Teaching interaction method and system based on ai

Abstract

The invention provides an ai-based teaching interaction method and system. The method comprises the steps of collecting multi-modal data of students; constructing an individual absorption model to calculate the knowledge absorption rate of the student and predict the knowledge absorption rate sequence trend of the student; constructing feature mapping to embed the knowledge absorptivity and the predicted knowledge absorptivity sequence, and constructing a dynamic adaptive content generator according to the feature mapping; and obtaining a teaching content feature vector generated by the adaptive content generator, and mapping the teaching content feature vector to the social behavior features of the students to obtain a social feature vector. Dynamically adjusting interactive recommendation and group distribution among the students; building an adaptive learning curve model according to the second dynamic social weight updating model, designing a feedback mechanism, and updating interactive recommendation and group distribution among the students; and constructing data output by the block chain network updating model to encrypt and upload, and constructing a learning feedback smart contract. According to the invention, a more intelligent and more efficient solution is provided for modern education.

Description

Technical Field

[0001] The present invention belongs to the field of teaching interaction, and in particular to an AI-based teaching interaction method and system. Background Art

[0002] In modern education, personalized learning and teaching interaction are increasingly valued. The traditional classroom teaching model has been unable to meet the diverse learning needs of students due to its singleness and lack of pertinence. In order to improve this situation, artificial intelligence (AI)-based teaching systems have emerged in recent years. These systems provide personalized learning content and suggestions by analyzing students' learning data. However, existing AI teaching systems still have many limitations in achieving personalized teaching and enhancing interactivity.

[0003] First, current AI teaching systems usually rely on students' historical learning data, such as test scores and homework performance, which can only reflect students' learning status at a specific point in time, but cannot fully capture students' cognitive changes and emotional reactions. The limitations of this static data make it impossible for the system to adjust teaching content in real time, unable to respond to changes in students' needs during the learning process in a timely manner, and lack dynamic adaptability. In addition, the feedback mechanism of existing systems is usually relatively simple, mainly based on standardized scoring and simple prompts, and lacks accurate perception of students' deep understanding and knowledge absorption process. Therefore, these systems often cannot effectively help students overcome learning difficulties or stimulate their interest in learning.

[0004] Secondly, although some AI teaching systems have begun to introduce emotional computing and behavior analysis technologies, such as judging students' emotional states through facial expression recognition and voice emotion detection, these technologies are not yet mature enough in the field of education, and there are challenges such as insufficient accuracy and inability to cope with complex emotional states. This makes it easy for the system to misjudge students' learning status, thus affecting the effectiveness of personalized teaching. In addition, the existing systems are designed to focus more on the learning experience of individual students, but ignore the optimization of social interaction and group learning among students. Modern learning research shows that social interaction and collaborative learning are crucial to students' knowledge construction and skill development, but existing technologies still have limited support in this regard. The lack of in-depth analysis of interactions within learning groups cannot effectively promote knowledge exchange and collaborative cooperation among students.

[0005] Furthermore, most AI teaching systems still remain at a superficial level in terms of adaptability and personalization, such as simply recommending more similar learning content or adjusting the difficulty of the content without in-depth analysis of the specific characteristics of students' learning curves and knowledge absorption processes. Such system design results in the inability to provide the most optimized learning path and strategy adjustment when facing students with different backgrounds and learning abilities. In addition, existing technologies also have certain challenges in data privacy and security, especially when sharing and verifying learning data across platforms, which is prone to the risk of data leakage or tampering. These shortcomings significantly limit the potential and application effect of AI in the field of teaching interaction. Summary of the invention

[0006] The purpose of this invention is to propose an AI-based teaching interaction method and system. Through the combination of multimodal data fusion, adaptive learning algorithm, social network analysis, adaptive learning curve modeling and blockchain technology, the personalization, interactivity and security of the AI ​​teaching interaction system are significantly improved, and many defects of the existing technology are overcome, providing a smarter and more efficient solution for modern education.

[0007] In order to achieve the above object, the present invention provides an AI-based teaching interaction method, the method comprising:

[0008] S1. Deploy sensor equipment to collect multimodal data of students and preprocess the multimodal data, extract and fuse features of the preprocessed data to obtain a comprehensive feature vector, and store the comprehensive feature vector as a multimodal data set; wherein the comprehensive feature vector includes EEG spectrum features, heart rate variability features, eye movement features and facial expression features;

[0009] S2, receiving the multimodal data set sent by the sensor and constructing an individual absorption model based on the multimodal data set to calculate the student's knowledge absorption rate at each time point and predict the student's knowledge absorption rate sequence trend;

[0010] S3, constructing a feature map to embed the knowledge absorption rate and the predicted knowledge absorption rate sequence, constructing a dynamic first adaptive content generator according to the feature map, and designing a joint loss function to jointly train and optimize the discriminator and the first adaptive content generator according to the feature map to obtain a second adaptive content generator;

[0011] S4. Obtain the teaching content feature vector generated by the second adaptive content generator, and map it to the social behavior characteristics of the students to obtain a social feature vector, use the social feature vector to build a student social network, build a first dynamic social weight update model based on the student social network, and design a group interaction optimization algorithm to adjust the weight of the first dynamic social weight update model to obtain a second dynamic social weight update model, generate a new social interaction strategy based on the second dynamic social weight update model, and dynamically adjust the interaction recommendation and group allocation between students;

[0012] S5. Construct an adaptive learning curve model based on the second dynamic social weight update model, and design a feedback mechanism to update the interactive recommendations and group assignments between students;

[0013] S6. Construct a blockchain network to encrypt and upload the data output by the second dynamic social weight update model, and construct a learning feedback smart contract to automatically perform data verification, storage and feedback optimization, and design a learning data consensus algorithm to judge the reception of data blocks; wherein, the learning data consensus algorithm is that all nodes decide whether to accept new data blocks through a voting mechanism, and only if all nodes vote in favor of accepting new data blocks will they accept new data blocks;

[0014] In the student social network, each node represents a student, and the weight of the edge represents the strength of the social connection between students; the first dynamic social weight update model updates the weight of the edge in real time according to the students' knowledge absorption rate and social behavior; the weight of the edge is calculated as follows:

[0015] w i,j,t =α·sim(F social,i,t ,F social,j,t )+β·|KAR i,t -KAR j,t |

[0016] Among them, w i,j,t represents the edge weight between student i and student j at time point t; α and β represent weight coefficients, which are used to balance the influence of social behavior similarity and knowledge absorption rate difference; sim(F social,i,t ,F social,j,t ) represents the similarity of social characteristics between student i and student j at time point t; |KAR i,t -KAR j,t | represents the difference in knowledge absorption rate between student i and student j at time point t;

[0017] A group interaction optimization algorithm is designed to increase the interaction between students with the highest knowledge absorption rate and students with the lowest knowledge absorption rate. The group interaction optimization algorithm is expressed as follows:

[0018]

[0019] Among them, L opt represents the optimization objective function, which is used to minimize the difference in knowledge absorption rate among students in the social network; E represents the edge set in the social network; w i,j,t represents the weight of edge i and j; KAR i,t and KAR j,t represents the knowledge absorption rate of students i and j at time point t;

[0020] The edge weights are updated according to the output of the group interaction optimization algorithm to obtain a second dynamic social weight update model. The edge weights of the second dynamic social weight update model are expressed as follows:

[0021] w i,j,t+1 =w i,j,t +γ·Δw i,j,t

[0022] Among them, w i,j,t+1 represents the edge weight between student i and student j at time point t+1; γ represents the learning rate, which is used to control the step size of weight update; Δw i,j,t Indicates the change in edge weight.

