An intelligent real-time advertising push method based on multi-level graph neural network

Abstract

The present invention belongs to the technical field of advertising push and relates to an intelligent real-time advertising push method based on a multi-level graph neural network. The method first collects dynamic data and static data of a parking lot and performs preprocessing. Then, two graph structures, a global graph and a local graph, are designed. The self-attention mechanism is introduced to construct and train a multi-level graph neural network to obtain a trained model. Finally, the trained model is used to calculate the embedding of each license plate and frequently visited places, thereby generating a predicted score. For each user, advertisements are sorted according to the predicted score, and the advertisement with the highest score is selected as the push result. The multi-level graph structure enables the model to capture user behavior patterns more comprehensively and meticulously, thereby improving the accuracy and relevance of recommendations. In addition, the computational complexity is reduced, the computational efficiency of the model is improved, and it can run efficiently in actual application scenarios with limited resources, meeting the needs of real-time recommendations.

Description

Technical Field

[0001] The present invention belongs to the technical field of advertisement push, and relates to an intelligent real-time advertisement push method based on a multi-level graph neural network. Background Art

[0002] Modern parking lots experience high volumes of traffic and frequent movement. Drivers passing through the gates are often attentive to their surroundings, particularly the content displayed on the gate displays. This presents an excellent opportunity for advertising. Given the short dwell time of vehicles, ads must precisely capture drivers' attention within a short period of time to maximize conversion rates and effectiveness. However, there are currently no algorithms specifically tailored to recommend ads based on parking vehicle behavior. This results in low accuracy and relevance in advertising, hindering maximum effectiveness.

[0003] Existing advertising recommendation algorithms are mostly used in e-commerce and social media scenarios, and cannot fully utilize data resources unique to parking lots, such as information about activity locations frequently visited by vehicles, vehicle brands and models, etc. These data resources are of great value in parking lot advertising recommendations because they can help the system understand the interests and needs of car owners more accurately. Among existing technical solutions, although recommendation algorithms have made considerable progress, especially in the field of graph convolutional networks (GCN), they still have some significant shortcomings in practical applications: First, data integration is insufficient and the data source is single. Traditional recommendation systems mainly rely on a single type of data, such as users' historical ratings or browsing history. This method fails to fully utilize information from multiple data sources, such as dynamic information (frequently visited locations) and static information (vehicle brands and models) of vehicles, resulting in low accuracy and relevance of recommendations; second, the response speed is slow and the computational complexity is high. Existing graph neural network models, such as standard GCN, often contain complex feature transformations and nonlinear activation operations, which increase Computational complexity and training difficulty make it difficult to achieve rapid response and instant recommendations in real-time application scenarios, especially in parking lots where vehicle dwell times are very short. Furthermore, due to computational resource and time constraints, traditional algorithms are prone to delays when processing large amounts of real-time data and cannot meet high real-time requirements. Thirdly, the model complexity is high. Many existing recommendation systems use complex deep learning models that require a large amount of computing resources and time for training and inference. This is a huge challenge for practical application scenarios with limited resources. Furthermore, the complex model structure makes it difficult to interpret and debug, increasing the maintenance cost and difficulty of the model, and affecting the promotion of practical applications. In summary, traditional algorithms are unable to effectively integrate and analyze this data, resulting in advertisements that are not relevant enough and cannot effectively attract the attention of car owners, thereby reducing the economic benefits of advertising. Summary of the Invention

[0004] To solve the above problems, the present invention proposes an intelligent real-time advertising push method based on a multi-level graph neural network, which integrates the dynamic and static information of the vehicle, and optimizes the information aggregation process through the self-attention mechanism to improve the accuracy and response speed of the recommendation. It is particularly suitable for real-time advertising push on parking gate display screens.

[0005] To achieve the above object, the present invention adopts the following technical solutions:

[0006] An intelligent real-time advertising push method based on a multi-level graph neural network specifically includes the following steps:

[0007] S1. Data collection: Collect dynamic and static data of the parking lot. Dynamic data includes vehicle entry and exit times and frequented locations, while static data includes vehicle brands and models.

[0008] S2. Data preprocessing: Use Apache Flink or Spark Streaming to preprocess the dynamic data collected in step S1.

