A graph convolutional heart disease detection method fusing trust loss and decision reclassification

Abstract

The application provides a kind of fusion trust loss and decision reclassification method for graph convolution heart disease detection.Belongs to medical information intelligent processing technology field, solved in heart disease detection, the problem of uncertain data affecting classification accuracy.Its technical scheme is: first, collect heart disease information data sample set, then, the above data is pretreated, and an adjacency matrix is constructed;Then, the processed heart disease information data is input into the graph convolution neural network GCN for model training;Finally, the model enters the test stage, gets the preliminary classification result and enters the decision layer, filters out the uncertain node information according to the threshold value, and maps it using the Gaussian kernel, so that it is high-dimensional separable, and achieves the effect of reclassification.

Description

Technical Field

[0001] The present invention relates to the field of intelligent processing of medical information, and more particularly to a graph convolutional heart disease detection method integrating trust loss and decision reclassification. Background Art

[0002] Accurately identifying heart disease types is crucial for improving early disease detection, enhancing treatment outcomes, and reducing misdiagnoses. Traditional heart disease classification methods rely primarily on physician experience and rule-based systems. These methods often rely on the judgment of medical experts and are susceptible to subjective factors, limiting accuracy and generalization.

[0003] Researchers are currently trying to classify heart disease through deep learning methods. However, although existing deep learning-based models have achieved good results in some tasks, they still have some problems.

[0004] In the medical field, data is often noisy, incomplete, and uncertain. This can lead traditional deep learning models to produce unstable predictions when processing uncertain data. This is particularly true when data quality is low or there are significant differences between samples. Furthermore, heart disease classification relies not only on data from a single patient but also requires consideration of similarities between patients. Existing models often overlook this inter-patient similarity information during classification, failing to fully utilize this information for model optimization.

[0005] How to solve the above problems is the subject faced by the present invention. Summary of the Invention

[0006] The purpose of the present invention is to provide a graph convolutional heart disease detection method that integrates trust loss and decision reclassification. This method integrates trust adaptive weighted loss and negative log-likelihood loss, and has good effects in processing complex data. In addition, a re-decision mechanism is introduced to ensure good results in processing uncertain data.

[0007] The present invention is achieved through the following measures: a graph convolutional heart disease detection method integrating trust loss and decision reclassification includes the following steps:

[0008] Step 1: Collect a sample set U of heart disease data, which includes heart disease feature information and heart disease categories. Give each data a unique ID, use similarity to select neighbors for each ID, and build an adjacency relationship.

[0009] Step 2: Preprocess the original sample set, convert the original categories into one-hot encoded labels, construct a sparse adjacency matrix based on the adjacency relationship, and divide the heart disease sample set data into training sample set, validation sample set, and test sample set;

[0010] Step 3: Input the above training sample data into the improved graph convolutional neural network. Based on the actual categories of heart disease, the graph convolutional neural network is trained. The trust-based adaptive loss function is integrated on the basis of the negative log-likelihood loss function. The output value of the new loss function is used for backpropagation to update the model parameters.

[0011] Step 4: Input the validation sample set data into the model and calculate the uncertain classification node threshold δ based on the classification accuracy of the last round of validation;

[0012] Step 5: Input the test sample set into the trained model to obtain the preliminary classification probability distribution Z = {z1,z2,z3,…z n}, the probability distribution enters the decision layer, the information of the uncertain nodes is filtered according to the uncertain classification node threshold δ, and the Gaussian kernel function is used to reclassify the uncertain nodes.

[0013] As a graph convolution heart disease detection method integrating trust loss and decision reclassification provided by the present invention, step 3 includes the following steps:

[0014] Step 3.1 Set the random seed seed to 42, the total number of training epochs to 200, the learning rate lr to 0.01, the dropout rate to 0.5, and set the model to enter training mode. In training mode, Dropout randomly drops some neurons. Input the sample data features X and the adjacency matrix A into the graph convolutional neural network. The formula is as follows:

[0015]

[0016] in W (0) is the weight from the input layer to the hidden layer, W (1) is the weight from the hidden layer to the output layer, Z={z1,z2,z3,…z n} is the probability distribution of n nodes corresponding to each category, and the nonlinear activation function ReLU of the hidden layer is expressed as:

[0017] ReLU(x i )=max(0,x i ) (2)