[0023] Preferably, the sensor device includes an EEG headset, a heart rate monitor, an eye tracker and a facial expression recognition camera; the preprocessing includes time synchronization processing, applying an adaptive bandpass filter to the data of the EEG headset to remove noise in the EEG data, processing the data of the heart rate monitor using a window smoothing algorithm, using a peak detection algorithm to identify gaze and saccade processing on the eye movement data to obtain eye movement features, and using a convolutional neural network to extract features from micro-expression data to obtain facial expression features; the feature extraction includes using short-time Fourier transform on the data of the EEG headset to obtain power spectrum features of the α band and the β band in different time windows to obtain EEG spectrum features, and using the standard deviation of the RR interval to calculate the heart rate variability features;

[0024] A joint feature space mapping method is used to map different modality data into a shared feature space.

[0025] Preferably, the individual absorption model is constructed as follows:

[0026] The student state at each moment on the comprehensive feature vector is standardized, and a multimodal time series embedding model that introduces a modal attention weight mechanism is constructed. The output of the modal attention mechanism is combined with the update formula of the multimodal time series embedding model to obtain an updated representation of the student's learning state; wherein the modal attention weight mechanism assigns a weight to each modal data to indicate its relative contribution to the knowledge absorption rate at the current time point, and the multimodal time series embedding model processes the multimodal data and captures temporal dynamic changes;

[0027] The knowledge absorption rate at each time point is calculated and the future knowledge absorption trend is predicted based on the hidden states and modal attention weights output by the multimodal time series embedding model.

[0028] Preferably, in the calculation of the knowledge absorption rate at each time point, a knowledge absorption rate regularization term based on L2 regularization is introduced to avoid overfitting and encourage weight sparsity; the prediction of future knowledge absorption trends is performed by inputting the hidden state into a fully connected layer with an adaptive learning rate, and the output of the fully connected layer is the prediction of the knowledge absorption rate at several future time points.

[0029] Preferably, the first adaptive content generator generates a first generated teaching content feature vector according to the state vector and the parameters of the generator; the joint loss function L(G, D) is expressed as follows:

[0030]

[0031] Among them, G(S t ) represents the content features generated by the generator; D(G(S t )) represents the adaptability score of the discriminator to the generated content; S t represents the state vector; D represents the learning state discriminator; G represents the knowledge enhancement generator; represents the average content feature vector in the training set, which is used to calculate content diversity; λ represents the regularization parameter, which controls the influence of the diversity term;

[0032] The knowledge enhancement generator generates content features and inputs them into a learning state discriminator for learning state discrimination. The learning state discriminator is used to discriminate whether the generated content is suitable for the current learning state.

[0033] Preferably, after the first adaptive content generator is trained using the joint loss function L(G,D), a second adaptive content generator is obtained, and the second adaptive content generator is used to generate teaching content features that adapt to the student's learning status in real time, and these features are converted into a second generated teaching content feature vector through a predefined content library and mapping rules.

[0034] Preferably, the S5 specifically includes:

[0035] S501, extracting the learning behavior characteristics of each student, generating a learning behavior characteristic vector for describing the learning behavior and state of student i at time point t;

[0036] S502. Construct an adaptive learning curve model based on a dynamic Bayesian network according to the learning behavior feature vector, capture the changes in students' learning status and potential dependencies in the time series according to students' knowledge absorption rate and social behavior characteristics, and reflect students' learning progress and knowledge mastery at different time points;

[0037] S503. Designing a real-time feedback mechanism based on the adaptive learning curve model to provide students with personalized learning suggestions and strategy adjustments; wherein the real-time feedback mechanism is set according to a learning efficiency function, and the learning efficiency function is used to evaluate each student's current learning status and potential improvement direction;

[0038] S504. Generate personalized feedback suggestions for students based on the calculation results of the learning efficiency function, and optimize learning strategies.

[0039] Preferably, the learning efficiency function is expressed as follows:

[0040]

[0041] Among them, E i,t represents the learning efficiency of student i at time point t; KAR i,t+1 and KAR i,t represents the knowledge absorption rate of student i at time points t+1 and t; Δt represents the time interval; ω represents the weight parameter, which is used to control the impact of behavioral characteristic differences on learning efficiency; represents the average learning behavior characteristics;

[0042] The knowledge absorption rate represents the state transition equation of the adaptive learning curve model and is calculated as follows:

[0043]

[0044] Among them, KAR i,t+1 represents the knowledge absorption rate of student i at time point t+1; γ and δ represent model parameters, which are used to balance the current absorption rate and new learning contribution; λ represents the learning rate parameter, which is used to adjust the impact of learning behavior characteristics on the absorption rate; F learn,i,t Represents the learning behavior characteristics of student i at time point t.

[0045] Preferably, the encryption uses an elliptic curve cryptographic algorithm to encrypt the data; the verification formula of the learning feedback smart contract is expressed as follows:

[0046]

[0047] Among them, F' data,i,t represents the encrypted data of student i at time point t; F data,i,t represents the unencrypted learning data feature vector; V(F' data,i,t ,key priv ) represents the verification function of the smart contract; key priv Represents the private key of the blockchain network, which is used for data decryption; ECC_decrypt represents the elliptic curve cryptography algorithm.

[0048] In a second aspect of the present invention, an AI-based teaching interaction system is provided, the system comprising:

[0049] A teaching modality data collection module is used to deploy sensor equipment to collect students' multimodal data and preprocess the multimodal data, and perform feature extraction and fusion on the preprocessed data to obtain a comprehensive feature vector, and store the comprehensive feature vector as a multimodal data set; wherein the comprehensive feature vector includes EEG spectrum features, heart rate variability features, eye movement features and facial expression features;

[0050] The knowledge absorption analysis module is used to receive the multimodal data set sent by the sensor and build an individual absorption model based on the multimodal data set to calculate the student's knowledge absorption rate at each time point and predict the trend of the student's knowledge absorption rate sequence;

[0051] The teaching content output module is used to construct a feature map to embed the knowledge absorption rate and the predicted knowledge absorption rate sequence, construct a dynamic first adaptive content generator according to the feature map, and design a joint loss function to jointly train and optimize the discriminator and the first adaptive content generator according to the feature map to obtain a second adaptive content generator;

[0052] A teaching interaction module is used to obtain the teaching content feature vector generated by the second adaptive content generator, and map it to the social behavior characteristics of the students to obtain a social feature vector, use the social feature vector to build a student social network, build a first dynamic social weight update model based on the student social network, and design a group interaction optimization algorithm to adjust the weight of the first dynamic social weight update model to obtain a second dynamic social weight update model, generate a new social interaction strategy based on the second dynamic social weight update model, and dynamically adjust the interaction recommendation and group allocation between students;

[0053] An interactive feedback module is used to construct an adaptive learning curve model based on the second dynamic social weight update model, and to design a feedback mechanism to update interactive recommendations and group assignments between students;

[0054] A data encryption module is used to build a blockchain network to encrypt and upload the data output by the second dynamic social weight update model, and to build a learning feedback smart contract to automatically perform data verification, storage and feedback optimization, and to design a learning data consensus algorithm to judge the reception of data blocks; wherein, the learning data consensus algorithm is that all nodes decide whether to accept new data blocks through a voting mechanism, and only when all nodes vote in favor will they accept new data blocks;

[0055] In the student social network, each node represents a student, and the weight of the edge represents the strength of the social connection between students; the first dynamic social weight update model updates the weight of the edge in real time according to the students' knowledge absorption rate and social behavior; the weight of the edge is calculated as follows:

[0056] w i,j,t =α·sim(F social,i,t ,F social,j,t )+β·|KAR i,t -KAR j,t |

[0057] Among them, w i,j,t represents the edge weight between student i and student j at time point t; α and β represent weight coefficients, which are used to balance the influence of social behavior similarity and knowledge absorption rate difference; sim(F social,i,t ,F social,j,t ) represents the similarity of social characteristics between student i and student j at time point t; |KAR i,t -KAR j,t | represents the difference in knowledge absorption rate between student i and student j at time point t;

[0058] A group interaction optimization algorithm is designed to increase the interaction between students with the highest knowledge absorption rate and students with the lowest knowledge absorption rate. The group interaction optimization algorithm is expressed as follows:

[0059]

[0060] Among them, L opt represents the optimization objective function, which is used to minimize the difference in knowledge absorption rate among students in the social network; E represents the edge set in the social network; w i,j,t represents the weight of edge i and j; KAR i,t and KAR j,t represents the knowledge absorption rate of students i and j at time point t;

[0061] The edge weights are updated according to the output of the group interaction optimization algorithm to obtain a second dynamic social weight update model. The edge weights of the second dynamic social weight update model are expressed as follows:

[0062] w i,j,t+1 =wi,j,t +γ·Δw i,j,t

[0063] Among them, w i,j,t+1 represents the edge weight between student i and student j at time point t+1; γ represents the learning rate, which is used to control the step size of weight update; Δw i,j,t Indicates the change in edge weight.

[0064] The beneficial technical effects of the present invention are at least as follows:

[0065] (1) The present invention uses multimodal data fusion technology to integrate students' physiological data (such as brain waves, heart rate), behavioral data (such as eye movements, micro-expressions), and traditional learning performance data. Through deep learning algorithms, the system constructs a dynamic individual knowledge absorption model that can track students' understanding and absorption of teaching content in real time. Compared with the existing technology that only relies on static analysis of historical learning data, the multimodal data fusion of the present invention can fully reflect students' real-time cognitive status and emotional response, thereby achieving more accurate personalized teaching.

[0066] (2) The present invention introduces a reinforcement learning algorithm that allows the system to dynamically adjust teaching content and interaction strategies based on students’ knowledge absorption models and real-time learning performance. The system can not only adjust the content difficulty based on students’ current level of understanding, but also adjust teaching style and interaction methods in real time through sentiment analysis, such as giving more encouragement or appropriate rest suggestions when students feel frustrated. This adaptive capability greatly enhances the flexibility and pertinence of teaching, overcoming the limitation of existing systems that cannot respond to changes in student needs in real time.

[0067] (3) In view of the defect that the existing technology does not pay enough attention to the interaction between students, the present invention adopts social network analysis (SNA) technology to analyze and optimize the interaction structure and knowledge dissemination path within the learning group in real time. The system can identify key nodes (such as knowledge disseminators) and potential knowledge islands (such as students with low participation) in the learning group, and dynamically adjust the group members and collaborative task configuration to promote group learning efficiency and synergy. This innovation significantly improves the interactivity and effectiveness of group teaching.

[0068] (4) The present invention proposes an adaptive learning curve modeling technology that combines students' multi-dimensional data and historical learning records to dynamically predict students' learning progress and depth of understanding. Based on the changes in the learning curve model, the system provides instant intelligent feedback and reinforcement suggestions to ensure that each student can receive personalized guidance suitable for their learning status. This intelligent feedback mechanism can correct learning deviations in a timely manner and stimulate students' learning interest, solving the problem that the existing system has a single feedback and cannot deeply analyze the student learning process.

[0069] (5) The present invention also introduces blockchain technology to ensure the security and credibility of learning data. Through decentralized data storage and smart contract mechanisms, the system can share and verify learning data securely and transparently, preventing the risk of data tampering and leakage. This innovation not only solves the data privacy issues of existing technologies, but also promotes cross-platform data collaboration and trusted verification. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] The present invention is further described using the accompanying drawings, but the embodiments in the accompanying drawings do not constitute any limitation to the present invention. A person skilled in the art can obtain other drawings based on the following drawings without creative work.

[0071] Figure 1 This is a flow chart of the AI-based teaching interaction method of the present invention. DETAILED DESCRIPTION

[0072] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be understood as limiting the present invention.

[0073] like Figure 1 As shown, the AI-based teaching interaction method provided by the embodiment of the present invention includes the following steps S1-S6:

[0074] S1. Deploy sensor equipment to collect students' multimodal data and preprocess the multimodal data, extract and fuse features of the preprocessed data to obtain a comprehensive feature vector, and store the comprehensive feature vector as a multimodal data set; wherein the comprehensive feature vector includes EEG spectrum features, heart rate variability features, eye movement features and facial expression features.

[0075] Specifically, to ensure the time alignment of multimodal data, all sensor devices (EEG headset, heart rate monitor, eye tracker, facial expression recognition camera) need to be calibrated before data collection to ensure the accuracy of data collection. The EEG device is set to a sensitivity of 0.1μV, the heart rate monitor error is controlled at ±0.5bpm, the eye tracker accuracy is 0.05°, and the facial expression recognition camera adjusts the light to adapt to environmental changes.

[0076] Device time synchronization uses the weighted minimum variance time synchronization algorithm (WMVTS) to calculate a global timestamp to ensure that the data of all devices are aligned at the same time point. This method assigns a weight to the timestamp of each device and calculates the weighted average time as the synchronization time.

[0077]

[0078] Among them, T sync is the global timestamp, T i is the timestamp of the ith sensor, w i is the weight of the sensor (assigned based on the device's time accuracy and latency). For example, if the present invention has three sensor devices, their timestamps are T1 = 100ms, T2 = 105ms, T3 = 102ms, and their weights are w1 = 0.2, w2 = 0.5, w3 = 0.3, then the global timestamp

[0079] Furthermore, the EEG data preprocessing is to apply an adaptive bandpass filter (ABF) to remove noise (such as electromyographic noise and baseline drift) in the EEG data. The bandpass frequency range of the filter is 0.5 Hz to 70 Hz.

[0080] X EEG =ABF(EEG,[0.5,70])

[0081] Among them, X EEG is the filtered EEG signal data.

[0082] Heart rate data preprocessing uses a windowed smoothing algorithm (WSA) to process the heart rate data to eliminate short-term fluctuations and ensure data stability.

[0083]

[0084] Among them, X HR is the smoothed heart rate data, and N is the size of the smoothing window. Assuming N = 5, for HR = [70, 72, 71, 73, 75], then

[0085] Eye movement data preprocessing is to use the peak detection algorithm to identify fixations and saccades, and record the fixation duration D respectively. f and scanning speed S s The micro-expression data is extracted using a convolutional neural network (CNN).

[0086] Furthermore, for EEG data, short-time Fourier transform (STFT) is used to obtain the power spectrum characteristics of the alpha band (8-13Hz) and beta band (13-30Hz) in different time windows:

[0087] F EEG = STFT(X EEG )

[0088] Among them, F EEG is a spectral feature matrix containing the power spectral density of each time window.

[0089] Calculate the heart rate variability (HRV) features and use the standard deviation of the RR interval (SDNN) to quantify the variability of heart rate:

[0090]

[0091] Among them, F HR is the standard deviation of heart rate variability, RR i is the RR interval of each heartbeat, is the mean of the RR intervals.

