Social network trust evaluation method and system based on multi-view neural network

Abstract

The invention discloses a social network trust evaluation method and system based on a multi-view graph neural network, and the method comprises the steps: obtaining social network behavior information, constructing social behavior network graphs of different views, and constructing a user trust relationship network graph according to the trust relationship between users; based on social behavior network diagrams of different perspectives, a multi-head attention mechanism is built to extract influences of different types of neighbor nodes on target user node features and fuse the influences, and social attribute features are output; on the basis of the user trust relationship network diagram, utilizing an attention mechanism to fuse neighbor information of the nodes to form trust features of the user nodes; and in combination with the social attribute features and the trust features of the user nodes, performing multi-view feature fusion, and outputting a predicted trust relationship. The problems that a traditional trust evaluation method depends on a predefined model and is poor in universality, and an existing trust relation prediction scheme based on a graph neural network is difficult to obtain a trust relation graph and is poor in feature fusion effect are solved.

Description

Technical Field

[0001] The present invention belongs to the technical field of trust assessment, and in particular to a social network trust assessment method and system based on a multi-view graph neural network. Background Art

[0002] Trust assessment is the process of evaluating trust by quantifying the attributes that affect trust. It has many applications in sensor networks, social networks, mobile crowdsourcing, blockchain and other fields. Traditional trust assessment schemes are based on heuristic trust theory. Usually, based on user attributes and scores between users, the above information is integrated based on a predefined trust model to obtain a comprehensive trust score of the user or a trust relationship between users. The effectiveness of such methods depends greatly on the predefined model, has poor universality, and cannot be flexibly applied to different scenarios. In view of the limitations of traditional trust assessment algorithms, some scholars have applied machine learning algorithms to trust assessment, and automatically processed trust-related data through machine learning models, extracted user trust features and predicted trust relationships between users. Traditional trust assessment algorithms based on machine learning are usually based on matrix decomposition theory. By representing the trust relationship between users as a matrix and decomposing the matrix, the node trust features and the trusted features are obtained, and then the matrix multiplication is performed to obtain the complete trust relationship matrix. For example, Fazeli et al. explored the effectiveness of different measurement methods in trust relationship evaluation based on matrix decomposition and proposed a matrix decomposition trust evaluation algorithm based on implicit trust score. By incorporating implicit trust score instead of explicit trust relationship into matrix decomposition, the model was experimentally verified to achieve similar performance without explicit trust score. Similar work was also applied in [6]. The matrix decomposition-based method has shown good effectiveness, but because it is based on a shallow model, it cannot mine the deep complex features of nodes and it is difficult to effectively discover the complex nonlinear mapping relationship between user trust features and trust relationships.

[0003] In response to the above problems, some scholars have introduced deep learning models for the extraction of trust features and the prediction of trust relationships. By modeling the trust relationship as a graph and processing it with a graph convolutional neural network, this type of model has achieved remarkable results. For example: Guardian uses a graph convolutional neural network to process and fuse the two trust features of the user separately, and builds a trust relationship prediction scheme based on a graph convolutional neural network; on this basis, KGTrust further integrates other social networks besides the trust relationship network to expand the trust relationship graph and achieve a certain performance improvement; GATTrust integrates trust features from different fields based on the attention mechanism to improve trust prediction performance; some work further considers the dynamic nature of trust and builds a dynamic prediction mechanism for trust relationships based on dynamic graph neural networks, such as TrustGuard. The above work has solved the problem of trust relationship prediction very well.

[0004] However, the above work has the following problems: 1. Most of the existing trust relationship prediction schemes based on graph neural networks are oriented towards trust relationship graphs. However, in practical applications, trust relationship graphs are very difficult to obtain, especially trust relationship networks with sufficient density to support accurate trust evaluation. 2. Some works based on external knowledge to supplement the missing trust relationship (such as KGTrust) fuse trust features and social attribute features through the attention mechanism. However, the two are highly heterogeneous in feature distribution. Although direct fusion can enrich trust features, it cannot achieve complementary enhancement of features. It is difficult to effectively extract the consistent information hidden in social attributes and trust features. Summary of the invention

[0005] The purpose of the present invention is to provide a social network trust evaluation method and system based on a multi-view graph neural network to solve the above problems.

