TWI911703B - Merchant management methods, devices, equipment, media and products - Google Patents

Abstract

This invention discloses a merchant management method, apparatus, equipment, medium, and product. The merchant management method includes acquiring user transaction data corresponding to multiple merchants; determining the relationships between the multiple merchants based on the user transaction data; constructing a merchant relationship network graph with merchants as nodes and relationships as connecting edges based on the relationships between the multiple merchants; extracting node characteristic information corresponding to each merchant from the merchant relationship network graph; and clustering the multiple merchants based on the node characteristic information to obtain a business district identification result. According to embodiments of this invention, the accuracy of business district identification results can be improved.

Description

Merchant management methods, devices, equipment, media, and products

This invention pertains to data prediction models, and particularly relates to a merchant management method, device, equipment, medium, and product.

In the digital age, merchant management models are gradually shifting towards refinement and intelligence. Among these, the business district is a crucial dimension for merchant management. For example, in merchant activation programs, since business districts concentrate a large proportion of offline traffic, accurately matching and linking merchants to business districts allows for precise targeting of promotional activities for merchants within the district, thereby activating their businesses.

Currently, the main method for identifying business districts in merchant management is as follows: Managers match merchants to their respective streets, shopping malls, or other commercial entities, and then group the merchants according to manually set business rules to form business districts.

Therefore, relying on the individual business experience of managers to delineate the business districts of merchants reduces the accuracy of business district identification results. Furthermore, simply focusing on the geographical location of merchants makes it difficult to accurately identify all merchants within a business district, easily overlooking potential merchants that should belong to that district, thus further reducing the accuracy of business district identification results.

This invention provides a merchant management method, device, equipment, medium, and product that can improve the accuracy of business district identification results.

Firstly, embodiments of the present invention provide a merchant management method, which includes:

Retrieve user transaction data corresponding to multiple merchants;

The relationships between the multiple merchants are determined based on the user transaction data.

Based on the relationships between the various merchants, a merchant relationship network graph is constructed, with the merchants as nodes and the relationships as connecting edges.

Extract the node representation information corresponding to each merchant from the merchant relationship network diagram;

Based on the node characteristic information, the multiple merchants are clustered to obtain the business district identification results.

Secondly, embodiments of the present invention provide a merchant management device, the device comprising:

The data acquisition module is used to acquire user transaction data corresponding to multiple merchants.

The relationship determination module is used to determine the relationships between the multiple merchants based on the user transaction data.

A network architecture modeling group is used to construct a merchant relationship network graph based on the relationships between the multiple merchants, with the merchants as nodes and the relationships as connecting edges;

A shape extraction module is used to extract node representation information corresponding to each merchant from the merchant relationship network graph;

The merchant clustering module is used to cluster the multiple merchants based on the node characteristic information to obtain the business district identification result.

Thirdly, embodiments of the present invention provide an electronic device comprising: a processor and memory storing computer program instructions;

When the processor executes the computer program instructions, it implements the steps of the merchant management method as described in any embodiment of the first aspect.

Fourthly, embodiments of the present invention provide a computer-readable storage medium storing computer program instructions, which, when executed by a processor, implement the steps of the merchant management method as described in any embodiment of the first aspect.

Fifthly, embodiments of the present invention provide a computer program product in which instructions, when executed by the processor of an electronic device, cause the electronic device to perform the steps of the merchant management method as described in any embodiment of the first aspect.

The merchant management method, device, equipment, medium, and product of this invention determine the relationships between merchants by utilizing the user transaction data corresponding to each merchant. Based on these relationships, a merchant relationship network graph is constructed with merchants as nodes and relationships as connecting edges. Node characteristic information corresponding to each merchant is extracted from this merchant relationship network graph, and merchants are then clustered according to this node characteristic information to obtain the business district identification result. Thus, since this invention automatically mines business district identification results from massive amounts of user transaction data, it avoids excessive reliance on human experience to determine business districts, thereby improving the accuracy of the business district identification results. Furthermore, since this invention constructs a merchant relationship network graph by utilizing the relationships between merchants as presented during transactions with users, and then performs representational clustering on the merchant nodes, it can segment merchants into business districts based on the dimension of user transaction preference relationships, rather than simply mining based on geographical location. Therefore, it can effectively reduce the possibility of merchants being excluded from business districts due to location, and uncover potential merchants that should be classified into business districts, thereby further improving the accuracy of business district identification results.

1,2,3,4,5: Merchants

500: Merchant Management Device

501: Data Acquisition Module

502: Relationship Confirmation Module

503: Network Architecture Modeling Group

504: Shape Extraction Module

505: Merchant Clustering Module

600: Electronic Equipment

601: Processor

602: Memory

603: Communication Interface

610: Busbar

A,a: business district

S110, S120, S130, S140, S150, S310, S320, S330, S340, S350, S360: Steps

To more clearly illustrate the technical solution of this invention's embodiments, a brief introduction to the drawings used in these embodiments will be provided below. For those skilled in the art, other drawings can be derived from these drawings without requiring further effort.