[0092] Furthermore, the extracted features are fused and the joint feature space mapping method (Joint Feature Space Mapping, JFSM) is used to map the different modal data into a shared feature space. EEG , heart rate variability feature F HR , eye movement characteristics (D f ,S s ), and facial expression features F MF Combined to form a comprehensive feature vector F Fusion :

[0093] F Fusion =JSFM(F EEG ,F HR ,D f ,S s ,F MF )

[0094] Among them, F Fusion It is the comprehensive feature vector after fusion.

[0095] Furthermore, the fused comprehensive feature vector is finally stored as a multimodal dataset D multi , used for the subsequent construction of individual knowledge absorption model:

[0096]

[0097] S2. Receive the multimodal data set sent by the sensor and build an individual absorption model based on the multimodal data set to calculate the student's knowledge absorption rate at each time point and predict the series trend of the student's knowledge absorption rate.

[0098] Specifically, the multimodal dataset generated in step S1 is used As input, an innovative individual knowledge absorption model (KAM) is constructed. This model captures students' real-time understanding of learning content and knowledge absorption rate, and predicts their future learning status. This model needs to be able to handle the heterogeneity and time dynamic changes of multimodal data, and provide deep insights into students' learning status.

[0099] Specifically, the input data is a multimodal dataset D multi The fused feature vector F in Fusion,t , these vectors represent the student status at each time point t, including EEG spectrum features, heart rate variability features, eye movement features and facial expression features. In order to capture the temporal dynamic characteristics of the data, the present invention first standardizes the data so that the mean of each feature is 0 and the variance is 1:

[0100]

[0101] in, is the normalized eigenvector, μ F and σ F are the mean and standard deviation of the eigenvectors, respectively.

[0102] In order to extract the time series features of students' learning status, the present invention uses an extended gated recurrent unit (GRU) model, called Multimodal Time Series Embedding Model (MTSEM). This model is able to process multimodal data and capture temporal dynamic changes, especially in terms of the impact of students' emotional and physiological reactions on learning.

[0103] In order to make full use of the relationship between different modal data, this paper adds a new modality attention mechanism (MAM) based on GRU. This mechanism assigns a weight to each modality of data, indicating its relative contribution to the knowledge absorption rate at the current time point. Let the feature vector of each modality be M i,t , then the modal attention weight α i,t The calculation is:

[0104]

[0105] Among them, α i,t is the attention weight of modality i at time point t; W m is the modality weight matrix, which is used to map the modality features to the attention weight space; M i,tis the eigenvector of mode i at time point t.

[0106] For example, suppose there are three modes of eigenvectors M 1,t =[0.2,0.3],M 2,t =[-0.1,0.5],M 3,t =[0.4,-0.2], and the weight matrix W m =[0.5,0.1]. The attention weight is calculated as α 1,t =0.3,α 2,t =0.4,α 3,t =0.3.

[0107] The output of the modal attention mechanism will be combined with the update formula of the GRU to update the student's learning state representation:

[0108]

[0109] Among them, h t is the hidden state of GRU, representing the learning state representation at time point t; z t is the update gate (UpdateGate), which is calculated in combination with the attention weight and is used to control the previous state h t-1 With the current candidate status combination of; is a candidate state, calculated using the current modality weight and input features.

[0110] Assume that the input feature vector is F Fusion,t =[0.5,-0.1,0.3,...], the previous state h t-1 =[0.2,0.4,0.6,...], update gate z t =[0.7,0.5,0.3,...] and candidate states The current hidden state h can be obtained t .

[0111] Furthermore, using the hidden state h output by GRU t The knowledge absorption rate (KAR) at each time point is calculated by using the modal attention weights. To this end, the present invention introduces a new knowledge absorption rate regularization term (KARRT), which introduces L2 regularization into the calculation formula to avoid overfitting and encourage weight sparsity:

[0112]

[0113] Among them, KAR tis the knowledge absorption rate at time point t; W k and b k is a trainable weight matrix and bias; σ is a sigmoid activation function that normalizes the output to between 0 and 1; It is the L2 regularization term, which controls the size of the weight to avoid overfitting.

[0114] For example, suppose the weight matrix W k =[0.2,0.5,-0.3], bias b k = 0.1, hidden state h t =[0.6,0.3,0.9], regularization parameter λ = 0.01, calculated

[0115] KAR t =σ(0.2·0.6+0.5·0.3-0.3·0.9+0.1)-0.01(0.2 2 +0.5 2 +(-0.3) 2 )≈0.515-0.011=0.504

[0116] Finally, the present invention uses the hidden state and attention mechanism of GRU to predict the future knowledge absorption trend. This is done by inputting the hidden state into an adaptive learning rate fully connected layer (AdaptiveLearningRateFullyConnectedLayer, ALR-FC). The output of the fully connected layer is the knowledge absorption rate prediction at several future time points, which is used to guide the adjustment of teaching strategies:

[0117] KAR t+1:t+k =ALR-FC(h t ,α t θ ALR-FC )

[0118] Among them, KAR t+1:t+k is the prediction of the knowledge absorption rate at k time points in the future; ALR-FC is a fully connected layer with an adaptive learning rate, with parameter θ ALR-FC ; α t is the modality attention weight, which is used to adjust the learning rate.

[0119] Assume that the hidden state h at the current time point t is t =[0.6,0.2,0.5,...], modal attention weight is α t = [0.3, 0.4, 0.3], the fully connected layer calculates the prediction for k = 2 time points in the future and outputs KAR t+1 =0.65,KAR t+2 =0.68.

[0120] Through the above steps, the present invention constructs an innovative individual knowledge absorption model that can reflect and predict students' learning status in real time. Each step and formula takes into account the special requirements in the patent scenario, such as the complexity of multimodal data, time dynamic characteristics, and the need for prediction accuracy and robustness. The knowledge absorption rate and trend prediction output by the model will be used for the next step of personalized teaching content generation to enhance the teaching interaction effect.

[0121] S3. Construct a feature map to embed the knowledge absorption rate and the predicted knowledge absorption rate sequence, construct a dynamic first adaptive content generator based on the feature map, design a joint loss function, and jointly train and optimize the discriminator and the first adaptive content generator according to the feature map to obtain a second adaptive content generator.

[0122] Specifically, the input data is the knowledge absorption rate sequence KAR generated in step S2 1:t ={KAR1,KAR2,…,KAR t} and future forecast absorption rate trend KAR t+1:t+k These data reflect the students’ learning status at each point in time and possible trends in future learning.

[0123] The present invention first performs feature mapping on these input data to generate a state vector S t , which contains the current and predicted learning state information. This state vector is used to characterize the student's current learning state and trend.

[0124] S t =φ(KAR 1:t ,KAR t+1:t+k )

[0125] Among them, φ is the feature mapping function, which is used to transform the knowledge absorption rate and trend into a comprehensive state representation.

[0126] Furthermore, in order to generate personalized teaching content, the present invention designs a dynamic adaptive content generator (DACG). The generator is based on the state vector S t Generate the characteristic vector C of teaching content t , the feature vector describes the teaching content attributes (such as difficulty, type, media format, etc.) that are suitable for the current time point t.

[0127] The present invention uses a variant of Generative Adversarial Network (GAN) to generate teaching content feature vectors. Specifically, a Knowledge-Enhanced Generator (KEG) is used to generate content features, and a Learning State Discriminator (LSD) is used to determine whether the generated content is suitable for the current learning state. The output of the generator is expressed by the following formula:

[0128] C t =KEG(S t θ G )

[0129] Among them, C t is the generated teaching content feature vector, S t is the state vector, θ G is the parameter of the generator. The generator is based on the input state vector S t Generate a content feature vector that meets the students' current learning needs.