[0006] To achieve the above object, the present invention adopts the following technical solutions: In a first aspect, the present invention provides a social network trust evaluation method based on a multi-view graph neural network, comprising: Obtain social network behavior information, construct social behavior network diagrams from different perspectives, and construct user trust relationship network diagrams based on the trust relationship between users; Based on social behavior network graphs from different perspectives, a multi-head attention mechanism is built to extract the influence of different types of neighbor nodes on the target user node features and fuse them to output social attribute features; Based on the user trust relationship network graph, the attention mechanism is used to fuse the node's neighbor information to form the trust characteristics of the user node; Combining social attribute features and trust features of user nodes, multi-view feature fusion is performed to output predicted trust relationships.

[0007] Furthermore, the obtaining of social network behavior information and constructing social behavior network graphs from different perspectives includes: Collect social data of target users in different social behaviors / platforms; Each social behavior / platform is processed separately to form a user relationship graph from each perspective. With the user as the center, the interaction relationship between the user and the social platform objects is collected, and these interaction objects are regarded as nodes and the interaction relationship as edges to construct a social relationship graph from this perspective. , ; According to the interaction type, node attributes and other information, different methods are used to process Initial attribute features of midpoints and edge attributes.

[0008] Furthermore, constructing a user trust relationship network diagram based on the trust relationship between users includes: Standardize the trust characteristics between users to form discrete trust levels, and model the trust relationship between users as a directed weighted graph; Initialize the out-degree trust features according to the trust relationship between the user and other users. and in-degree trust characteristics , forming a complete trust relationship network diagram .

[0009] Furthermore, the social behavior network graph based on different perspectives builds a multi-head attention mechanism to extract the influence of different types of neighbor nodes on the target user node features and fuse them to output social attribute features, including: Given a node , which is in the perspective The corresponding figure The neighbor set is ,node , and nodes The edge of the record , the feature is recorded as ,but The attention parameter The calculation is performed using the following formula:

[0010] Get attention score Afterwards, given The layer output is , then The output of the layer is:

[0011] Based on the attention mechanism, we further design a multi-head attention mechanism, which includes Independent attentions extract features and perform feature fusion respectively. Assume that The features extracted by the attention mechanism are , then The output of the layer is:

[0012] When all is completed After extracting the features of the layer, the final features are output .

[0013] Furthermore, based on the user trust relationship network graph, the neighbor information of the node is fused using the attention mechanism to form the trust feature of the user node, including: Build The graph convolutional neural network of the layer is based on the user's initial in-degree trust feature The initial feature , assuming that The output of the layer is ,if Existence right Trust Edge , then it is believed that yes The in-degree neighbors of , first calculate the attention score as follows:

[0014] Based on the score, the fused features are calculated:

[0015] go through After the layer is trained, its in-degree features are output

[0016] Build The graph convolutional neural network of the layer is based on the user's initial in-degree trust feature The initial feature , assuming that The output of the layer is ,if Existence right Trust Edge , then it is believed that yes The out-degree neighbors of , first calculate the attention score as follows:

[0017] Based on the score, the fused features are calculated:

[0018] go through After the layer is trained, its in-degree features are output

[0019] Trust feature fusion and output: and After concatenation, input the fully connected network and activation layer, output Trust characteristics .

[0020] Furthermore, the combining of social attribute features and trust features of user nodes, performing multi-view feature fusion, and outputting predicted trust relationships includes: by As a user No. perspective, forming the user The characteristics of the perspective, namely , summarized as follows:

[0021] Next, we construct a trust feature fusion module for multi-view learning. Merge into a unified feature , and construct the reconstruction loss Used for subsequent model training; Trust relationship prediction based on multi-layer perceptron: given a user and users and its characteristics , , input it into the multi-layer perceptron to output the predicted trust relationship .