Figure 1 is a flowchart illustrating a merchant management method provided in an embodiment of the present invention;

Figure 2 is a flowchart illustrating an exemplary merchant management method provided by the present invention;

Figure 3 is a flowchart illustrating a merchant management method provided in another embodiment of the present invention;

Figure 4 is a schematic diagram of one type of terminal device display interface provided by the present invention;

Figure 5 is a schematic diagram of the structure of a merchant management device provided in one embodiment of the present invention;

Figure 6 is a schematic diagram of the structure of an electronic device provided in one embodiment of the present invention.

The features and exemplary embodiments of the present invention will be described in detail below. To make the purpose, technical solution, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to drawings and specific embodiments. It should be understood that the specific embodiments described herein are intended only to illustrate the present invention and not to limit it. For those skilled in the art, the present invention can be implemented without some of these specific details. The following description of the embodiments is only to provide a better understanding of the present invention by showing examples.

It should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variation thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes said element.

Currently, two main methods are used for business district identification in merchant management: One method involves managers matching merchants to their respective streets, shopping malls, or other commercial entities, and then grouping them according to manually set business rules to form business districts. The other method utilizes big data technology to track population movement patterns, obtaining the distribution of data for each area. For example, user travel data can be used to identify areas of high population density, and managers then manually refine these areas using pre-defined business rules to form business districts.

Regarding the first method mentioned above, it heavily relies on the individual business experience of managers. Different operators may have different perceptions of the same business district, resulting in different business district identification rules and reducing the accuracy of the identification results. Regarding the second method, in addition to relying on adjustments to the results by managers, different business districts vary depending on their location, user profiles, transaction volume, etc. Different features result in different geographic radii and merchant selection criteria, leading to a lack of uniform threshold parameters applicable to the identification process of all business districts. Therefore, simple big data mapping algorithms struggle to accurately identify all merchants within a business district. Even overlay density clustering algorithms for community mining may miss potential merchants that should be associated with a business district, thus reducing the accuracy of business district identification results.

To address the problems of existing technologies, this invention provides a merchant management method, apparatus, equipment, medium, and product. This merchant management method can be applied to scenarios involving merchant management and is executed by a merchant management server. The merchant management method provided by this invention will be introduced below.

Figure 1 is a flowchart illustrating a merchant management method provided in an embodiment of the present invention. As shown in Figure 1, the merchant management method may specifically include the following steps:

Step S110: Obtain user transaction data corresponding to multiple merchants;

Step S120: Determine the relationships between multiple merchants based on user transaction data;

Step S130: Based on the relationships between multiple merchants, construct a merchant relationship network graph with merchants as nodes and relationships as connecting edges;

Step S140: Extract the node representation information corresponding to each merchant from the merchant relationship network diagram;

Step S150: Cluster multiple merchants based on node characteristic information to obtain the business district identification results.

Thus, since this embodiment of the invention automatically mines business district identification results from massive amounts of user transaction data, it avoids excessive reliance on human experience to determine business districts, thereby improving the accuracy of the business district identification results. Furthermore, because this embodiment of the invention constructs a merchant relationship network graph using the relationships between merchants presented during transactions with users, and then performs representational clustering on merchant nodes, it can segment merchants into business districts based on the dimension of user transaction preference relationships, rather than simply mining based on geographical location. Therefore, it can effectively reduce the possibility of merchants being excluded from business districts due to location, and uncover potential merchants that should be classified into business districts, thereby further improving the accuracy of the business district identification results.

The specific implementation methods for each of the above steps are described below.

In some implementations, in step S110, the user transaction data can be transaction-related information recorded when the user and merchant conduct a transaction, such as the transaction amount, transaction time, merchant information, user information, and other transaction-related data.

For example, user transaction data recorded within the most recent preset time period can be obtained from the historical transaction data of multiple merchants, where the multiple merchants may include all merchants requiring business district identification.

In some implementations, in step S120, since the user transaction data contains information about both parties to the transaction—namely, merchant information and user information—the relationship between merchants can be determined based on this information. For example, a relationship can be established between any two merchants who have transacted with the same user.

Here, the relationship may include the degree of correlation. Based on this, in some embodiments, step S120 above may specifically include:

Based on a collaborative filtering algorithm, the correlation between multiple merchants is calculated using user transaction data.

In this embodiment, the collaborative filtering algorithm can specifically be the i2i itemCF algorithm. For example, user transaction data can be recorded in the i2i itemCF model, and the i2i itemCF model can be used to calculate the correlation between merchants, using this correlation as a measure of the degree of correlation between merchants. Specifically, the more shared transaction users two merchants have, the higher the correlation between the two merchants.

In some implementations, in step S130, the merchant relationship network diagram can be a network diagram used to represent the relationships between merchants. This merchant relationship network diagram may include multiple nodes, each node corresponding to a merchant, and the connecting edges between nodes are used to represent the relationships between merchants.

In cases where the relationship includes the degree of correlation, in some implementations, step S140 may specifically include:

Based on the relationships between multiple merchants, a merchant relationship network graph is constructed, with merchants as nodes, relationships as connecting edges, and the degree of relationship as the weight of the connecting edges.