[0130] Furthermore, in order to train the generator and the discriminator, the present invention introduces a joint loss function L(G,D). This loss function combines the adaptability score of the content (the degree of matching with the student's knowledge absorption state) and the content diversity regularization term (ensuring the diversity of generated content and avoiding mode collapse). The output of the discriminator is used to update the parameters of the generator so that the generator can generate content features that better meet the needs of students.

[0131]

[0132] Among them, G(S t ) is the content feature generated by the generator; D(G(S t )) is the adaptability score of the discriminator to the generated content; is the average content feature vector in the training set, which is used to calculate content diversity; λ is the regularization parameter that controls the impact of the diversity term.

[0133] For example, suppose the generator output is C t =[0.6,0.8,0.1], the discriminator is on the true state vector S t The output of is 0.9, the output of generated content is 0.1, and the diversity regularization term is The joint loss function can be calculated as:

[0134] L(G,D)=log(0.9)+log91-0.1)+0.1×0.05=-0.1053

[0135] Furthermore, once the training is completed, the present invention uses the generator to generate teaching content features C that adapt to the student's learning status in real time. t , and transform these features into actual teaching content through pre-defined content libraries and mapping rules. t Indicates a content with medium difficulty and video type. The system will select teaching videos that meet these characteristics from the content library.

[0136] The system is based on the new knowledge absorption rate KAR t+1 Feedback is used to dynamically adjust the content to ensure that the generated teaching content can continue to adapt to the students' learning status.

[0137] C t+1 =KEG(S t+1 θ G ),where S t+1 =φ(KAR 1:t+1 ,KAR t+2:t+k+1 )

[0138] Assume that at time t+1, the new state vector S t+1 is the updated knowledge absorption rate sequence KAR 1:t+1 and predicted future absorption rate trends KAR t+2:t+k+1 Generate, the generator uses the updated state to generate new content features C t+1 .

[0139] Through this solution, the dynamic adaptive teaching content generation system can generate and adjust teaching content in real time according to students' learning status, ensuring that each student's learning experience is personalized and can maximize their knowledge absorption rate.

[0140] S4. Obtain the teaching content feature vector generated by the second adaptive content generator, and map it to the student's social behavior characteristics to obtain a social feature vector, use the social feature vector to build a student social network, build a first dynamic social weight update model based on the student social network, and design a group interaction optimization algorithm to adjust the weight of the first dynamic social weight update model to obtain a second dynamic social weight update model, generate a new social interaction strategy based on the second dynamic social weight update model, and dynamically adjust the interaction recommendation and group allocation between students.

[0141] Specifically, the input data is the teaching content feature vector C generated in step 3 t , and the student's knowledge absorption rate sequence KAR 1:t First, the present invention maps these data to the social behavior characteristics of students to generate a social feature vector F social,t , represents the status of students in social networks.

[0142] F social,t =ψ(C t ,KAR 1:t )

[0143] Among them, ψ is a feature mapping function that converts teaching content characteristics and knowledge absorption rate into social behavior characteristics.

[0144] Further, using the generated social feature vector F social,t , the present invention builds a social network between students. Each node represents a student, and the edge weight represents the strength of the social connection between students. The present invention designs a dynamic social weight update model (DynamicSocialWeightUpdateModel, DSWUM) to update the edge weight in real time according to the students' knowledge absorption rate and social behavior. The edge weight w between nodes in the social network i,j,t Calculated by the following formula:

[0145] w i,j,t =α·sim(F social,i,t ,F social,j,t )+β·|KAR i,t -KAR j,t |

[0146] Among them, w i,j,t is the edge weight between student i and student j at time point t; α and β are weight coefficients used to balance the impact of social behavior similarity and knowledge absorption rate differences; sim(F social,i,t ,F social,j,t ) is the similarity of social characteristics between student i and student j at time point t; |KAR i,t -KAR j,t | is the difference in knowledge absorption rate between student i and student j at time point t.

[0147] For example, suppose the social feature vectors of student i and student j are F social,i,t =[0.6,0.8] and F social,j,t =[0.5,0.7], the knowledge absorption rates are KAR i,t =0.7 and KAR j,t = 0.6, weight coefficients are α = 0.5 and β = 0.3, then the edge weight w i,j,t =0.5·sim([0.6,0.8],[0.5,0.7])+0.3·|0.7-0.6|. Assuming the similarity sim([0.6,0.8],[0.5,0.7])=0.95, then w i,j,t =0.5·0.95+0.3·0.1=0.505.

[0148] Furthermore, based on the constructed social network, the present invention uses the Group Interaction Optimization Algorithm (GIOA) to optimize the interaction between students. The core of the algorithm is to promote effective interaction between students by adjusting the weights of the edges, increase the interaction between students with higher knowledge absorption rates and students with lower absorption rates, and achieve the effect of complementary learning. The optimization process minimizes the following objective function L opt accomplish:

[0149]

[0150] Among them, L opt is the optimization objective function, which is used to minimize the difference in knowledge absorption rate among students in the social network; E is the edge set in the social network; w i,j,t is the weight of edge i and j; KAR i,t and KAR j,t is the knowledge absorption rate of students i and j at time point t.

[0151] For example, if the edge set E contains two edges (i, j) and (k, l), their weights are w i,j,t = 0.5 and w k,l,t =0.8, the knowledge absorption rate is KAR i,t =0.7, KAR j,t =0.6, KAR k,t =0.8 and KAR l,t =0.5, then the objective function is:

[0152]

[0153] Furthermore, based on the optimization results, the present invention generates new social interaction strategies to improve learning effects by adjusting the interactive recommendations and group assignments between students. The system will continuously monitor the students’ knowledge absorption rate KAR t+1 and updating the social network model to ensure that student interactions remain at an optimal level.

[0154] The updated edge weight calculation formula is:

[0155] w i,j,t+1 =w i,j,t +γ·Δw i,j,t

[0156] Among them, w i,j,t+1 is the edge weight between student i and student j at time point t+1; γ is the learning rate, which is used to control the step size of weight update; Δw i,j,t is the change in edge weight, output by the optimization algorithm.

[0157] S5. Construct an adaptive learning curve model based on the second dynamic social weight update model, and design a feedback mechanism to update the interactive recommendations and group assignments between students.

[0158] Specifically, the input data includes the edge weights w output by the social network model i,j,t And the updated knowledge absorption rate sequence KAR i,1:t First, the present invention extracts the learning behavior characteristics of each student and generates a feature vector F learn,i,t , used to describe the learning behavior and status of student i at time point t.

[0159] F learn,i,t =η(w i,j,t ,KAR i,1:t )

[0160] Among them, η is the feature extraction function, which is used to convert the edge weights and knowledge absorption rate of the social network into learning behavior features.

[0161] Furthermore, using the extracted learning behavior features F learn,i,t , the present invention constructs an adaptive learning curve model (Adaptive Learning Curve Model, ALCM) to reflect the learning progress and knowledge mastery of students at different time points. The model uses a dynamic Bayesian network (Dynamic Bayesian Network, DBN) to capture the changes in students' learning status and potential dependencies in time series. The core of the learning curve model is to predict future learning performance based on students' knowledge absorption rate and social behavior characteristics. The state transition equation of the model is:

[0162]

[0163] Among them, KAR i,t+1 is the knowledge absorption rate of student i at time point t+1; γ and δ are model parameters used to balance the current absorption rate and new learning contribution; λ is the learning rate parameter used to adjust the impact of learning behavior characteristics on the absorption rate; F learn,i,t is the learning behavior characteristic of student i at time point t.