[0018] The output layer softmax function uses logarithmic transformation to output logarithmic probability distribution. The formula is as follows:

[0019]

[0020] where x i is the i-th component in the input vector x;

[0021] Step 3.2 Use the exp function to convert the logarithmic probability distribution Z = {z1,z2,z3,…z n}, and restored to the ordinary probability distribution P = {p1,p2,p3,…p n}, retaining the logarithmic probability distribution, the calculation expression is as follows:

[0022] p i =σ exp (x i )=exp(σ(x i )) (4)

[0023] Step 3.3 According to the probability distribution p of the i-th node i ={p i1 ,p i2 ,p i3 ,…p im}The maximum probability v for each node i,1 =max(p i1 ,p i2 ,p i3 ,…p im ) and the second highest probability v i,2 =max({p i1 ,p i2 ,p i3 ,…p im}\{v i,1}) Make the difference, the operation is:

[0024] Δ i =v i,1 -v i,2 (5)

[0025] Get the maximum difference matrix Δ=[Δ1,Δ2,…,Δ n ], the difference is used as the confidence of node classification;

[0026] Step 3.4 Given the logarithmic probability distribution z of the i-th node i ={z i1 ,z i2 ,z i3 ,…z im}, training set sample index idx train , the number of training set samples N, the true category label y of the i-th node i , calculate the negative log-likelihood loss, the function calculation formula is:

[0027]

[0028] Step 3.5 uses the maximum difference matrix Δ=[Δ1,Δ2,…,Δ n], the negative log-likelihood loss is given an adaptive weight by the Hadamard product to obtain the trust adaptive loss. The loss function formula is as follows:

[0029]

[0030] The trust adaptive loss function is combined with the negative log-likelihood loss function to obtain the final loss function, which is expressed as:

[0031]

[0032] The higher the node trust, the greater the impact on the model. The lower the node trust, the smaller the impact on the model. For nodes with uncertain properties, reduce their interference with the model and improve the model's classification ability for nodes with certain properties. Set λ = 0.2, perform backpropagation based on the loss value, update the model parameters, and output the loss and classification accuracy of each round of training.

[0033] As a graph convolution heart disease detection method integrating trust loss and decision reclassification provided by the present invention, step 4 includes the following steps:

[0034] Step 4.1 The model enters the evaluation mode, Dropout is disabled, all neurons participate in the calculation, the model uses the complete network structure for verification, the verification sample set data is input into the model, and the number of correctly classified samples is calculated. The calculation formula is:

[0035]

[0036] in, is the indicator function, when 1 when it is, otherwise 0;

[0037] Step 4.2 calculates the classification accuracy based on the number of correctly classified samples. The output is the ratio of the number of correctly classified samples to the total number of input samples. The calculation formula is as follows:

[0038]

[0039] Step 4.3: The classification accuracy of the last round of verification Rounded to two decimal places, the formula is:

[0040]

[0041] According to classification accuracy Calculate the threshold for uncertain classification nodes

[0042] As a graph convolution heart disease detection method integrating trust loss and decision reclassification provided by the present invention, step 5 includes the following steps:

[0043] Step 5.1 Input the test sample data into the trained model and retain the initial classification result y=[y1,y 2, …,y n ], calculate the maximum difference matrix Δ=[Δ1,Δ2,…,Δ n ], calculate the number of selected elements k according to the threshold δ min , the calculation expression is as follows:

[0044]

[0045] Select the smallest k from the maximum difference matrix min elements, record the difference value and the index in the original set, the operation expression is:

[0046]

[0047] where |S| = k min ;

[0048] Step 5.2 calculates the Euclidean distance matrix D between nodes based on the node feature X. The matrix form is as follows:

[0049]

[0050] Set σ = 1, input the Euclidean distance matrix D into the Gaussian kernel function, and obtain the similarity matrix S based on the Gaussian kernel. The matrix is ​​expressed as:

[0051]

[0052] Step 5.3: The neighborhood similarity matrix N is obtained by element-by-element multiplication of the adjacency matrix A and the similarity matrix S. The expression is:

[0053] N=A⊙S (16)

[0054] If there is an edge connection between node i and node j, then N ij It is the similarity value S between node i and node j ij , if there is no edge connection, then N ij =0;