[0022] Furthermore, the loss function is: View-dependent feature encoder : Based on the neural network, a multi-layer deep encoder is constructed. User features for each view Encoded as latent features ; The user The features of each view are fused to obtain the user fusion feature ; View-dependent feature decoder :Build a multi-layer deep decoder based on a neural network, and form an autoencoder structure with the above deep encoder to combine the fused features Decode to reconstruct Features of a view ; Constructing autoencoder-based reconstruction loss , which is defined as follows:

[0023] Extract a set of edges from the trust relationship for end-to-end training, using the real trust relationship For supervision, the loss function is constructed as follows:

[0024] At the same time, combined with the loss function in (5), the final loss function is constructed as follows: .

[0025] In a second aspect, the present invention provides a social network trust evaluation system based on a multi-view graph neural network, comprising: The data acquisition module is used to obtain social network behavior information, construct social behavior network diagrams from different perspectives, and construct user trust relationship network diagrams based on the trust relationship between users; The social attribute feature acquisition module is used to build a multi-head attention mechanism based on social behavior network graphs from different perspectives to extract the influence of different types of neighbor nodes on the target user node features and fuse them to output social attribute features; The trust feature acquisition module is used to fuse the neighbor information of nodes based on the user trust relationship network graph using the attention mechanism to form the trust feature of the user node; The trust relationship output module is used to combine social attribute features and trust features of user nodes, perform multi-view feature fusion, and output predicted trust relationships.

[0026] In a third aspect, the present invention provides a computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the social network trust assessment method based on a multi-view graph neural network when executing the computer program.

[0027] In a fourth aspect, the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the social network trust assessment method based on a multi-view graph neural network are implemented.

[0028] Compared with the prior art, the present invention has the following technical effects: This technical solution proposes a social network trust assessment method based on a multi-perspective graph neural network, aiming to solve the problems of traditional trust assessment methods relying on pre-defined models and poor universality, as well as the difficulties in obtaining trust relationship graphs and poor feature fusion effects faced by existing trust relationship prediction schemes based on graph neural networks.

[0029] By constructing user social behavior network graphs from different perspectives, the solution can integrate user behavior information from multiple social platforms and different types. This helps to supplement and enrich trust features through user social behavior data in scenarios where trust relationship data is scarce or insufficient, thereby improving the accuracy of trust assessment.

[0030] Using multi-view learning technology, the scheme can mine consistent trust information contained in different view graphs. This consistent information helps to eliminate redundant and irrelevant features, complete and enhance trust features, and further improve the accuracy of trust assessment.

[0031] The heterogeneous graph attention mechanism in the scheme can effectively process node information and edge information and adapt to different types of graph structures. This means that the scheme can be flexibly applied to various complex social network environments, including different types of social behavior graphs and trust relationship graphs, improving the adaptability and generalization ability of the model.

[0032] Through multi-view learning technology, the scheme can fuse trust features from different perspectives to form a more comprehensive and accurate user trust representation. This fusion strategy helps capture the trust features of users in different social scenarios, thereby improving the accuracy of trust relationship prediction.

[0033] The autoencoder structure in the scheme optimizes model training by constructing reconstruction loss to ensure that the fused features can accurately reconstruct the features of the original perspective. This optimization strategy helps to retain important information in the original data and further improve the accuracy of trust assessment.

[0034] In summary, this technical solution achieves effective evaluation of trust relationships in social networks by introducing multi-view graph neural networks and heterogeneous graph attention mechanisms. It not only solves the limitations of traditional trust evaluation methods, but also improves the adaptability and prediction accuracy of the model. In addition, this solution also has potential advantages such as interpretability and scalability, providing new ideas and methods for future trust evaluation research. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 It is a flow chart of the present invention. DETAILED DESCRIPTION

[0036] The present invention is further described below in conjunction with the accompanying drawings: Example 1, please refer to Figure 1, a social network trust assessment method based on multi-view graph neural network, including: Obtain social network behavior information, construct social behavior network diagrams from different perspectives, and construct user trust relationship network diagrams based on the trust relationship between users; Based on social behavior network graphs from different perspectives, a multi-head attention mechanism is built to extract the influence of different types of neighbor nodes on the target user node features and fuse them to output social attribute features; Based on the user trust relationship network graph, the attention mechanism is used to fuse the node's neighbor information to form the trust characteristics of the user node; Combining social attribute features and trust features of user nodes, multi-view feature fusion is performed to output predicted trust relationships.