[0009] S3. Multi-level graph neural network construction: Based on the CNN model, a global graph and a local graph are constructed. The global graph is constructed based on vehicle movement data at different locations. Nodes include vehicles, locations, and advertisements, and edges represent the relationships between vehicles and locations, vehicles and advertisements, and locations and advertisements. The local graph includes a frequently visited location-advertising graph and a vehicle brand-advertising graph, which refines the advertising relationships between users' location and brand preferences.

[0010] S4. Multi-level graph neural network training: Use historical data to pre-train the multi-level graph structure, adopt incremental update technology, and use online learning algorithms to update the embedding representations of users and items in real time to obtain a trained model;

[0011] S5. Real-time advertising push: Using the trained model, we calculate the embedding of each license plate and frequently visited location to generate a predicted score. For each user, we sort the ads based on the predicted score and select the highest-scoring ad as the push result.

[0012] As a further technical solution of the present invention, the dynamic data in step S1 is obtained through sensors and cameras in the parking lot, and the static data is obtained through a license plate recognition system.

[0013] As a further technical solution of the present invention, the global graph constructed in step S3 is represented as , by the node set and edge sets Composition; Frequently visited places - advertising pictures (partial Figure 1 ) is expressed as , by the node set and edge sets Composition; Vehicle brand - advertising picture (partial picture 2) is represented as , by the node set and edge sets composition.

[0014] As a further technical solution of the present invention, the specific process of step S4 is:

[0015] S41. Initialize the embedding representation of each node 、 、 , 、 、 Represents the global map, the frequently visited places-advertising map (local Figure 1 ) and nodes in the vehicle brand-advertising graph (partial figure 2) The node embedding representation of layer 0 (initialized);

[0016] S42. Perform graph convolution on the global graph to update the node embedding representations of ads, brands, and locations:

[0017] ,

[0018] ,

[0019] ,

[0020] in 、 and They represent advertising nodes, brand nodes, and location nodes respectively; 、 and Represents nodes respectively 、 and The set of neighbor nodes in the global graph; 、 and Represents nodes respectively 、 and In the global graph The node embedding representation of the layer; similarly, the graph convolution operation is performed on the two local graphs to obtain the updated node embedding representation:

[0021] For frequently visited places - advertising pictures (partial Figure 1 ):

[0022] ,

[0023] ,

[0024] For vehicle brand-advertising image (partial image 2):

[0025] ,

[0026] ,

[0027] in and Represents nodes respectively and In the frequently visited places - advertising picture (partial Figure 1 ) in the neighbor node set; and Represents nodes respectively and The set of neighbor nodes in the vehicle brand-advertisement graph (partial Figure 2); and Represents nodes respectively and In the frequently visited places - advertising picture (partial Figure 1 ) in Node embedding representation of the layer; and Represents nodes respectively and In the vehicle brand-advertising picture (partial picture 2) Node embedding representation of the layer;

[0028] S43, introduce the self-attention mechanism, and Figure 1 ) and nodes in the vehicle brand-advertising graph (partial figure 2) Each layer of embedding representation , , Calculate the attention scores separately:

[0029] ,

[0030] ,

[0031] ;

[0032] in is the weight vector in the self-attention mechanism, is the weight matrix in the self-attention mechanism;

[0033] S44, convert the attention score into attention weight through the Softmax function so that the sum of all weights is 1, the global map, the frequently visited places-advertising map (local Figure 1) and the vehicle brand-advertising image (partial image 2) The attention weights of the layer embedding representation are:

[0034] ,

[0035] ,

[0036] ;

[0037] S45. Weight the attention weights on the embedded representations of neighboring nodes to form nodes The new embedding representation of is:

[0038] ,

[0039] ,

[0040] ;

[0041] S46, introduce multi-graph fusion mechanism, for each node , the node In the frequently visited places - advertising picture (partial Figure 1 ) and the vehicle brand-advertisement image (partial figure 2) are fused through MLP to obtain the final embedding representation : ,in It is a learnable multi-layer perceptron;

[0042] S47. Using Bayesian Personalized Ranking (BPR) loss function Calculation error:

[0043] ,

[0044] in is the Sigmoid function, Indicates brand often go Location of advertising The prediction score of Indicates brand often go Location of advertising The prediction score of Controls the L2 regularization strength.

[0045] As a further technical solution of the present invention, the embedding of each license plate and frequently visited place calculated in step S5 is: , by calculating the inner product About Advertising Prediction score: .