[0055] Step 5.4 Get the similarity S between the current uncertain node i and other adjacent nodes i =[S i1 ,S i2 ,…,S ik ], get the label of the adjacent node, the operation expression is as follows:

[0056] Yi ={y j |j∈N(i)} (17)

[0057] Extract the probability value of node i corresponding to the adjacent node label in the probability distribution to obtain the weight vector W i =[W i1 ,W i2 ,…,W ik ], similarity S i =[S i1 ,S i2 ,…,S ik ] are weighted to obtain weighted similarity The operation is as follows:

[0058]

[0059] Step 5.5 finds the node index with the largest similarity in the weighted similarity matrix, as described below:

[0060]

[0061] Similarity node index i max Get reclassification labels Update y=[y1,y2,…,y n ], loop step 5.4 to step 5.5 until all uncertain nodes are reclassified, and output the final classification result.

[0062] Compared with the prior art, the present invention has the following beneficial effects:

[0063] 1. The present invention integrates trust adaptive loss with negative log-likelihood loss, assigns a trust value to each node according to the uncertainty measurement, reduces the influence of nodes with low trust values ​​on the model, and enhances the influence of nodes with high trust values ​​on the model, which can effectively improve classification accuracy and provide medical personnel with more accurate detection basis.

[0064] 2. The present invention introduces a reclassification mechanism to screen out uncertain nodes according to the threshold, and reclassifies the uncertain nodes through implicit dimensionality increase of the Gaussian kernel function, thereby improving the model's ability to process uncertain data and further improving the classification accuracy.

[0065] 3. In view of the characteristics of heart disease data such as fuzzy category boundaries and uncertainty, the present invention proposes a graph convolution heart disease detection method that integrates trust loss and decision reclassification. This method starts from both the loss function and probability distribution, fully considers the uncertainty of data and probability, uses trust weighting to "denoise" the model, and uses Gaussian kernel function for reclassification, which effectively improves the classification accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention.

[0067] Figure 1 This is the overall framework diagram of the graph convolutional heart disease detection method that integrates trust loss and decision reclassification in the present invention.

[0068] Figure 2 This is a flowchart of the algorithm improved based on graph convolution in the present invention.

[0069] Figure 3 This is a diagram of the decision reclassification architecture of the present invention. DETAILED DESCRIPTION

[0070] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. Of course, the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0071] Example 1

[0072] See also Figures 1 to 3 This embodiment provides a technical solution for detecting heart disease using graph convolutional neural network that integrates trust loss and decision reclassification, including the following steps:

[0073] Step 1: Collect 200 heart disease data samples, including heart disease feature information and heart disease categories. Give each data a unique ID, use similarity to select neighbors for each ID, and build an adjacency relationship.

[0074] Step 2: Preprocess the original sample set, convert the original categories into one-hot encoded labels, construct a sparse adjacency matrix based on the adjacency relationship, and divide the heart disease sample set data into training sample set, validation sample set, and test sample set;

[0075] Step 3: Input the above training sample data into the improved graph convolutional neural network. Based on the actual categories of heart disease, the graph convolutional neural network is trained. The trust-based adaptive loss function is integrated on the basis of the negative log-likelihood loss function. The output value of the new loss function is used for backpropagation to update the model parameters.

[0076] Step 4: Input the validation sample set data into the model and calculate the uncertain classification node threshold δ based on the classification accuracy of the last round of validation;

[0077] Step 5: Input the test sample set into the trained model to obtain the preliminary classification probability distribution Z = {z1,z2,z3,…z n}, the probability distribution enters the decision layer, the information of the uncertain nodes is filtered according to the uncertain classification node threshold δ, and the Gaussian kernel function is used to reclassify the uncertain nodes.