For example, to more clearly express the degree of connection between merchants, the degree of connection can be represented by the weight of the edges connecting nodes to construct a merchant relationship network graph. That is, the merchant relationship network graph can include multiple nodes, each corresponding to a merchant. The edges connecting nodes represent the relationships between merchants, and the weights of the edges represent the degree of connection between them.

In some implementations, in step S140, a graph neural network model can be used to extract node representation information corresponding to each merchant from the merchant relationship network graph. This node representation information can, for example, be a node vector representation. This node representation information can characterize the relationship features between the corresponding node of a merchant and other nodes in the merchant relationship network graph.

For example, a merchant relationship network graph can be input into a graph neural network (GNN) model. The GNN model calculates the node vector representation corresponding to each merchant node, and outputs the node vector representation corresponding to each merchant. The GNN model can include one or more of the following: a deep walk model, a node2vec model, and a graph attention network (GAT) model.

To improve the accuracy of node representation information, in some possible embodiments, step S140 may specifically include:

The first-graph neural network model is used to perform characterization calculations on the nodes corresponding to each merchant in the merchant relationship network graph, obtaining the initial node characterization information for each merchant;

Using the second-graph neural network model, based on the initial node representation information, the representation of each node corresponding to each merchant in the merchant relationship network graph is calculated, thus obtaining the node representation information corresponding to each merchant.

Here, the first-order graph neural network model can include, for example, the DeepWalk model or the node2vec model; it is not limited to these two models. The first-order graph neural network model can also include, for example, the GAT model or other graph neural network models.

For example, the merchant relationship network graph is input into the DeepWalk model. Through random walk sampling of the merchant relationship network graph, the DeepWalk model calculates the graph embedding node vector representation, outputting the initial node vector representation corresponding to each merchant in the merchant relationship network graph. Then, the initial node vector representation corresponding to each merchant is input into the GAT model. The GAT model uses an attention mechanism to operate within the graph neural network for further node vector representation mining, outputting the final node vector representation corresponding to each merchant in the merchant relationship network graph.

This two-stage node representation calculation improves the accuracy of the final node representation information for each merchant.

In some possible embodiments, the step of using the second graph neural network model to calculate the representation of each node corresponding to each merchant in the merchant relationship network graph based on the initial node representation information, and obtaining the node representation information corresponding to each merchant, may specifically include:

The initial node representation information is used as initialization information and input into the second graph neural network model;

The neural network model in the second graph was trained using the relationships between merchants in the merchant relationship network graph and the industry types of the merchants as training objectives, resulting in node representation information corresponding to each merchant.

For example, the merchant relationship network graph and the initial node representation information corresponding to each merchant node in the graph can be input into the second graph neural network model for model training. The training task is to predict the industry type of each merchant and the relationships between merchants; that is, the training objective is set as the industry type of each merchant and the connection edges between each merchant node in the merchant relationship network graph. After the second graph neural network model is trained, the final node representation information corresponding to each merchant can be obtained.

In some implementation methods, in step S150, the business districts of multiple merchants can be divided by clustering the node representation information corresponding to each merchant.

For example, the density-based spatial clustering of applications with noise (DBSCAN) algorithm is used to cluster the node representation information corresponding to multiple merchants. The resulting clustering is the business district identification result. Each merchant cluster can be identified as a business district.

In addition, to further improve the accuracy of business district identification results, in some implementation methods, the merchant management methods provided by this invention may also include:

Obtain the location information of multiple merchants;

Based on location information, multiple businesses are clustered to identify at least one primary business district.

Accordingly, step S150 may specifically include:

Clustering multiple merchants based on node representation information identifies at least one secondary business district;

Merchants within the same area of at least one primary business district and at least one secondary business district are merged to obtain business district identification results.

Here, in addition to using merchant relationship network maps for business district identification, business district identification can also be performed based on the location of merchants. The results from these two methods are then combined to obtain the final business district identification result.

For example, based on the merchants' Global Positioning System (GPS) coordinates, density clustering can be performed on the GPS coordinates corresponding to multiple merchants using DBSCAN to obtain one or more first business districts. Combining the results of clustering merchants based on node representation information—that is, one or more second business districts obtained through clustering—business districts in the same location areas within the first and second business districts are merged to obtain the final business district identification result. The business district merging method includes, but is not limited to, taking the union of merchants in the two business districts, or taking the union of merchants in the two business districts that meet preset distance conditions, etc., and is not limited here.

For example, if at least one first business district includes business district A, at least one second business district includes business district a, business district A and merchant a are located in the same area, and business district A includes merchant 1, merchant 2 and merchant 3, and business district a includes merchant 1, merchant 2, merchant 4 and merchant 5, then the business district obtained by merging business districts A and business district a can include merchant 1, merchant 2, merchant 3, merchant 4 and merchant 5. Additionally, merging conditions can be set separately. For example, if the merging condition is set to the distance between a merchant and other merchants not exceeding a preset distance threshold, then if the distances between merchants 3 and 4 and other merchants do not exceed the preset distance threshold, and the distance between merchant 5 and other merchants exceeds the preset distance threshold, the merging of merchants in business districts A and a will result in a business district that includes merchants 1, 2, 3, and 4.