[0164] For example, suppose that the current knowledge absorption rate of student i is KAR i,t =0.7, the learning behavior characteristic is F learn,i,t =1.2, model parameters are γ=0.6, δ=0.4, λ=0.5, then the future knowledge absorption rate is predicted to be:

[0165] KAR i,t+1 =0.6·0.7+0.4·(1-e -0.5·1.2)≈0.42+0.4·(1-0.5488)=0.42+0.1805=0.6005

[0166] Furthermore, based on the adaptive learning curve model, the present invention designs a real-time feedback mechanism to provide students with personalized learning suggestions and strategy adjustments. The feedback mechanism is based on the "Learning Efficiency Function" (LEF), which is used to evaluate each student's current learning status and potential improvement directions. The learning efficiency function is defined as:

[0167]

[0168] Among them, E i,t is the learning efficiency of student i at time point t; KAR i,t+1 and KAR i,t is the knowledge absorption rate of student i at time points t+1 and t; Δt is the time interval; ω is the weight parameter used to control the impact of behavioral characteristic differences on learning efficiency; is the average learning behavior characteristic.

[0169] For example, suppose the knowledge absorption rate of student i changes from KAR i,t =0.6 increases to KAR i,t+1 =0.7, the learning behavior characteristic is F learn,i,t =1.2, the average learning behavior characteristics are Time interval Δt = 1, weight parameter ω = 0.3, then the learning efficiency is:

[0170]

[0171] Furthermore, based on the calculation results of the learning efficiency function, the system generates personalized feedback suggestions for students to help them optimize their learning strategies. The system will continuously monitor students’ learning progress and knowledge absorption rate, and dynamically adjust the learning curve model parameters to ensure that the model can accurately reflect students’ learning status and improvement directions.

[0172] The dynamic update formula of model parameters is:

[0173]

[0174] Among them, θ t+1 and θ t are the model parameters at time points t+1 and t; η is the learning rate; It is the gradient of the loss function to the model parameters, which is used to guide parameter updates.

[0175] S6. Construct a blockchain network to encrypt and upload the data output by the second dynamic social weight update model, and construct a learning feedback smart contract to automatically perform data verification, storage and feedback optimization, and design a learning data consensus algorithm to judge the reception of data blocks; wherein, the learning data consensus algorithm is a voting mechanism for all nodes to decide whether to accept new data blocks, and new data blocks will be accepted only if all nodes vote in favor.

[0176] Specifically, the input is the output data of the adaptive learning curve model and feedback mechanism in step 5, which includes the student's knowledge absorption rate sequence KAR i,1:t and learning efficiency E i,t The present invention uploads this data to the blockchain network to ensure the security and transparency of the data, and uses smart contracts to optimize the feedback mechanism. The following are the specific steps of the solution:

[0177] The input data includes the learning data feature vector F data,i,t =(KAR i,1:t ,E i,t ). First, the present invention encrypts these data to generate a ciphertext F' data,i,t To ensure the security of data before uploading it to the blockchain. The present invention uses a method based on elliptic curve cryptography (ECC) to encrypt data.

[0178] F' data,i,t =ECC_encrypt(F data,i,t ,key pub )

[0179] Among them, F' data,i,t is the encrypted data of student i at time point t; F data,i,t is the unencrypted learning data feature vector; key pub It is the public key of the blockchain network and is used for data encryption.

[0180] For example, suppose F data,i,t =(0.7,0.5),key pub =12345, then the encrypted data F' data,i,t =ECC_encrypt((0.7,0.5),12345)=(0.9,0.8).

[0181] Furthermore, in order to manage the secure sharing and feedback optimization of learning data on the blockchain, the present invention defines a learning feedback smart contract (LFSC). The smart contract automatically performs data verification, storage, and feedback optimization. The core of the smart contract is to verify the authenticity and integrity of the uploaded data and perform feedback optimization according to predefined rules. The core verification formula of the contract is:

[0182]

[0183] Among them, V(F' data,i,t ,key priv ) is the verification function of the smart contract; key priv is the private key of the blockchain network, used for data decryption; ? means that the first item is equal to the second item, and then determines whether the third item is equal to the first two items.

[0184] The contract verifies the integrity of the data by decrypting the encrypted data and comparing it with the original data. data,i,t =(0.9,0.8),key priv =54321, the decryption result is ECC_decrypt((0.9,0.8),54321)=(0.7,0.5), which is the same as the original data F data,i,t =(0.7,0.5) matches, the verification passes.

[0185] Furthermore, once the data is verified, the smart contract stores the data on the blockchain and allocates the data according to the learning efficiency E. i,t Perform dynamic feedback optimization. The present invention designs a feedback optimization function (FeedbackOptimizationFunction, FOF) to adjust the learning curve model parameters and optimize the learning strategy. The feedback optimization function is defined as:

[0186]

[0187] Among them, θ t+1 is the model parameter at time point t+1; θ t is the model parameter at time point t; μ is the optimization step size; It is the gradient of learning efficiency to model parameters and is used to guide parameter updates.

[0188] For example, suppose E i,t =0.04, model parameter θ t =0.5, optimization step size μ=0.1, and the gradient is calculated The updated model parameters are:

[0189] θt+1 =0.5+0.1·0.3=0.53

[0190] Furthermore, the blockchain network uses the Learning Data Consensus Algorithm (LDCA) to ensure the secure sharing and transparency of data. All nodes use a voting mechanism to decide whether to accept a new data block. The learning data contained in each data block is verified by multiple signatures to ensure that the data cannot be tampered with. The data update process uses the following formula:

[0191] D t+1 =D t +σ·ΔD t

[0192] Among them, D t+1 is the learning data state at time point t+1; D t is the learning data state at time point t; σ is the consensus parameter used to control the data update step size; ΔD t It is the change in data state, output by the consensus algorithm.

[0193] Through this solution, blockchain-based secure sharing and feedback optimization of learning data can dynamically adjust learning strategies according to students' learning status and feedback, ensuring the security, transparency and immutability of learning data.

[0194] In another embodiment of the present invention, an AI-based teaching interaction system is provided, the system comprising:

[0195] The teaching modality data collection module 101 is used to deploy sensor equipment to collect multimodal data of students and preprocess the multimodal data, and perform feature extraction and fusion on the preprocessed data to obtain a comprehensive feature vector, and store the comprehensive feature vector as a multimodal data set; wherein the comprehensive feature vector includes EEG spectrum features, heart rate variability features, eye movement features and facial expression features;

[0196] The knowledge absorption analysis module 102 is used to receive the multimodal data set sent by the sensor and construct an individual absorption model according to the multimodal data set to calculate the knowledge absorption rate of the student at each time point and predict the trend of the knowledge absorption rate sequence of the student;

[0197] The teaching content output module 103 is used to construct a feature map to embed the knowledge absorption rate and the predicted knowledge absorption rate sequence, construct a dynamic first adaptive content generator according to the feature map, and design a joint loss function to jointly train and optimize the discriminator and the first adaptive content generator according to the feature map to obtain a second adaptive content generator;

[0198] The teaching interaction module 104 is used to obtain the teaching content feature vector generated by the second adaptive content generator, and map it to the social behavior feature of the student to obtain a social feature vector, use the social feature vector to build a student social network, build a first dynamic social weight update model based on the student social network, and design a group interaction optimization algorithm to adjust the weight of the first dynamic social weight update model to obtain a second dynamic social weight update model, generate a new social interaction strategy based on the second dynamic social weight update model, and dynamically adjust the interaction recommendation and group allocation between students;

[0199] An interactive feedback module 105 is used to construct an adaptive learning curve model according to the second dynamic social weight update model, and to design a feedback mechanism to update interactive recommendations and group assignments between students;