[0078] Preferably, step 3 includes the following steps:

[0079] Step 3.1 Set the random seed seed to 42, the total number of training epochs to 200, the learning rate lr to 0.01, the dropout rate to 0.5, and set the model to enter training mode. In training mode, Dropout randomly drops some neurons. Input the sample data features X and the adjacency matrix A into the graph convolutional neural network. The formula is as follows:

[0080]

[0081] in W (0) is the weight from the input layer to the hidden layer, W (1) is the weight from the hidden layer to the output layer, Z={z1,z2,z3,…z n} is the probability distribution of n nodes corresponding to each category, and the nonlinear activation function ReLU of the hidden layer is expressed as:

[0082] ReLU(x i )=max(0,x i ) (4)

[0083] The output layer softmax function uses logarithmic transformation to output logarithmic probability distribution. The formula is as follows:

[0084]

[0085] where x i is the i-th component in the input vector x;

[0086] Step 3.2 Use the exp function to convert the logarithmic probability distribution Z = {z1,z2,z3,…z n}, and restored to the ordinary probability distribution P = {p1,p2,p3,…p n}, retaining the logarithmic probability distribution, the calculation expression is as follows:

[0087] p i =σ exp (x i )=exp(σ(x i )) (6)

[0088]

[0089] Step 3.3 According to the probability distribution p of the i-th node i ={pi1 ,p i2 ,p i3 ,…p im}maximal probability v i,1 = max(p i1 ,p i2 ,p i3 ,…p im}and the second maximal probability v i,2 = max(p i1 ,p i2 ,p i3 ,…p im}\{v i,1}are differentiated, and the operation is:

[0090] Δ i = v i,1 - v i,2 (8)

[0091]

[0092] The maximum difference matrix Δ = [Δ1, Δ2, …, Δ n ] is obtained, and the difference value is used as the trust value of node classification;

[0093] Step 3.4, given the log probability distribution z i = {z i1 ,z i2 ,z i3 ,…z im}of the ith node, the training set sample index idx train , the number of training set samples N, and the true class label y i of the ith node, the negative log-likelihood loss is calculated, and the function calculation formula is:

[0094]

[0095] Step 3.5, using the maximum difference matrix Δ = [Δ1, Δ2, …, Δ n ], the Hadamard product is used to give adaptive weight to the negative log-likelihood loss, and the trust adaptive loss is obtained, and the loss function formula is as follows:

[0096]

[0097] The trust adaptive loss function and the negative log-likelihood loss function are fused to obtain the final loss function, which is represented as:

[0098]

[0099] The higher the node trust value, the greater the impact on the model. The lower the node trust value, the smaller the impact on the model. For nodes with uncertain properties, reduce their interference with the model and improve the model's classification ability for nodes with certain properties. Set λ = 0.2, perform backpropagation based on the loss value, update the model parameters, and output the loss and classification accuracy of each round of training.

[0100] Preferably, step 4 includes the following steps:

[0101] Step 4.1 The model enters the evaluation mode, Dropout is disabled, all neurons participate in the calculation, the model uses the complete network structure for verification, the verification sample set data is input into the model, and the number of correctly classified samples is calculated. The calculation formula is:

[0102]

[0103] in, is the indicator function, when 1 when it is, otherwise 0;

[0104] Step 4.2 calculates the classification accuracy based on the number of correctly classified samples. The output is the ratio of the number of correctly classified samples to the total number of input samples. The calculation formula is as follows:

[0105]

[0106] Step 4.3: The classification accuracy of the last round of verification Rounded to two decimal places, the formula is:

[0107]

[0108] According to classification accuracy Calculate the threshold for uncertain classification nodes

[0109] Preferably, step 5 includes the following steps:

[0110] Step 5.1 Input the test sample data into the trained model and retain the initial classification result y=[y1,y2,…,y n ], calculate the maximum difference matrix Δ=[Δ1,Δ2,…,Δ n ], calculate the number of selected elements k according to the threshold δ min , the calculation expression is as follows:

[0111]

[0112] Select the smallest k from the maximum difference matrix minAn element, record difference value and index in the original set, operation expression is:

[0113]

[0114] Where |S| = k min ;

[0115] Step 5.2 Calculate the Euclidean distance matrix D between nodes according to the node characteristics X, the matrix form is as follows:

[0116]

[0117] Set σ = 1, input the Euclidean distance matrix D into the Gaussian kernel function, get the similarity matrix S based on the Gaussian kernel, the matrix representation is:

[0118]

[0119] Step 5.3 Get the neighborhood similarity matrix N by element-wise multiplication of the adjacency matrix A and the similarity matrix S, the expression is:

[0120]

[0121] If there is an edge between node i and node j, then N ij is the similarity value S ij between node i and node j, if there is no edge, then N ij = 0;

[0122] Step 5.4 Get the similarity S i = [S i1 , S i2 , …, S ik ] of the current uncertain node i and other adjacent nodes, get the label of the adjacent node, the operation expression is as follows:

[0123] S i = [S i1 , S i2 , …, S ik ] = [0.1977 0.1982 … 0.1978] (21)

[0124] Y i = {y j |j∈N(i)} = [0 3 … 3] (22)

[0125] Extract the probability value of the label of the adjacent node corresponding to node i in the probability distribution, get the weight vector W i = [W i1 , W i2 , …, W ik ], to the similarity Si =[S i1 ,S i2 ,…,S ik ] are weighted to obtain weighted similarity The operation is as follows:

[0126]

[0127] Step 5.5 finds the node index with the largest similarity in the weighted similarity matrix, as described below:

[0128]

[0129] Similarity node index i max Get reclassification labels Update y=[y1,y2,…,y n ], loop step 5.4 to step 5.5 until all uncertain nodes are reclassified, and output the final classification result.

[0130] Example 2

[0131] Referring to Example 1, this example uses the parameters and results calculated in Example 1 to compare with traditional methods to demonstrate the superiority of this example. In this specific comparison, we compared the accuracy of different models. The final results show that this example is superior to the traditional method.

[0132] 1. Traditional Model

[0133] With reference to relevant research at home and abroad, this embodiment selected the following comparison models:

[0134] 1) Deep Walk Model:

[0135] Deep Walk is a graph embedding method based on random walks. It maps graph nodes into a low-dimensional vector space, thereby learning node representations. It traverses the entire graph using random walks and then uses these random walk sequences to learn vector representations for the nodes. Specifically, Deep Walk first performs several random walks on each node, generating a series of random walk sequences. These sequences are then treated as sentences, and word embedding models are used to learn vector representations for the nodes.

[0136] 2) GCN model:

[0137] GCN operates on graph data, and its core concept is to propagate information through adjacency relationships within the graph structure. Through graph convolution, nodes not only retain their own information but also obtain information about their neighbors, thereby updating the node's feature representation at each layer. Increasing the number of GCN layers allows nodes to obtain information about their neighbors at greater distances. The graph convolution operation aggregates information about node neighbors through the adjacency matrix A, and the node features at each layer depend on the features of the previous layer.

[0138] 2. Comparative indicators

[0139] This example is used for classification tasks. In classification tasks, we use accuracy (denoted as ACC) as the evaluation metric. ACC is the ratio of the number of correctly classified samples to the total number of samples. It reflects the accuracy of the model's overall classification.

[0140] This embodiment selects an accuracy evaluation method to objectively compare this embodiment with other models, aiming to highlight the superiority of this embodiment.

[0141] 3. Comparison results

[0142] As can be seen from the table, this example performs exceptionally well on accuracy evaluation metrics, significantly outperforming other models. The classification accuracy of this example reached 82.5%, exceeding the 80.3% achieved by the GCN model and significantly exceeding the 67.9% achieved by the DeepWalk model. This demonstrates that this example achieves higher overall classification accuracy and is able to better distinguish between multiple classes of samples.

[0143] Table 1 Experimental comparison of the present invention, Deep Walk, and GCN models

[0144] method Dataset ACC (%) This article UCI 82.5 GCN UCI 80.3 Deep Walk UCI 67.9