[0046] Compared with the prior art, the present invention has the following advantages:

[0047] (1) We designed two graph structures: a global graph and a local graph. The global graph captures the movement patterns of vehicles at different locations and their exposure to advertisements; the local graph focuses on specific details, such as the direct relationship between frequently visited locations and advertisements, and the association between vehicle brands and advertisements. This multi-level graph structure enables the model to capture user behavior patterns more comprehensively and meticulously, improving the accuracy and relevance of recommendations. It overcomes the problem that traditional GCN models typically only utilize a single graph structure and fail to fully integrate and analyze information from multiple data sources, resulting in suboptimal recommendation results.

[0048] (2) By introducing the self-attention mechanism, the model can adaptively assign weights according to the importance of nodes and edges, thereby capturing key information more effectively. The specific implementation of the self-attention mechanism includes the calculation of attention scores and the aggregation of information, so that the model can aggregate the information of neighbor nodes according to the calculated attention weights in each layer, and form the final embedding representation of each node through the layer combination strategy, which provides the accuracy and real-time performance of advertising recommendations and overcomes the problem that the traditional GCN model fails to fully utilize the attention mechanism in the information aggregation process, resulting in weak information aggregation effect, which affects the accuracy and real-time performance of recommendations.

[0049] (3) The model architecture is simplified. By removing complex feature transformations and nonlinear activation operations, the computational complexity is reduced and the computational efficiency of the model is improved. It can run efficiently in practical application scenarios with limited resources and meet the needs of real-time recommendation. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 This is a flowchart of the intelligent real-time advertising push method based on a multi-level graph neural network provided by the present invention. DETAILED DESCRIPTION

[0051] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0052] It should be noted that the following detailed descriptions are exemplary and intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which the present invention belongs.

[0053] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0054] In the absence of conflict, the embodiments of the present invention and the features thereof may be combined with each other.

[0055] Example:

[0056] like Figure 1 As shown, this embodiment provides an intelligent real-time advertising push method based on a multi-level graph neural network, which specifically includes the following steps:

[0057] S1. Data collection: Collect dynamic and static data of the parking lot. Dynamic data includes the entry and exit time of vehicles and the places where they frequently go, which is obtained through sensors and cameras in the parking lot. Static data includes the brand and model of vehicles, which is obtained through the license plate recognition system.

[0058] S2. Data preprocessing: The dynamic data collected in step S1 is preprocessed using Apache Flink or Spark Streaming.

[0059] S3. Multi-level graph neural network construction: Based on the CNN model, the global graph and local graph are constructed respectively. It is constructed based on the movement data of vehicles at different locations. Nodes include vehicles, locations, and advertisements. Edges represent the relationship between vehicles and locations, vehicles and advertisements, and locations and advertisements. All nodes constitute a node set. , all edges form an edge set ; Partial map includes frequently visited places - advertising map (partial Figure 1 ) and vehicle brand-advertising image (partial image 2) , refine the advertising relationship between users in terms of location and brand preferences, Frequently visited places - advertising pictures (partial Figure 1 )’s node set and edge set; are the node sets and edge sets of the vehicle brand-advertisement graph (partial figure 2).

[0060] S4. Multi-level graph neural network training: Use historical data to pre-train the multi-level graph structure, adopt incremental update technology, and use online learning algorithms to update the embedding representations of users and items in real time to obtain a trained model. The specific process is as follows:

[0061] S41. Initialize the embedding representation of each node 、 、 , 、 、 Represents the global map, the frequently visited places-advertising map (local Figure 1 ) and nodes in the vehicle brand-advertising graph (partial figure 2) The node embedding representation of layer 0 (initialized);

[0062] S42. Perform graph convolution on the global graph to update the node embedding representations of ads, brands, and locations:

[0063] ,

[0064] ,

[0065] ,

[0066] in 、 and They represent advertising nodes, brand nodes, and location nodes respectively; 、 and Represents nodes respectively 、 and The set of neighbor nodes in the global graph; 、 and Represents nodes respectively 、 and In the global graph The node embedding representation of the layer; similarly, the graph convolution operation is performed on the two local graphs to obtain the updated node embedding representation:

[0067] For frequently visited places - advertising pictures (partial Figure 1 ):

[0068] ,

[0069] ,

[0070] For vehicle brand-advertising image (partial image 2):

[0071] ,

[0072] ,

[0073] in and Represents nodes respectively and In the frequently visited places - advertising picture (partial Figure 1 ) in the neighbor node set; and Represents nodes respectively and The set of neighbor nodes in the vehicle brand-advertisement graph (partial Figure 2); and Represents nodes respectively and In the frequently visited places - advertising picture (partial Figure 1 ) in Node embedding representation of the layer; and Represents nodes respectively and In the vehicle brand-advertising picture (partial picture 2) Node embedding representation of the layer;