Based on the above-described implementation methods, specific examples are provided to illustrate in detail the business district identification method in this invention's embodiments.

As shown in Figure 2, the process can include two branches for business district identification and processing. In one branch, user transaction data is entered into the i2i itemCF model to calculate the correlation data between merchants. This correlation data is used as the edge weight parameters to construct a merchant relationship network graph. Then, the DeepWalk model is used to calculate the graph embedding node vector representation of this merchant relationship network graph, outputting the initial node vector representation corresponding to each merchant node. This initial node vector representation is then used as the initialization result and input into the GAT model for training. The training tasks are to predict the merchant industry type and the connection edge corresponding to the node. The node vector representation corresponding to each merchant node output by the GAT model is then clustered using DBSCAN to obtain the merchant vector clustering result.

In another processing branch, merchant GPS data is acquired, and density clustering is performed using DBSCAN based on the merchant GPS coordinates to obtain the merchant GPS clustering results.

By fusing the vector clustering results and the GPS clustering results for the merchant that fall within the same region, the final business district identification result can be obtained.

After identifying business districts, relevant information about the merchants within each district can be uploaded to the corresponding business district management device, preparing users for district-based merchant recommendations. When users visit a specific business district online or offline, they can receive personalized recommendations for the merchants within that district.

Based on this, and in order to improve the effectiveness of merchant management, as another implementation method of this invention, this invention also provides another implementation method for merchant management, as detailed in the following embodiments.

Referring to Figure 3, another implementation of the merchant management method provided by this invention includes the following steps: S310-S360, which are explained in detail below.

Step S310: Receive a merchant recommendation request sent by the user through a first terminal device, which is a terminal device located in the target business district.

Here, the primary terminal device can be any device placed within the target business district, such as a smart guidance robot.

For example, a user can perform facial recognition on a terminal device located anywhere within the target business district, and then send a merchant recommendation request to the backend merchant management server via that terminal device.

Step S320: Respond to merchant recommendation requests, obtain user-specific profile information, and merchant profile information for each merchant in the target business district.

Here, the profile feature information can be information used to characterize the temporal behavioral characteristics of a user, where temporal behavior may include, for example, transaction behavior over a period of time.

For example, after a user scans a QR code displayed on a smart terminal using facial recognition or their mobile phone, the user's corresponding user ID can be identified based on the facial image or user information obtained from the mobile phone. Then, based on the user ID, corresponding image feature information can be retrieved. Simultaneously, merchant feature information for each merchant listed in the current target business district can be retrieved.

Step S330: Match the image feature information with the corresponding merchant feature information of each merchant in the target business district to determine the target merchants that match the user.

In this embodiment, the user's profile information and the corresponding merchant information in the target business district can be input into a personalized matching model. This model then scores the match between the user and each merchant in the target business district, assigning a recommendation value to each merchant relative to the user. A higher recommendation value indicates a higher personalized consumption preference from the user towards that merchant.

Furthermore, the aforementioned personalized matching models may include, for example, Deep Interest Networks (DIN), which utilize deep neural networks and user temporal behavioral characteristics to model and obtain the personalized matching degree between users and merchants.

After determining the referral value of each merchant within the target business district relative to the user, the merchants can be ranked from highest to lowest based on this referral value. This determines the merchants in the pre-set ranking as target merchants matching the user. Alternatively, all merchants with referral values higher than the set threshold can be designated as target merchants matching the user; this is not limited to this approach. The number of target merchants can be one or more.

To target paid advertising merchants, step S330 above may specifically include, in some possible implementation methods, the following:

The image feature information is matched with the corresponding merchant feature information of each merchant in the target business district to determine the initial recommendation value of each merchant in the target business district relative to the user;

Obtain payment information for each merchant in the target business district;

The initial referral value is adjusted based on payment information to obtain the corresponding referral value for each merchant;

Based on the referral values of each merchant, target merchants that match the user are identified.

Here, the recommended value output by the personalized matching model can be used as the initial recommended value. This initial recommended value can then be adjusted accordingly, taking into account the payment status of each merchant in the target business district. For example, the initial recommended value can be increased for merchants who pay for advertising.

In this way, a final referral value can be obtained for each merchant. Based on this referral value, target merchants matching the user can be determined. The specific method for determining target merchants is the same as described above and will not be repeated here.

Step S340: Obtain the location information corresponding to the target merchant to obtain the target location information.

In this embodiment, after identifying the target merchant, the merchant's location information, such as its GPS coordinates, can be retrieved.

To facilitate users in selecting their preferred merchants, in some possible implementation methods where there are multiple target merchants, step S340 may specifically include:

The system sends merchant information corresponding to multiple target merchants to a first terminal device, enabling the first terminal device to display the merchant information, and receives user input selecting at least one target merchant, the merchant information including at least one target merchant's information;

Receive at least one target merchant information sent by the first terminal device;

Obtain the location information of at least one target merchant, and thus obtain the target location information.

