according to the output variation, including: Inputting the isolated digital information into the digital twin virtual teaching model to simulate a virtual teaching scene; Obtaining the actual output of the digital twin virtual teaching model and performing a subtraction calculation with the preset output to obtain an output change; According to the output variation, relevant parameters in the isolated digital information are set.

8. The virtual teaching data interaction method based on digital twin according to claim 1 is characterized in that: Determine the data abnormality according to the isolated digital information after the parameter setting, and formulate a virtual teaching data interaction decision in combination with the prediction accuracy of the digital twin virtual teaching model, including: Calculate the data anomaly degree of isolated digital information after parameter adjustment; Obtaining the prediction accuracy of the digital twin virtual teaching model; Combining the data abnormality and the prediction accuracy, a virtual teaching data interaction decision is made.

9. A virtual teaching data interaction system based on digital twins, characterized in that: include: A data acquisition module is used to acquire audio data and video data in teaching data; A fluctuation range determination module, used to determine the fluctuation range according to the audio data and the video data; A change type judgment module, used to judge the change type according to the fluctuation range and a preset standard fluctuation range; A clustering module, used for clustering the audio data and the video data according to the change type to obtain a clustered data set and isolated digital information; A model construction and training module, used to construct an initial digital twin virtual teaching model according to the feature information, and classify the digital information according to the change type, iteratively adjust the digital twin virtual teaching model according to the classification result, obtain the error between the output result of the digital twin virtual teaching model and the preset teaching result, and train the digital twin virtual teaching model through error feedback according to the error. When the preset error standard is reached, it is determined that the training is completed, and the digital twin virtual teaching model is obtained; A simulation and parameter setting module, used for performing virtual teaching simulation on the isolated digital information, obtaining an output variation, and performing parameter setting on the isolated digital information according to the output variation; The decision-making module is used to determine the data abnormality according to the isolated digital information after the parameter setting, and to make a virtual teaching data interaction decision in combination with the prediction accuracy of the digital twin virtual teaching model.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored computer program, wherein, when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute a virtual teaching data interaction method based on digital twins as described in any one of claims 1 to 7.

Images