[0037] First, by acquiring rich social network behavior information, such as the frequency of interaction between users, content sharing, likes, comments, etc., a social behavior network diagram from different perspectives was constructed. This multi-dimensional data collection method ensures that the user's behavior patterns in social networks can be fully captured, providing a solid foundation for subsequent trust evaluation.

[0038] This method can extract and fuse the influence of different types of neighbor nodes on the target user node features in a refined manner. This means that it can not only identify which neighbor nodes have important influences on the target user, but also accurately quantify these influences, thereby generating more representative social attribute features.

[0039] In the user trust relationship network graph, this method integrates the node's neighbor information through the attention mechanism to form the trust feature of the user node. This mechanism can automatically pay attention to the nodes that are closely related to the user's trust relationship, and adjust the trust feature representation of the user node accordingly, thereby improving the accuracy of trust evaluation.

[0040] This method combines social attribute features and user node trust features into multi-view feature fusion, making full use of feature information from different sources and types. This fusion strategy not only enhances the expressiveness of features, but also improves the model's adaptability to complex social environments, thereby achieving more accurate prediction of trust relationships.

[0041] Since this method uses a multi-perspective graph neural network structure and combines the attention mechanism for feature extraction and fusion, it is highly robust to data noise and outliers. At the same time, through training and learning on large-scale social network data, this method can learn more general social rules and trust patterns, thereby improving its generalization ability in different social scenarios.

[0042] Accurate trust evaluation is crucial for many applications in social networks, such as recommendation systems, social search, fraud detection, etc. The trust evaluation results provided by this method can provide strong data support for these applications, helping them better meet user needs, improve user experience, and effectively prevent potential risks and threats.

[0043] In summary, the social network trust assessment method based on multi-view graph neural network achieves accurate assessment of trust relationships by comprehensively capturing social behavior information, accurately extracting user features, effectively integrating trust relationships, and integrating multi-view features. This method not only improves the accuracy and robustness of trust assessment, but also provides strong data support for social network applications, and has broad application prospects and value.

[0044] Embodiment 2, the present invention provides a social network trust assessment method based on a multi-view graph neural network, specifically comprising: Social behavior information composition construction module: by pre-processing the user's behavior information on social networks (such as likes, purchases, comments, etc.), according to different types of user social behaviors, construct user social behavior network diagrams from different perspectives ; Trust relationship network diagram construction module: Build a user trust relationship network diagram based on the trust relationship between users , and based on a single user Build its initial trust characteristics ; User node social attribute feature extraction module based on graph attention mechanism: Construct a user feature extraction module based on graph neural network, build a multi-head attention mechanism to extract the influence of different types of neighbor nodes on the target user node features and fuse them. , and output the user The corresponding feature embedding

[0045] User trust attribute feature extraction module based on graph attention mechanism: construct the user's initial features based on the trust relationship of the nodes and trust relationship network , using the attention mechanism to fuse the node’s neighbor information and form the trust feature of the user node ; Trust feature fusion module based on multi-view learning: As a user No. perspective, forming the user The characteristics of the perspective, namely , summarized as follows:

[0046] Next, we construct a trust feature fusion module for multi-view learning. Merge into a unified feature , and construct the reconstruction loss For subsequent model training Trust relationship prediction based on multi-layer perceptron: given a user and users and its characteristics , , input it into the multi-layer perceptron to output the predicted trust relationship

[0047] Model training: Extract a set of edges from the trust relationship for end-to-end training, using the real trust relationship For supervision, the loss function is constructed as follows:

[0048] At the same time, combined with the loss function in (5), the final loss function is constructed as follows:

[0049] The social behavior information composition construction module mainly includes the following processes: Data collection: Collect social data of target users in different social behaviors / platforms; Social relationship graph construction: Each social behavior / platform is processed separately to form a user relationship graph from each perspective, with users (such as ) as the center, collect its interactive relationships with other objects (such as commodities, articles, etc.), regard these interactive objects as nodes, and regard the interactive relationships as edges, and construct a social relationship graph from this perspective. ; Initial social node / edge attribute modeling: Different methods are used to process the interaction type, node attributes, and other information. Initial attribute features of midpoints and edge attributes, for example: The BERT model can be used to process the text evaluation information of the product to obtain edge features; The trust relationship network diagram construction module mainly includes the following processes: User trust relationship extraction: standardize the trust characteristics between users to form discrete trust levels, and model the trust relationship between users as a directed weighted graph; User trust feature initialization: For example, according to the trust relationship between the user and other users and the trust relationship between other users and the user, the out-degree trust features are initialized respectively. and in-degree trust characteristics , thus forming a complete trust relationship network diagram ; The user node social attribute feature extraction module based on the graph attention mechanism mainly includes the following processes: Attention mechanism: given a node , which is in the perspective The corresponding figure The neighbor set is ,node , and nodes The edge of the record , the feature is recorded as ,but The attention parameter The calculation is performed using the following formula:

[0050] Get attention score Afterwards, given The layer output is , then The output of the layer is:

[0051] Multi-head attention mechanism: Based on the attention mechanism, a multi-head attention mechanism is further designed, which includes Each independent attention extracts features and performs feature fusion. Assume that The features extracted by the attention mechanism are , then The output of the layer is:

[0052] Social attribute feature output: When all After extracting the features of the layer, the final features are output ; The user trust attribute feature extraction module based on the graph attention mechanism mainly includes the following processes: In-degree trust feature embedding module based on multi-layer graph convolutional network: Building The graph convolutional neural network of the layer is based on the user's initial in-degree trust feature The initial feature , assuming that The output of the layer is ,if Existence right Trust Edge , then it is believed that yes The in-degree neighbors of , first calculate the attention score as follows:

[0053] Based on the score, the fused features are calculated:

[0054] go through After the layer is trained, its in-degree features are output

[0055] Out-degree trust feature embedding module based on multi-layer graph convolutional network: Construction The graph convolutional neural network of the layer is based on the user's initial in-degree trust feature The initial feature , assuming that The output of the layer is ,if Existence right Trust Edge , then it is believed that yes The out-degree neighbors of , first calculate the attention score as follows:

[0056] Based on the score, the fused features are calculated:

[0057] go through After the layer is trained, its in-degree features are output

[0058] Trust feature fusion and output: and After concatenation, input the fully connected network and activation layer, output Trust characteristics ; The trust feature fusion module based on multi-view learning mainly includes the following processes: View-dependent feature encoder : Based on the neural network, a multi-layer deep encoder is constructed. User features for each view Encoding as latent features ; Multi-view feature fusion module: The features of each view are fused to obtain the user fusion feature ; View-dependent feature decoder :Build a multi-layer deep decoder based on a neural network, and form an autoencoder structure with the above deep encoder to combine the fused features Decode to reconstruct Features of a view ; Constructing autoencoder-based reconstruction loss , which is defined as follows:

[0059] The present invention is aimed at the trust evaluation problem in social network platforms including ubiquitous social networks, mobile crowdsourcing, etc. The platform mainly includes a social network platform and For ease of representation, this patent uses Represents a social platform, which mainly provides online social services, such as instant messaging, information, shopping and other services. It is also responsible for predicting trust relationships based on users' trust relationships and social behaviors to better serve users. The algorithms involved in this patent are deployed in this middle, Maintaining social behavior graphs of users in different social scenarios And the user's trust relationship diagram ; The user set is represented as , user use The social services provided by users are used to form trust or distrust information about other users based on personal preferences. A social network trust assessment method based on multi-view graph neural network is proposed. This method uses multi-view graphs to model user trust relationship graphs and cross-platform and cross-scenario social relationship graphs. It uses graph neural networks and multi-view representation learning technology to effectively obtain user trust features from graphs of different perspectives and mine consistent trust information. While using multiple social relationship graphs to supplement the limited trust features caused by insufficient trust relationships, the mining of consistent information effectively ensures the exclusion of redundant and irrelevant features, thus completing and enhancing trust features.

