A method, device and system for precision marketing based on a user relationship graph

Abstract

The application relates to the technical field of Internet, and provides a precise marketing method, device and system based on a user relationship graph, which comprises the following steps: adding a unique identifier corresponding to a creating user in a marketing activity sharing link created by a first targeted user by establishing the unique identifier of the first targeted user and a targeted node of each first targeted user in a relationship graph; when a potential user accesses the marketing activity through the sharing link, establishing a unique identifier of the potential user, establishing a corresponding first node in the relationship graph, and establishing a connection relationship between the first node and the targeted node in the relationship graph according to the unique identifier in the link; calculating the importance of each node according to the connection relationship between the nodes in the relationship graph, determining a second targeted user according to the node importance, and accurately selecting a user who can play a greater role in marketing activity promotion to distribute marketing activity information, so that the effect of full-quantity user promotion is realized with less marketing cost, and the marketing efficiency is improved.

Description

Technical Field

[0001] This invention relates to the field of Internet technology, and in particular to a precision marketing method, apparatus and system based on user relationship graphs. Background Technology

[0002] In an era of rapid information development, users' daily lives are filled with all kinds of marketing information. Most users will ignore marketing information directly delivered by companies as junk information.

[0003] However, most companies, due to a lack of attention to or ability to maintain and deeply mine user data security, can only choose to directly deliver marketing information to users or adopt a full-coverage delivery model to all users. This is not only unfriendly to users, but also results in high marketing costs and low marketing efficiency for companies. Summary of the Invention

[0004] This invention provides a precision marketing method based on user relationship graphs, which solves the problem of high marketing costs and low efficiency caused by the existing technology of directly delivering marketing information to all users.

[0005] The first aspect of this invention provides a precision marketing method based on user relationship graphs, comprising:

[0006] Establish a unique identifier for the first target user, and create a target node corresponding to the unique identifier of each first target user in the relationship graph;

[0007] Distribute marketing campaign information to the first-target users and add a unique identifier corresponding to the first-target users to the marketing campaign sharing links created by the first-target users;

[0008] When a potential user accesses a marketing campaign through a shared link, a unique identifier for the potential user is established, and a corresponding first node is created in the relationship graph. Based on the unique identifier in the link, a connection relationship is established between the first node and the target node in the relationship graph.

[0009] The importance of each node is calculated based on the connections between nodes in the relationship graph, and the second target user is determined based on the importance of the nodes.

[0010] Optionally, the step of calculating the importance of each node based on the connection relationships between nodes in the relationship graph specifically involves:

[0011] The importance of a node is calculated using a node importance model based on the number of other nodes it connects to and the importance of those other nodes. The node importance model is as follows:

[0012]

[0013] In the formula, x i Let C be the importance of node i, and let x be the scaling constant. ij To determine the importance of node j connected to node i, a ij Let i be the adjacency relationship between i and j, which is connected to i.

[0014] Optionally, the weight value of node j is set according to the number of consumptions, consumption intervals, and consumption levels of the user corresponding to node j.

[0015] Optionally, determining the second target user based on node importance includes:

[0016] Nodes are sorted from highest to lowest importance. A predetermined number of nodes are selected based on their importance. Users corresponding to these selected nodes are then designated as the second target users, and marketing campaign information is distributed to them. The predetermined number is determined based on the company's marketing costs; the higher the marketing costs, the higher the predetermined number.

[0017] Optionally, determining the second target user based on node importance further includes:

[0018] Obtain basic data such as the province, city, user type, and consumption level of the corresponding users in the nodes, and further filter the nodes according to the needs of the marketing campaign.

[0019] Optionally, after establishing the connection between the first node and the target node in the relationship graph, the method further includes:

[0020] Track user behavior data by embedding tracking points on the marketing interface and incentivizing users based on this data. This behavior data includes the time spent browsing product detail pages, the number of times the user browses the detail pages, the types of products viewed, the number of times the user clicks on product details, and the ordering method.

[0021] Optionally, the step of incentivizing the user based on the behavioral data specifically includes:

[0022] A marketing strategy model is established. After inputting user behavior data and marketing cost budget into the model, a marketing incentive strategy for each user is obtained, and the incentive content is displayed to users through promotional SMS, APP push notifications, and other means.

[0023] A second aspect of this application provides a precision marketing device based on a user relationship graph, comprising:

[0024] The node creation module is used to create a unique identifier for the first target user and to create a target node corresponding to the unique identifier of each first target user in the relationship graph;

[0025] The marketing delivery module is used to distribute marketing campaign information to the first-target users and add a unique identifier corresponding to the first-target user to the marketing campaign sharing link created by the first-target user.