Here, the automatically identified merchant information for all target merchants can be sent to the first-end device for display. This merchant information may include the merchant name, promotional information, etc.

For example, as shown in Figure 4, after receiving merchant information from all target merchants automatically recommended by the system, the first terminal device can display this merchant information in the "You May Also Like" area of the display interface. The user can select N merchant information as target merchant information, which is then sent by the first terminal device to the backend server. The backend server obtains the location information corresponding to the user-selected merchants so that it can subsequently generate a recommended shopping route that includes the user's final target merchants based on this location information.

Step S350: Generate a recommended shopping route passing through target merchants based on target location information.

Here, the recommended shopping route can be a route planned from the location of the first terminal device that passes through the target merchants. If there are multiple target merchants, the recommended shopping route can pass through the locations of all of them.

Step S360: Send a recommended shopping route to the first terminal device so that the first terminal device displays the recommended shopping route.

For example, as shown in Figure 4, the first terminal device can display recommended shopping routes received from the backend server in the "Today's Shopping Route" area of the display interface, allowing users to shop according to the displayed recommended routes.

This way, after visiting their target shopping area, users can receive personalized shopping route guidance through smart terminals within the area. This enables personalized recommendations for merchants within the area, providing users with a highly targeted merchant recommendation experience and a customized offline service experience.

In addition, to incentivize users to shop according to recommended shopping routes, in some possible implementations, after step S360 described above, the merchant management method provided by this embodiment may also include:

Receive user check-in information from a second terminal device, where the user check-in information is generated by the second terminal device in response to the user's check-in input. The second terminal device is located at any point along the recommended shopping route.

Based on user check-in information, add corresponding virtual resources to the user's corresponding account.

Here, the second terminal device can be located anywhere along the recommended shopping route. For example, it can be installed at the target merchant or in other locations along the recommended shopping route. That is, users can check in via the smart terminal set up along the recommended shopping route, or by shopping at the recommended target merchant. Specific check-in methods include, but are not limited to, facial recognition and/or scanning QR codes. Additionally, virtual resources can be issued to the user's account as check-in rewards. These virtual resources could be, for example, points.

For example, if a user completes a cumulative check-in task according to the system's recommended shopping route, they can choose to receive personalized rewards. In this embodiment, facial recognition or QR code scanning can be used for system login and check-in.

In addition, to update the personalized matching model dynamically and adjust the personalized merchant recommendation results in a timely manner, in some possible embodiments, after the step of receiving user check-in information sent by the second terminal device, the merchant management method provided by this invention may also include:

The system adjusts and updates portrait and merchant profile information based on user check-in data.

Here, because users check in and provide feedback on the shopping routes recommended by the system, user behavior and the transaction behavior of target merchants may change as a result. Therefore, merchant and user data can be updated in a timely manner to adjust and update user profile information and target merchant information accordingly.

It should be noted that the application scenarios described above in the embodiments of the present invention are for the purpose of more clearly illustrating the technical solutions of the present invention, and do not constitute a limitation on the technical solutions provided by the embodiments of the present invention. Those skilled in the art will understand that with the emergence of new application scenarios, the technical solutions provided by the embodiments of the present invention are equally applicable to similar technical problems.

Based on the same inventive concept, this invention also provides a merchant management device. A detailed description is provided with reference to Figure 5.

Figure 5 is a schematic diagram of the structure of a merchant management device provided in one embodiment of the present invention.

As shown in Figure 5, the merchant management device 500 may include:

Data acquisition module 501 is used to acquire user transaction data corresponding to multiple merchants;

Relationship determination module 502 is used to determine the relationship between the multiple merchants based on the user transaction data;

Network architecture module 503 is used to construct a merchant relationship network graph based on the relationships between the multiple merchants, with the merchants as nodes and the relationships as connecting edges;

Shape extraction module 504 is used to extract node representation information corresponding to each merchant from the merchant relationship network graph;

The merchant clustering module 505 is used to cluster the multiple merchants based on the node characteristic information to obtain the business district identification result.

The following is a detailed description of the aforementioned merchant management device 500, as shown below:

In some embodiments, the shape extraction module 504 includes:

The first operator module is used to perform characterization calculations on the nodes corresponding to each merchant in the merchant relationship network graph using the first graph neural network model, to obtain initial node characterization information corresponding to each merchant;

The second operator module is used to perform representation calculations on the nodes corresponding to each merchant in the merchant relationship network graph based on the initial node representation information using the second graph neural network model, thereby obtaining node representation information corresponding to each merchant.

In some embodiments, the second operator module includes:

The information input unit is used to input the initial node representation information as initialization information into the second graph neural network model;

The model training unit is used to train the second graph neural network model using the relationships between merchants in the merchant relationship network graph and the industry types of the merchants as training targets, to obtain node representation information corresponding to each merchant.

In some embodiments, the aforementioned merchant management device 500 also includes:

A location acquisition module is used to acquire the location information corresponding to each of the multiple merchants.