[0060] In yet another embodiment of the present invention, a social network trust evaluation system based on a multi-view graph neural network is provided, which can be used to implement the above-mentioned social network trust evaluation method based on a multi-view graph neural network. Specifically, the system includes: Obtain social network behavior information, construct social behavior network diagrams from different perspectives, and construct user trust relationship network diagrams based on the trust relationship between users; Based on social behavior network graphs from different perspectives, a multi-head attention mechanism is built to extract the influence of different types of neighbor nodes on the target user node features and fuse them to output social attribute features; Based on the user trust relationship network graph, the attention mechanism is used to fuse the node's neighbor information to form the trust characteristics of the user node; Combining social attribute features and trust features of user nodes, multi-view feature fusion is performed to output predicted trust relationships.

[0061] The division of modules in the embodiments of the present invention is schematic and is only a logical function division. There may be other division methods in actual implementation. In addition, each functional module in each embodiment of the present invention may be integrated into one processor, or may exist physically separately, or two or more modules may be integrated into one module. The above-mentioned integrated modules may be implemented in the form of hardware or in the form of software functional modules.

[0062] In another embodiment of the present invention, a computer device is provided, the computer device including a processor and a memory, the memory is used to store a computer program, the computer program includes program instructions, and the processor is used to execute the program instructions stored in the computer storage medium. The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc., which are the computing core and control core of the terminal, which are suitable for implementing one or more instructions, and are specifically suitable for loading and executing one or more instructions in the computer storage medium to implement the corresponding method flow or corresponding function; the processor described in the embodiment of the present invention can be used for the operation of the social network trust evaluation method based on a multi-view graph neural network.

[0063] In another embodiment of the present invention, the present invention also provides a storage medium, specifically a computer-readable storage medium (Memory), which is a memory device in a computer device for storing programs and data. It is understandable that the computer-readable storage medium here can include both built-in storage media in a computer device and, of course, extended storage media supported by the computer device. The computer-readable storage medium provides a storage space, which stores the operating system of the terminal. In addition, one or more instructions suitable for being loaded and executed by a processor are also stored in the storage space, and these instructions can be one or more computer programs (including program codes). It should be noted that the computer-readable storage medium here can be a high-speed RAM memory or a non-volatile memory, such as at least one disk memory. The processor can load and execute one or more instructions stored in the computer-readable storage medium to implement the corresponding steps of the social network trust evaluation method based on a multi-view graph neural network in the above embodiment.

[0064] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0065] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0066] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0067] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0068] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A social network trust assessment method based on multi-view graph neural network, characterized in that: include: Obtain social network behavior information, construct social behavior network diagrams from different perspectives, and construct user trust relationship network diagrams based on the trust relationship between users; Based on social behavior network graphs from different perspectives, a multi-head attention mechanism is built to extract the influence of different types of neighbor nodes on the target user node features and fuse them to output social attribute features; Based on the user trust relationship network graph, the attention mechanism is used to fuse the node's neighbor information to form the trust characteristics of the user node; Combining social attribute features and trust features of user nodes, multi-view feature fusion is performed to output predicted trust relationships.

2. The social network trust assessment method based on multi-view graph neural network according to claim 1 is characterized in that: The step of obtaining social network behavior information and constructing social behavior network graphs from different perspectives includes: Collect social data of target users in different social behaviors / platforms; Each social behavior / platform is processed separately to form a user relationship graph from each perspective. With the user as the center, the interaction relationship between the user and the social platform objects is collected, and these interaction objects are regarded as nodes and the interaction relationship as edges to construct a social relationship graph from this perspective. , ; According to the interaction type, node attributes and other information, different methods are used to process Initial attribute features of midpoints and edge attributes.

3. The social network trust assessment method based on multi-view graph neural network according to claim 1 is characterized in that: The step of constructing a user trust relationship network diagram based on the trust relationship between users includes: Standardize the trust characteristics between users to form discrete trust levels, and model the trust relationship between users as a directed weighted graph; Initialize the out-degree trust features according to the trust relationship between the user and other users. and in-degree trust characteristics , forming a complete trust relationship network diagram .