[0074] S43. In the global graph and local graph, the relationship and importance of different nodes are different. The traditional graph convolution operation assigns the same weight to all neighbor nodes, ignoring the importance differences of different neighbor nodes. The attention mechanism can adaptively assign weights according to the characteristics of the nodes, assigning different weights to each neighbor node, thereby more accurately reflecting its importance in information aggregation. This can effectively process heterogeneous data and enable the model to better adapt to different types of graph structures and relationships. Therefore, this embodiment introduces the self-attention mechanism to the global graph, the frequently visited places-advertising graph (local graph), and the self-attention mechanism to the local graph. Figure 1 ) and nodes in the vehicle brand-advertising graph (partial figure 2) Each layer of embedding representation , , Calculate the attention scores separately:

[0075] ,

[0076] ,

[0077] ;

[0078] in is the weight vector in the self-attention mechanism, is the weight matrix in the self-attention mechanism;

[0079] S44, convert the attention score into attention weight through the Softmax function so that the sum of all weights is 1, the global map, the frequently visited places-advertising map (local Figure 1 ) and the vehicle brand-advertising image (partial image 2) The attention weights of the layer embedding representation are:

[0080] ,

[0081] ,

[0082] ;

[0083] S45. Weight the attention weights on the embedded representations of neighboring nodes to form nodes The new embedding representation of is:

[0084] ,

[0085] ,

[0086] ;

[0087] S46. In order to unify the node embedding representation of the global graph and the local graph, a multi-graph fusion mechanism is introduced. For each node , the node In the frequently visited places - advertising picture (partial Figure 1 ) and the vehicle brand-advertisement image (partial figure 2) are fused through MLP to obtain the final embedding representation : ,in It is a learnable multi-layer perceptron;

[0088] S47. Using Bayesian Personalized Ranking (BPR) loss function Computational error, Bayesian Personalized Ranking (BPR) loss function It is a pairwise loss function that encourages observed interaction terms to have higher prediction scores than unobserved interaction terms. Its formula is:

[0089] ,

[0090] in is the Sigmoid function, Indicates brand often go Location of advertising The prediction score of Indicates brand often go Location of advertising The prediction score of Controls the L2 regularization strength.

[0091] S5. Real-time advertising push: Using the trained model, calculate the embedding of each license plate and frequently visited location: , by calculating the inner product About Advertising Prediction score: ,For each user, the ads are sorted according to the predicted ratings, and the ads with the highest ratings are selected as push results.

[0092] This embodiment combines dynamic and static vehicle data for the first time, integrating currently available data and leveraging prior knowledge of dynamic vehicle information, such as frequent locations, and static information, such as vehicle brand. By combining a multi-level graph structure with a self-attention mechanism, the improved lightweight graph convolutional network approach enables more accurate ad recommendations, particularly suitable for real-time ad push on parking gate displays. This method not only improves recommendation accuracy but also maintains high computational efficiency, making it suitable for applications requiring high real-time performance and high precision. The multi-level graph structure more comprehensively captures the complex relationships between vehicles and ads. The global graph provides a holistic perspective, capturing the movement patterns and ad exposure of vehicle brands across different locations. The local graph focuses on specific details, such as the direct relationship between frequently visited locations and ads, and the association between vehicle brands and ads. This multi-level graph structure enables the model to more meticulously capture user behavior patterns, improving the accuracy and relevance of recommendations. Furthermore, a self-attention mechanism is designed to address the heterogeneity of the global and local graphs, further optimizing the information aggregation process. Through the self-attention mechanism, the model can adaptively assign weights based on the importance of nodes and edges, thereby more effectively capturing key information. The specific implementation of the self-attention mechanism includes the calculation of attention scores and information aggregation, so that the model can aggregate the information of neighboring nodes according to the calculated attention weights in each layer, and form the final embedding representation of each node through a layer combination strategy. This solves the problem of real-time advertising push on parking lot gate display screens. It can not only fully utilize the advantage of high traffic volume in parking lots, but also improve the relevance and economic benefits of advertising through precise recommendation algorithms, bringing higher returns to advertisers and improving the user experience of car owners. It has important practical application value.