A location clustering module is used to cluster the multiple merchants based on the location information, identifying at least one primary business district;

The merchant clustering module 505 includes:

The feature clustering submodule is used to cluster the multiple merchants based on the node feature information, identifying at least one secondary business district;

The business district integration submodule is used to integrate merchants within the same area of the at least one first business district and the at least one second business district, thereby obtaining business district identification results.

In some embodiments, the relationship determination module 502 is specifically used for:

Based on a collaborative filtering algorithm, the correlation degree between the multiple merchants is calculated according to the user transaction data, and the correlation relationship includes the correlation degree;

The network architecture modeling group 503 is specifically used for:

Based on the relationships between the multiple merchants, a merchant relationship network graph is constructed, with the merchants as nodes, the relationships as connecting edges, and the relationship degree as the weight of the connecting edges.

In some embodiments, the aforementioned merchant management device 500 also includes:

A request receiving module is used to receive merchant recommendation requests sent by users through a first terminal device, which is a terminal device located in the target business district.

The request-response module is used to respond to the merchant recommendation request, obtain the user's corresponding image feature information, and the merchant feature information of each merchant in the target business district;

The merchant matching module is used to match the image feature information with the corresponding merchant feature information of each merchant in the target business district to determine the target merchants that match the user;

The target acquisition module is used to acquire the location information corresponding to the target merchant, thereby obtaining the target location information;

A route generation module is used to generate recommended shopping routes passing through the target merchants based on the target location information.

A route sending module is used to send the recommended shopping route to the first terminal device, so that the first terminal device displays the recommended shopping route.

In some embodiments, the target acquisition module includes:

The information transmission submodule is used to send merchant information corresponding to each of the multiple target merchants to the first terminal device, so that the first terminal device can display the merchant information, and to receive user input selecting at least one target merchant, wherein the merchant information includes the at least one target merchant information;

An information receiving submodule is used to receive the at least one target merchant information sent by the first terminal device;

The first acquisition submodule is used to acquire the location information of the target merchant corresponding to the information of the at least one target merchant, thereby obtaining the target location information.

In some embodiments, the merchant matching module includes:

A feature matching submodule is used to match the image feature information with the corresponding merchant feature information of each merchant in the target business district to determine the initial recommendation value of each merchant in the target business district relative to the user;

The second acquisition sub-module is used to acquire payment information for each merchant within the target business district;

The referral value adjustment submodule is used to adjust the initial referral value based on the payment information to obtain the corresponding referral value for each merchant;

The target determination submodule is used to identify target merchants that match the user based on the referral values corresponding to each merchant.

In some embodiments, the aforementioned merchant management device 500 also includes:

The user check-in module is used to receive user check-in information from a second terminal device after sending the recommended shopping route to the first terminal device. The user check-in information is generated by the second terminal device in response to the user's check-in input. The second terminal device is located at any position along the recommended shopping route.

The resource addition module is used to add corresponding virtual resources to the user's account based on the user's attendance information.

In some embodiments, the aforementioned merchant management device 500 also includes:

A feature adjustment module is used to adjust and update the image feature information and the merchant feature information based on the user check-in information received from the second terminal device.

Therefore, by utilizing the user transaction data corresponding to each merchant, the relationships between merchants are determined. Based on these relationships, a merchant relationship network graph is constructed, with merchants as nodes and relationships as connecting edges. The node characteristic information corresponding to each merchant is extracted from this graph, and then the merchants are clustered according to this node characteristic information to obtain the business district identification result. Thus, since this embodiment of the invention automatically mines business district identification results from massive amounts of user transaction data, it avoids excessive reliance on human experience to determine business districts, thereby improving the accuracy of the business district identification results. Furthermore, since this invention constructs a merchant relationship network graph by utilizing the relationships between merchants as presented during transactions with users, and then performs representational clustering on the merchant nodes, it can segment merchants into business districts based on the dimension of user transaction preference relationships, rather than simply mining based on geographical location. Therefore, it can effectively reduce the possibility of merchants being excluded from business districts due to location, and uncover potential merchants that should be classified into business districts, thereby further improving the accuracy of business district identification results.

Figure 6 is a schematic diagram of the structure of an electronic device provided in one embodiment of the present invention.

The electronic device 600 may include a processor 601 and memory 602 storing computer program instructions.

Specifically, the processor 601 described above may include a Central Processing Unit (CPU), or an Application Specific Integrated Circuit (ASIC), or one or more integrated circuits that can be configured to implement embodiments of the present invention.

Memory 602 may include a large storage area for data or instructions. For example, and not limitingly, memory 602 may include a hard disk drive (HDD), disk drive, flash memory, optical disc, magneto-optical disc, magnetic tape, or Universal Serial Bus (USB) drive, or a combination of two or more of these. Where appropriate, memory 602 may include removable or non-removable (or fixed) media. Where appropriate, memory 602 may be internal or external to an integrated gate disaster recovery device. In a particular embodiment, memory 602 is non-volatile solid-state memory.