[0200] The data encryption module 106 is used to construct a blockchain network to encrypt and upload the data output by the second dynamic social weight update model, and to construct a learning feedback smart contract to automatically perform data verification, storage and feedback optimization, and to design a learning data consensus algorithm to judge the reception of data blocks; wherein, the learning data consensus algorithm is that all nodes decide whether to accept new data blocks through a voting mechanism, and only when all nodes vote in favor of accepting new data blocks will they accept new data blocks;

[0201] In the student social network, each node represents a student, and the weight of the edge represents the strength of the social connection between students; the first dynamic social weight update model updates the weight of the edge in real time according to the students' knowledge absorption rate and social behavior; the weight of the edge is calculated as follows:

[0202] w i,j,t =α·sim(F social,i,t ,F social,j,t )+β·|KAR i,t -KAR j,t |

[0203] Among them, w i,j,t represents the edge weight between student i and student j at time point t; α and β represent weight coefficients, which are used to balance the influence of social behavior similarity and knowledge absorption rate difference; sim(F social,i,t ,F social,j,t ) represents the similarity of social characteristics between student i and student j at time point t; |KAR i,t -KAR j,t | represents the difference in knowledge absorption rate between student i and student j at time point t;

[0204] A group interaction optimization algorithm is designed to increase the interaction between students with the highest knowledge absorption rate and students with the lowest knowledge absorption rate. The group interaction optimization algorithm is expressed as follows:

[0205]

[0206] Among them, L opt represents the optimization objective function, which is used to minimize the difference in knowledge absorption rate among students in the social network; E represents the edge set in the social network; w i,j,t represents the weight of edge i and j; KAR i,t and KAR j,t represents the knowledge absorption rate of students i and j at time point t;

[0207] The edge weights are updated according to the output of the group interaction optimization algorithm to obtain a second dynamic social weight update model. The edge weights of the second dynamic social weight update model are expressed as follows:

[0208] w i,j,t+1 =w i,j,t +γ·Δw i,j,t

[0209] Among them, w i,j,t+1 represents the edge weight between student i and student j at time point t+1; γ represents the learning rate, which is used to control the step size of weight update; Δw i,j,t Indicates the change in edge weight.

[0210] It should be noted that the workflow described above is merely illustrative and does not limit the scope of protection of the present invention. In practical applications, technicians in this field can select part or all of them according to actual needs to achieve the purpose of the present embodiment, and no limitation is made here.

[0211] In addition, for technical details that are not described in detail in this embodiment, reference can be made to the parameter operation method provided in any embodiment of the present invention, and will not be repeated here.

[0212] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or system including the element.

[0213] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0214] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as a read-only memory / random access memory, a magnetic disk, or an optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0215] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. The AI-based teaching interactive method is characterized by: The method comprises: S1. Deploy sensor equipment to collect multimodal data of students and preprocess the multimodal data, extract and fuse features of the preprocessed data to obtain a comprehensive feature vector, and store the comprehensive feature vector as a multimodal data set; wherein the comprehensive feature vector includes EEG spectrum features, heart rate variability features, eye movement features and facial expression features; S2, receiving the multimodal data set sent by the sensor and constructing an individual absorption model based on the multimodal data set to calculate the student's knowledge absorption rate at each time point and predict the student's knowledge absorption rate sequence trend; S3, constructing a feature map to embed the knowledge absorption rate and the predicted knowledge absorption rate sequence, constructing a dynamic first adaptive content generator according to the feature map, and designing a joint loss function to jointly train and optimize the discriminator and the first adaptive content generator according to the feature map to obtain a second adaptive content generator; S4. Obtain the teaching content feature vector generated by the second adaptive content generator, and map it to the social behavior characteristics of the students to obtain a social feature vector, use the social feature vector to build a student social network, build a first dynamic social weight update model based on the student social network, and design a group interaction optimization algorithm to adjust the weight of the first dynamic social weight update model to obtain a second dynamic social weight update model, generate a new social interaction strategy based on the second dynamic social weight update model, and dynamically adjust the interaction recommendation and group allocation between students; S5. Construct an adaptive learning curve model based on the second dynamic social weight update model, and design a feedback mechanism to update the interactive recommendations and group assignments between students; S6. Construct a blockchain network to encrypt and upload the data output by the second dynamic social weight update model, and construct a learning feedback smart contract to automatically perform data verification, storage and feedback optimization, and design a learning data consensus algorithm to judge the reception of data blocks; wherein, the learning data consensus algorithm is that all nodes decide whether to accept new data blocks through a voting mechanism, and only if all nodes vote in favor of accepting new data blocks will they accept new data blocks; In the student social network, each node represents a student, and the weight of the edge represents the strength of the social connection between students; the first dynamic social weight update model updates the weight of the edge in real time according to the students' knowledge absorption rate and social behavior; the weight of the edge is calculated as follows: In i,j,t =α sim(F social,i,t ,F social,j,t )+β |PENALTY i,t -PENALTY j,t | Among them, w i,j,t represents the edge weight between student i and student j at time point t; α and β represent weight coefficients, which are used to balance the influence of social behavior similarity and knowledge absorption rate difference; sim(F social,i,t ,F social,j,t ) represents the similarity of social characteristics between student i and student j at time point t; |KAR i,t -KAR j,t | represents the difference in knowledge absorption rate between student i and student j at time point t; A group interaction optimization algorithm is designed to increase the interaction between students with the highest knowledge absorption rate and students with the lowest knowledge absorption rate. The group interaction optimization algorithm is expressed as follows: Among them, L opt represents the optimization objective function, which is used to minimize the difference in knowledge absorption rate among students in the social network; E represents the edge set in the social network; w i,j,t represents the weight of edge i and j; KAR i,t and KAR j,t represents the knowledge absorption rate of students i and j at time point t; The edge weights are updated according to the output of the group interaction optimization algorithm to obtain a second dynamic social weight update model. The edge weights of the second dynamic social weight update model are expressed as follows: In i,j,t+1 =in i,j,t +γ Δw i,j,t Among them, w i,j,t+1 represents the edge weight between student i and student j at time point t+1; γ represents the learning rate, which is used to control the step size of weight update; Δw i,j,t Indicates the change in edge weight.

2. The AI-based teaching interactive method according to claim 1, characterized in that: The sensor device includes an EEG headset, a heart rate monitor, an eye tracker and a facial expression recognition camera; the preprocessing includes time synchronization processing, applying an adaptive bandpass filter to the data of the EEG headset to remove noise in the EEG data, using a window smoothing algorithm to process the data of the heart rate monitor, using a peak detection algorithm to identify gaze and saccade processing on the eye movement data to obtain eye movement features, and using a convolutional neural network to extract features from micro-expression data to obtain facial expression features; the feature extraction includes using short-time Fourier transform on the data of the EEG headset to obtain power spectrum features of the α band and the β band in different time windows to obtain EEG spectrum features, and using the standard deviation of the RR interval to calculate the heart rate variability features; A joint feature space mapping method is used to map different modality data into a shared feature space.

3. According to the AI-based teaching interactive method according to claim 1, it is characterized in that: The individual absorption model is constructed as follows: The student state at each moment on the comprehensive feature vector is standardized, and a multimodal time series embedding model that introduces a modal attention weight mechanism is constructed. The output of the modal attention mechanism is combined with the update formula of the multimodal time series embedding model to obtain an updated representation of the student's learning state; wherein the modal attention weight mechanism assigns a weight to each modal data to indicate its relative contribution to the knowledge absorption rate at the current time point, and the multimodal time series embedding model processes the multimodal data and captures temporal dynamic changes; The knowledge absorption rate at each time point is calculated and the future knowledge absorption trend is predicted based on the hidden states and modal attention weights output by the multimodal time series embedding model.