4. The social network trust assessment method based on multi-view graph neural network according to claim 1, characterized in that: Based on the social behavior network graph from different perspectives, a multi-head attention mechanism is built to extract the influence of different types of neighbor nodes on the target user node features and fuse them to output social attribute features, including: Given a node , which is in the perspective The corresponding figure The neighbor set is ,node , and nodes The edge of the record , the feature is recorded as ,but The attention parameter The calculation is performed using the following formula: Get attention score Afterwards, given The layer output is , then The output of the layer is: Based on the attention mechanism, we further design a multi-head attention mechanism, which includes Independent attentions extract features and perform feature fusion respectively. Assume that The features extracted by the attention mechanism are , then The output of the layer is: When all is completed After extracting the features of the layer, the final features are output .

5. The social network trust assessment method based on multi-view graph neural network according to claim 1, characterized in that: Based on the user trust relationship network graph, the neighbor information of the node is fused using the attention mechanism to form the trust feature of the user node, including: Build The graph convolutional neural network of the layer is based on the user's initial in-degree trust feature The initial feature , assuming that The output of the layer is ,if Existence right Trust Edge , then it is believed that yes The in-degree neighbors of , first calculate the attention score as follows: Based on the score, the fused features are calculated: go through After the layer is trained, its in-degree features are output Build The graph convolutional neural network of the layer is based on the user's initial in-degree trust feature The initial feature , assuming that The output of the layer is ,if Existence right Trust Edge , then it is believed that yes The out-degree neighbors of , first calculate the attention score as follows: Based on the score, the fused features are calculated: go through After the layer is trained, its in-degree features are output Trust feature fusion and output: and After concatenation, input the fully connected network and activation layer, output Trust characteristics .

6. The social network trust assessment method based on multi-view graph neural network according to claim 1, characterized in that: The method combines the social attribute features and the trust features of the user nodes to perform multi-view feature fusion and output the predicted trust relationship, including: by As a user No. perspective, forming the user The characteristics of the perspective, namely , summarized as follows: Next, we construct a trust feature fusion module for multi-view learning. Merge into a unified feature , and construct the reconstruction loss Used for subsequent model training; Trust relationship prediction based on multi-layer perceptron: Given a user and users and its characteristics , , input it into the multi-layer perceptron to output the predicted trust relationship .

7. The social network trust assessment method based on multi-view graph neural network according to claim 6, characterized in that: Loss function: View-dependent feature encoder : Based on the neural network, a multi-layer deep encoder is constructed. User features for each view Encoded as latent features ; The user The features of each view are fused to obtain the user fusion feature ; View-dependent feature decoder :Build a multi-layer deep decoder based on a neural network, and form an autoencoder structure with the above deep encoder to combine the fused features Decode to reconstruct Features of a view ; Constructing autoencoder-based reconstruction loss , which is defined as follows: Extract a set of edges from the trust relationship for end-to-end training, using the real trust relationship For supervision, the loss function is constructed as follows: At the same time, combined with the loss function in (5), the final loss function is constructed as follows: 。 8. A social network trust evaluation system based on multi-view graph neural network, characterized in that: include: The data acquisition module is used to obtain social network behavior information, construct social behavior network diagrams from different perspectives, and construct user trust relationship network diagrams based on the trust relationship between users; The social attribute feature acquisition module is used to build a multi-head attention mechanism based on social behavior network graphs from different perspectives to extract the influence of different types of neighbor nodes on the target user node features and fuse them to output social attribute features; The trust feature acquisition module is used to fuse the neighbor information of nodes based on the user trust relationship network graph using the attention mechanism to form the trust feature of the user node; The trust relationship output module is used to combine social attribute features and trust features of user nodes, perform multi-view feature fusion, and output predicted trust relationships.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the social network trust evaluation method based on a multi-view graph neural network as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the social network trust evaluation method based on a multi-view graph neural network as described in any one of claims 1 to 7 are implemented.

Images