[0093] Those skilled in the art will appreciate that the units, i.e., algorithm steps, of the various examples described in conjunction with this embodiment can be implemented using electronic hardware or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0094] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

[0095] Although the above describes the specific embodiments of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art on the basis of the technical solution of the present invention without any creative work are still within the scope of protection of the present invention.

Claims

1. An intelligent real-time advertising push method based on a multi-level graph neural network, characterized in that: The specific steps include: S1. Data collection: Collect dynamic and static data of the parking lot. Dynamic data includes vehicle entry and exit times and frequented locations, while static data includes vehicle brands and models. S2. Data preprocessing: Use Apache Flink or Spark Streaming to preprocess the dynamic data collected in step S1. S3. Multi-level graph neural network construction: Based on the CNN model, a global graph and a local graph are constructed. The global graph is constructed based on vehicle movement data at different locations. Nodes include vehicles, locations, and advertisements, and edges represent the relationships between vehicles and locations, vehicles and advertisements, and locations and advertisements. The local graph includes a frequently visited location-advertising graph and a vehicle brand-advertising graph, which refines the advertising relationships between users' location and brand preferences. S4. Multi-level graph neural network training: Use historical data to pre-train the multi-level graph structure, adopt incremental update technology, and use online learning algorithms to update the embedding representations of users and items in real time to obtain a trained model. The specific process is as follows: S41. Initialize the embedding representation of each node 、 、 , 、 、 Represent nodes in the global graph, the frequently visited places-advertising graph, and the vehicle brand-advertising graph respectively The node embedding representation of the 0th layer; S42. Perform graph convolution on the global graph to update the node embedding representations of ads, brands, and locations: ， ， ， in 、 and They represent advertising nodes, brand nodes, and location nodes respectively; 、 and Represents nodes respectively 、 and The set of neighbor nodes in the global graph; 、 and Represents nodes respectively 、 and In the global graph The node embedding representation of the layer; similarly, the graph convolution operation is performed on the two local graphs to obtain the updated node embedding representation: For frequently visited places - advertising pictures: ， ， For vehicle brand-advertising images: ， ， in and Represents nodes respectively and The set of neighbor nodes in the frequented places-advertisement graph; and Represents nodes respectively and The set of neighbor nodes in the vehicle brand-advertisement graph; and Represents nodes respectively and In the frequently visited places - the first Node embedding representation of the layer; and Represents nodes respectively and In the vehicle brand-advertising picture Node embedding representation of the layer; S43, introduce the self-attention mechanism to the nodes in the global graph, the frequently visited places-advertising graph and the vehicle brand-advertising graph Each layer of embedding representation , , Calculate the attention scores separately: ， ， ； in is the weight vector in the self-attention mechanism, is the weight matrix in the self-attention mechanism; S44. Convert the attention scores into attention weights through the Softmax function so that the sum of all weights is 1. The attention weights of the layer embedding representation are: ， ， ； S45. Weight the attention weights on the embedded representations of neighboring nodes to form nodes The new embedding representation of is: ， ， ； S46, introduce multi-graph fusion mechanism, for each node , the node The embedding representations of the frequently visited places-advertising images and vehicle brands-advertising images are fused through MLP to obtain the final embedding representation. : ,in It is a learnable multi-layer perceptron; S47. Using Bayesian Personalized Ranking Loss Function Calculation error: ， in is the Sigmoid function, Indicates brand often go Location of advertising The prediction score of Indicates brand often go Location of advertising The prediction score of Control the L2 regularization strength; S5. Real-time advertising push: Using the trained model, we calculate the embedding of each license plate and frequently visited location to generate a predicted score. For each user, we sort the ads based on the predicted score and select the highest-scoring ad as the push result.

2. The intelligent real-time advertising push method based on a multi-level graph neural network according to claim 1 is characterized in that: The dynamic data in step S1 is obtained through sensors and cameras in the parking lot, and the static data is obtained through a license plate recognition system.

3. The intelligent real-time advertising push method based on a multi-level graph neural network according to claim 2 is characterized in that: The global graph constructed in step S3 is represented as , by the node set and edge sets Composition; Frequently visited places - advertising map is represented as , by the node set and edge sets Composition; Vehicle brand - advertising image is represented as , by the node set and edge sets composition.

4. The intelligent real-time advertising push method based on a multi-level graph neural network according to claim 3 is characterized in that: The embedding of each license plate and frequently visited location calculated in step S5 is: , by calculating the inner product About Advertising Prediction score: .

Images