4. According to the AI-based teaching interactive method according to claim 3, it is characterized in that: In the calculation of the knowledge absorption rate at each time point, a knowledge absorption rate regularization term based on L2 regularization is introduced to avoid overfitting and encourage weight sparsity; the prediction of future knowledge absorption trends is performed by inputting the hidden state into a fully connected layer with an adaptive learning rate, and the output of the fully connected layer is the knowledge absorption rate prediction at several future time points.

5. According to the AI-based teaching interactive method according to claim 1, it is characterized in that: The first adaptive content generator generates a first generated teaching content feature vector according to the state vector and the parameters of the generator; the joint loss function L(G, D) is expressed as follows: Among them, G(S t ) represents the content features generated by the generator; D(G(S t )) represents the adaptability score of the discriminator to the generated content; S t represents the state vector; D represents the learning state discriminator; G represents the knowledge enhancement generator; represents the average content feature vector in the training set, which is used to calculate content diversity; λ represents the regularization parameter, which controls the influence of the diversity term; The knowledge enhancement generator generates content features and inputs them into a learning state discriminator for learning state discrimination. The learning state discriminator is used to discriminate whether the generated content is suitable for the current learning state.

6. According to the AI-based teaching interactive method according to claim 5, it is characterized in that: The first adaptive content generator is trained using the joint loss function L(G,D) to obtain the second adaptive content generator, which is used to generate teaching content features adapted to the student's learning status in real time, and converts these features into a second generated teaching content feature vector through a predefined content library and mapping rules.

7. According to the AI-based teaching interactive method according to claim 1, it is characterized in that: The S5 specifically includes: S501, extracting the learning behavior characteristics of each student, generating a learning behavior characteristic vector for describing the learning behavior and state of student i at time point t; S502. Construct an adaptive learning curve model based on a dynamic Bayesian network according to the learning behavior feature vector, capture the changes in students' learning status and potential dependencies in the time series according to students' knowledge absorption rate and social behavior characteristics, and reflect students' learning progress and knowledge mastery at different time points; S503. Designing a real-time feedback mechanism based on the adaptive learning curve model to provide students with personalized learning suggestions and strategy adjustments; wherein the real-time feedback mechanism is set according to a learning efficiency function, and the learning efficiency function is used to evaluate each student's current learning status and potential improvement direction; S504. Generate personalized feedback suggestions for students based on the calculation results of the learning efficiency function, and optimize learning strategies.

8. According to the AI-based teaching interactive method according to claim 7, it is characterized in that: The learning efficiency function is expressed as follows: Among them, E i,t represents the learning efficiency of student i at time point t; KAR i,t+1 and KAR i,t represents the knowledge absorption rate of student i at time points t+1 and t; Δt represents the time interval; ω represents the weight parameter, which is used to control the impact of behavioral characteristic differences on learning efficiency; represents the average learning behavior characteristics; The knowledge absorption rate represents the state transition equation of the adaptive learning curve model and is calculated as follows: Among them, KAR i,t+1 represents the knowledge absorption rate of student i at time point t+1; γ and δ represent model parameters, which are used to balance the current absorption rate and new learning contribution; λ represents the learning rate parameter, which is used to adjust the impact of learning behavior characteristics on the absorption rate; F learn,i,t Represents the learning behavior characteristics of student i at time point t.

9. According to the AI-based teaching interactive method according to claim 1, it is characterized in that: The encryption adopts elliptic curve cryptography algorithm to encrypt data; the verification formula of the learning feedback smart contract is expressed as follows: Among them, F' data,i,t represents the encrypted data of student i at time point t; F data,i,t represents the unencrypted learning data feature vector; V(F' data,i,t ,key priv ) represents the verification function of the smart contract; key priv Represents the private key of the blockchain network, which is used for data decryption; ECC_decrypt represents the elliptic curve cryptography algorithm.

10. The AI-based teaching interactive system is characterized by: The system comprises: A teaching modality data collection module is used to deploy sensor equipment to collect students' multimodal data and preprocess the multimodal data, and perform feature extraction and fusion on the preprocessed data to obtain a comprehensive feature vector, and store the comprehensive feature vector as a multimodal data set; wherein the comprehensive feature vector includes EEG spectrum features, heart rate variability features, eye movement features and facial expression features; The knowledge absorption analysis module is used to receive the multimodal data set sent by the sensor and build an individual absorption model based on the multimodal data set to calculate the student's knowledge absorption rate at each time point and predict the trend of the student's knowledge absorption rate sequence; The teaching content output module is used to construct a feature map to embed the knowledge absorption rate and the predicted knowledge absorption rate sequence, construct a dynamic first adaptive content generator according to the feature map, and design a joint loss function to jointly train and optimize the discriminator and the first adaptive content generator according to the feature map to obtain a second adaptive content generator; A teaching interaction module is used to obtain the teaching content feature vector generated by the second adaptive content generator, and map it to the social behavior characteristics of the students to obtain a social feature vector, use the social feature vector to build a student social network, build a first dynamic social weight update model based on the student social network, and design a group interaction optimization algorithm to adjust the weight of the first dynamic social weight update model to obtain a second dynamic social weight update model, generate a new social interaction strategy based on the second dynamic social weight update model, and dynamically adjust the interaction recommendation and group allocation between students; An interactive feedback module is used to construct an adaptive learning curve model based on the second dynamic social weight update model, and to design a feedback mechanism to update interactive recommendations and group assignments between students; A data encryption module is used to build a blockchain network to encrypt and upload the data output by the second dynamic social weight update model, and to build a learning feedback smart contract to automatically perform data verification, storage and feedback optimization, and to design a learning data consensus algorithm to judge the reception of data blocks; wherein, the learning data consensus algorithm is that all nodes decide whether to accept new data blocks through a voting mechanism, and only when all nodes vote in favor will they accept new data blocks; In the student social network, each node represents a student, and the weight of the edge represents the strength of the social connection between students; the first dynamic social weight update model updates the weight of the edge in real time according to the students' knowledge absorption rate and social behavior; the weight of the edge is calculated as follows: In i,j,t =α sim(F social,i,t ,F social,j,t )+β |PENALTY i,t -PENALTY j,t | Among them, w i,j,t represents the edge weight between student i and student j at time point t; α and β represent weight coefficients, which are used to balance the influence of social behavior similarity and knowledge absorption rate difference; sim(F social,i,t ,F social,j,t ) represents the similarity of social characteristics between student i and student j at time point t; |KAR i,t -KAR j,t | represents the difference in knowledge absorption rate between student i and student j at time point t; A group interaction optimization algorithm is designed to increase the interaction between students with the highest knowledge absorption rate and students with the lowest knowledge absorption rate. The group interaction optimization algorithm is expressed as follows: Among them, L opt represents the optimization objective function, which is used to minimize the difference in knowledge absorption rate among students in the social network; E represents the edge set in the social network; w i,j,t represents the weight of edge i and j; KAR i,t and KAR j,t represents the knowledge absorption rate of students i and j at time point t; The edge weights are updated according to the output of the group interaction optimization algorithm to obtain a second dynamic social weight update model. The edge weights of the second dynamic social weight update model are expressed as follows: In i,j,t+1 =in i,j,t +γ Δw i,j,t Among them, w i,j,t+1 represents the edge weight between student i and student j at time point t+1; γ represents the learning rate, which is used to control the step size of weight update; Δw i,j,t Indicates the change in edge weight.

Images