[0145] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A graph convolutional heart disease detection method integrating trust loss and decision reclassification, characterized by The following steps are involved: Step 1: Collect heart disease data sample set ,The sample set includes the characteristic information of heart disease, ,and the categories of heart disease. Each data is given a unique ID, ,and the neighbors for each ID are selected using similarity to ,build an adjacency relationship; Step 2: Preprocess the original sample set, convert the original categories into one-hot encoded labels, construct a sparse adjacency matrix based on the adjacency relationship, and divide the heart disease sample set data into training sample set, validation sample set, and test sample set; Step 3: Input the above training sample data into the improved graph convolutional neural network. Based on the actual categories of heart disease, the graph convolutional neural network is trained. The trust-based adaptive loss function is integrated on the basis of the negative log-likelihood loss function. The output value of the new loss function is used for backpropagation to update the model parameters. The step 3 comprises the following steps: Step 3.1 Set the random seed seed=42, the total number of training epochs=200, the learning rate lr=0.01, the dropout rate Dropout=0.5, and set the model to enter the training mode. In the training mode, Dropout randomly drops a part of the neurons and transforms the sample data features into and the adjacency matrix Input graph convolutional neural network, the formula is as follows: ； in , , , is the weight from the input layer to the hidden layer, is the weight from the hidden layer to the output layer, for The nodes correspond to the probability distribution of each category, and the nonlinear activation function of the hidden layer Expressed as: ； Output layer The function uses logarithmic transformation to output logarithmic probability distribution, the formula is as follows: ； in is the input vector The Quantity Step 3.2 Use The function converts the logarithmic probability distribution , reduced to the ordinary probability distribution , retaining the logarithmic probability distribution, the calculation expression is as follows: ； Step 3.3 According to The probability distribution of nodes The maximum probability for each node and the second highest probability To do the difference, the operation is: ； Get the maximum difference matrix ,The difference is used as the trust value of node classification; Step 3.4 Known The logarithmic probability distribution of nodes , training set sample index , the number of training set samples , No. The true category labels of the nodes , calculate the negative log-likelihood loss, the function calculation formula is: ； Step 3.5 Using the maximum difference matrix , the negative log-likelihood loss is given an adaptive weight by the Hadamard product to obtain the trust adaptive loss. The loss function formula is as follows: ； The trust adaptive loss function is combined with the negative log-likelihood loss function to obtain the final loss function, which is expressed as: ； The higher the node trust, the greater the impact on the model; the lower the node trust, the smaller the impact on the model. For nodes with uncertain properties, reduce their interference with the model and improve the model's ability to classify nodes with certain properties. , perform backpropagation based on the loss value, update the model parameters, and output the loss and classification accuracy of each round of training; Step 4: Input the validation sample set data into the model and calculate the classification accuracy based on the last round of validation. Calculate the uncertainty classification node threshold ; Step 5: Input the test sample set into the trained model to obtain the preliminary classification probability distribution , the probability distribution enters the decision layer, according to the uncertainty classification node threshold Filter the information of uncertain nodes and use Gaussian kernel function to reclassify uncertain nodes.

2. The graph convolutional heart disease detection method integrating trust loss and decision reclassification according to claim 1 is characterized in that: The step 4 comprises the following steps: Step 4.1 The model enters evaluation mode, Dropout is disabled, all neurons participate in the calculation, and the model is verified using the complete network structure. The verification sample set data is input into the model, and the number of correctly classified samples is calculated using the following formula: ; in, is the indicator function, when 1 when it is, otherwise 0; Step 4.2 Based on the number of correctly classified samples , calculate the classification accuracy , the output is the ratio of the number of correctly classified input samples to the total number of input samples, and the calculation formula is as follows: ; Step 4.3: The classification accuracy of the last round of verification Rounded to two decimal places, the formula is: ; According to classification accuracy Calculate the threshold for uncertain classification nodes .

3. The graph convolutional heart disease detection method integrating trust loss and decision reclassification according to claim 1 is characterized in that: The step 5 comprises the following steps: Step 5.1 Input the test sample data into the trained model and retain the initial classification results , calculate the maximum difference matrix , according to the threshold Count the number of selected elements , the calculation expression is as follows: ; Select the smallest one from the maximum difference matrix elements, record the difference value and the index in the original set, the operation expression is: ; in ; Step 5.2 Based on node characteristics Calculate the Euclidean distance matrix between nodes , the matrix form is as follows: ; set up , the Euclidean distance matrix Input the Gaussian kernel function to obtain the similarity matrix based on the Gaussian kernel , the matrix is ​​represented as: ; Step 5.3 Through the adjacency matrix and similarity matrix The neighborhood similarity matrix is ​​obtained by element-by-element multiplication of , the expression is: ; If the node and nodes There are edges connecting them, then It's the node and nodes The similarity value between , if there is no edge connection, then ; Step 5.4 Get the current uncertain node Similarity with other adjacent nodes , get the label of the adjacent node, the operation expression is as follows: ; Extracting nodes from probability distribution The probability value corresponding to the adjacent node label is used to obtain the weight vector , for similarity Weighted to get weighted similarity , the operation is as follows: ; Step 5.5: Find the node index with the highest similarity in the weighted similarity matrix, as described below: ; Similarity node index Get reclassification labels ,renew , loop step 5.4 to step 5.5 until all uncertain nodes are reclassified and the final classification result is output.

Images