In certain embodiments, memory may include read-only memory (ROM), random access memory (RAM), magnetic storage media devices, optical storage media devices, flash memory devices, and electrical, optical, or other physical/tangible memory storage devices. Therefore, typically, memory includes one or more tangible (non-transient) computer-readable storage media (e.g., memory devices) encoded with software including computer-executable instructions, and is operable to perform the operations described with reference to a method according to one aspect of the invention when the software is executed (e.g., by one or more processors).

Processor 601 implements any of the merchant management methods described in the above embodiments by reading and executing computer program instructions stored in memory 602.

In some examples, the electronic device 600 may also include a communication interface 603 and a bus 610. As shown in Figure 6, the processor 601, memory 602, and communication interface 603 are connected via bus 610 and communicate with each other.

Communication interface 603 is mainly used to enable communication between modules, devices, units, and/or equipment in the embodiments of the present invention.

Bus 610 includes hardware, software, or both, that couples together the components of an online data flow metering device. For example, and not as a limitation, bus 610 may include an Accelerated Graphics Port (AGP) or other graphical buses, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a Hyper Transport (HT) interconnect, an Industry Standard Architecture (ISA) bus, a Wi-Fi interconnect, a Low Pin Count (LPC) bus, a memory bus, a Micro Channel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, and a Fast Peripheral Component Interconnect (PCI) bus. The bus may be an Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local Bus (VLB) bus, or other suitable buses, or a combination of two or more of these. Where suitable, bus 610 may include one or more buses. Although specific buses are described and shown in embodiments of the invention, the invention contemplates any suitable bus or interconnection.

For example, electronic device 600 can be a mobile phone, tablet computer, laptop computer, handheld computer, in-vehicle electronic device, ultra-mobile personal computer (UMPC), netbook, or personal digital assistant (PDA), etc.

The electronic device 600 can execute the merchant management method in the embodiments of the present invention, thereby realizing the merchant management method and device described in the above embodiments.

In addition, in conjunction with the merchant management methods in the above embodiments, the present invention can provide a computer-readable storage medium for implementation. The computer-readable storage medium stores computer program instructions; when these computer program instructions are executed by the processor, they implement any of the merchant management methods in the above embodiments. Examples of computer-readable storage media include non-transient computer-readable storage media, such as portable storage devices, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), compact disc read-only memory (CD-ROM), optical memory devices, magnetic memory devices, etc.

It should be clarified that this invention is not limited to the specific configuration and processing described above and shown in the figures. For the sake of simplicity, a detailed description of known methods is omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of this invention is not limited to the specific steps described and shown, and those skilled in the art can make various changes, modifications, and additions, or change the order of the steps, after understanding the spirit of this invention.

The functional blocks shown in the above-described block diagram can be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, they can be, for example, electronic circuits, application-specific integrated circuits (ASICs), suitable firmware, plug-ins, function cards, etc. When implemented in software, the elements of this invention are programs or code segments used to perform the required tasks. The programs or code segments can be stored in a machine-readable medium or transmitted via data signals carried in a carrier medium or communication link. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable read-only memory (EROM), floppy disks, CD-ROMs, optical discs, hard drives, optical fiber media, radio frequency (RF) chains, and so on. Code snippets can be downloaded via computer networks such as the Internet and intranets.

It should also be noted that the exemplary embodiments mentioned in this invention describe methods or systems based on a series of steps or devices. However, this invention is not limited to the order of the above steps; that is, the steps can be performed in the order mentioned in the embodiments, or in a different order, or several steps can be performed simultaneously.

Aspects of the present invention are described above with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and/or block diagrams, and combinations of blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device to produce a machine such that execution of the instructions via the processor of the computer or other programmable data processing device enables implementation of the functions/actions specified in one or more blocks of the flowcharts and/or block diagrams. Such a processor may be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field-programmable logic circuit. It is also understood that each box in the block diagram and/or flowchart, as well as combinations of boxes in the block diagram and/or flowchart, can be implemented by dedicated hardware that performs the specified function or action, or by a combination of dedicated hardware and computer instructions.

The above description is merely a specific embodiment of the present invention. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, modules, and units described above can be referred to the corresponding processes in the aforementioned method embodiments, and will not be repeated here. It should be understood that the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and all such modifications or substitutions should be covered within the scope of protection of the present invention.

S110, S120, S130, S140, S150: Steps

Claims (11)

A merchant management method, characterized by comprising: a merchant management server acquiring user transaction data corresponding to multiple merchants respectively; the merchant management server determining the relationships between the multiple merchants based on the user transaction data; the merchant management server constructing a merchant relationship network graph with the merchants as nodes and the relationships as connecting edges based on the relationships between the multiple merchants; the merchant management server using a first graph neural network model to perform a representation calculation on the node corresponding to each merchant in the merchant relationship network graph to obtain initial node representation information corresponding to each merchant; the merchant management server inputting the initial node representation information as initialization information into a second graph neural network model; the merchant management... The merchant management server trains the second graph neural network model using the relationships between merchants in the merchant relationship network graph and the industry types of the merchants as training targets, obtaining node representation information corresponding to each merchant. The merchant management server clusters the multiple merchants based on the node representation information, identifying at least one second business district. The merchant management server acquires the location information corresponding to each of the multiple merchants. The merchant management server clusters the multiple merchants based on the location information, identifying at least one first business district. The merchant management server merges merchants located in the same area within the at least one first business district and the at least one second business district, obtaining a business district identification result. As described in claim 1, wherein the merchant management server determines the relationships between the multiple merchants based on the user transaction data, including: the merchant management server calculating the correlation degree between the multiple merchants based on the user transaction data using a collaborative filtering algorithm, wherein the correlation degree is included in the correlation relationship; and the merchant management server constructing a merchant relationship network graph with the merchants as nodes and the relationships as connecting edges based on the relationships between the multiple merchants, including: the merchant management server constructing a merchant relationship network graph with the merchants as nodes, the relationships as connecting edges, and the correlation degree as the weight of the connecting edges based on the relationships between the multiple merchants. The method described in claim 1 further includes: the merchant management server receiving a merchant recommendation request sent by a user through a first terminal device, wherein the first terminal device is a terminal device located in the target business district; the merchant management server responding to the merchant recommendation request by obtaining image feature information corresponding to the user and merchant feature information corresponding to each merchant in the target business district; the merchant management server comparing the image feature information with the target business district... The system matches the merchant characteristics of each merchant within the circle to determine the target merchant that matches the user. The merchant management server obtains the location information corresponding to the target merchant to obtain target location information. Based on the target location information, the merchant management server generates a recommended shopping route passing through the target merchant. The merchant management server sends the recommended shopping route to the first terminal device, so that the first terminal device displays the recommended shopping route. The method described in claim 3, wherein, when there are multiple target merchants, the merchant management server obtains the location information corresponding to the target merchants to obtain target location information, including: the merchant management server sending merchant information corresponding to the multiple target merchants to the first terminal device, so that the first terminal device displays the merchant information; and receiving user selection input for at least one target merchant, the merchant information including the at least one target merchant information; the merchant management server receiving the at least one target merchant information sent by the first terminal device; and the merchant management server obtaining the location information of the target merchant corresponding to the at least one target merchant information to obtain the target location information. As described in claim 3, wherein the merchant management server matches the image feature information with the corresponding merchant feature information of each merchant in the target business district to determine target merchants matching the user, including: the merchant management server matching the image feature information with the corresponding merchant feature information of each merchant in the target business district to determine the initial recommendation value of each merchant in the target business district relative to the user; the merchant management server obtaining the corresponding payment information of each merchant in the target business district; the merchant management server adjusting the initial recommendation value according to the payment information to obtain the recommendation value corresponding to each merchant; and the merchant management server determining the target merchants matching the user based on the corresponding recommendation values of each merchant. The method described in claim 3, wherein, after the merchant management server sends the recommended shopping route to the first terminal device, the method further includes: the merchant management server receiving user check-in information sent by a second terminal device, wherein the user check-in information is information generated by the second terminal device in response to the user's check-in input, and the second terminal device is located at any position on the recommended shopping route; the merchant management server adding corresponding virtual resources to the user's corresponding user account based on the user check-in information. The method of claim 6, wherein, after the merchant management server receives user attendance information sent by the second terminal device, the method further includes: the merchant management server adjusting and updating the image feature information and the merchant feature information based on the user attendance information. A merchant management device, characterized in that it includes: a data acquisition module for acquiring user transaction data corresponding to multiple merchants respectively; a relationship determination module for determining the relationships between the multiple merchants based on the user transaction data; and a network construction module for constructing a merchant relationship network graph with the merchants as nodes and the relationships as connecting edges based on the relationships between the multiple merchants; a first calculation... The operator module is used by the merchant management server to perform characterization calculations on the nodes corresponding to each merchant in the merchant relationship network graph using the first graph neural network model, to obtain initial node characterization information corresponding to each merchant; the information input unit is used by the merchant management server to input the initial node characterization information as initialization information into the second graph neural network model; the model training unit is used by the merchant management server... The device trains the second graph neural network model using the relationships between merchants in the merchant relationship network graph and the industry types to which the merchants belong as training targets, obtaining node representation information corresponding to each merchant; a representation clustering submodule is used by the merchant management server to cluster the multiple merchants according to the node representation information, identifying at least one second business district; a location acquisition module is used by the merchant management... The merchant management server acquires the location information corresponding to the multiple merchants respectively; a location clustering module is used by the merchant management server to cluster the multiple merchants according to the location information, identifying at least one first business district; a business district fusion submodule is used by the merchant management server to merge merchants located in the same area within the at least one first business district and the at least one second business district, obtaining a business district identification result. An electronic device is characterized in that it includes: a processor and a memory storing computer program instructions; the processor, when executing the computer program instructions, implements the steps of the merchant management method as described in any one of claims 1 to 7. A computer-readable storage medium, characterized in that computer program instructions are stored on the computer-readable storage medium, which, when executed by a processor, implement the steps of the merchant management method as described in any one of claims 1 to 7. A computer program product characterized in that, when the instructions in the computer program product are executed by the processor of an electronic device, the electronic device performs the steps of the merchant management method as described in any one of claims 1 to 7.