[0026] The account login module is used to establish a unique identifier for potential users when they access marketing campaigns through shared links, create a corresponding first node in the relationship graph, and establish a connection relationship between the first node and the target node in the relationship graph based on the unique identifier in the link.

[0027] The precision marketing module is used to calculate the importance of each node based on the connection relationships between nodes in the relationship graph, and to determine the second target user based on the importance of the node.

[0028] Optionally, in the precision marketing module, determining the second target user based on node importance includes:

[0029] Nodes are sorted from highest to lowest importance. A predetermined number of nodes are selected based on their importance. Users corresponding to these selected nodes are then designated as the second target users, and marketing campaign information is distributed to them. The predetermined number is determined based on the company's marketing costs; the higher the marketing costs, the higher the predetermined number.

[0030] A third aspect of this application provides a precision marketing system based on a user relationship graph, the system comprising a processor and a memory:

[0031] The memory is used to store program code and transmit the program code to the processor;

[0032] The processor is used to execute the precision marketing method based on user relationship graphs according to any one of the first aspects of the present invention, according to the instructions in the program code.

[0033] As can be seen from the above technical solutions, the present invention has the following advantages: It establishes a unique identifier for the first target user and creates a corresponding target node in the relationship graph; it distributes marketing activity information to the first target user and adds a unique identifier corresponding to the user who created the marketing activity sharing link; when a potential user accesses the marketing activity through the sharing link, it establishes a unique identifier for the potential user, creates a corresponding first node in the relationship graph, and establishes a connection between the first node and the target node in the relationship graph based on the unique identifier in the link; it spreads the marketing activity to more users through sharing among users, avoiding direct contact between the enterprise and users, reducing user resistance and aversion, and establishing a desensitized relationship network with user information during the sharing process; it calculates the importance of each node based on the connection relationship between nodes in the relationship graph, determines the second target user based on the node importance, accurately selects users who can play a greater role in the promotion of the marketing activity for the distribution of marketing activity information, achieves the effect of full-user promotion with less marketing cost, and improves marketing efficiency. Attached Figure Description

[0034] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0035] Figure 1 This is the first flowchart of a precision marketing method based on user relationship graphs.

[0036] Figure 2 This is the second flowchart for a precision marketing approach based on user relationship graphs;

[0037] Figure 3 This is the third flowchart for a precision marketing approach based on user relationship graphs;

[0038] Figure 4 This is a diagram of a precision marketing device based on user relationship graphs. Detailed Implementation

[0039] To make the objectives, features, and advantages of this invention more apparent and understandable, the technical solutions of the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the embodiments described below are only some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0040] This invention provides a precision marketing method based on user relationship graphs, which solves the problem of high marketing costs and low efficiency caused by the existing technology of directly delivering marketing information to all users.

[0041] Please see Figure 1 , Figure 1 This is the first flowchart of a precision marketing method based on user relationship graphs provided in an embodiment of the present invention.

[0042] S100, establish a unique identifier for the first target user, and establish a target node corresponding to the unique identifier of each first target user in the relationship graph;

[0043] It should be noted that in the initial stage of building the relationship graph, you can choose users maintained by the enterprise itself or import customer information from historical orders and use these users as the first target users;

[0044] Specifically, when establishing a unique identifier, the province and city in the mobile phone number of the first target user are associated with the unique identifier, and this basic data is stored in the user node created in the relationship graph. Subsequently, only the unique identifier and the above basic data are processed, thereby ensuring user privacy and avoiding leakage. In actual implementation, the relationship graph is established based on the Neo4j graph database, and nodes are created in the graph database to store data.

[0045] S200 delivers marketing campaign information to the first-target user and adds a unique identifier corresponding to the first-target user to the marketing campaign sharing link created by the first-target user.

[0046] It should be noted that, in order to establish relationships among users, information about non-profit marketing campaigns can be distributed to the first-target users. These campaigns are rich in content, interesting, and have a wide audience and broad reach. The system then receives access information from the first-target users when their shared links are accessed, and rewards are given to them based on a preset access count. This increases the first-target users' enthusiasm for sharing marketing information and encourages sharing that starts with them. In practice, marketing campaigns can take the form of likes, analyses, or quizzes.

[0047] When the first targeted user shares a marketing campaign, the system obtains the marketing campaign information and the unique identifier in the first targeted user's node, and generates a marketing campaign sharing link with a unique identifier for the first targeted user to share. Subsequently, the first targeted user who created the sharing link can be identified through the unique identifier in the sharing link.

[0048] S300: When a potential user accesses a marketing campaign through a shared link, a unique identifier for the potential user is established. A first node corresponding to the unique identifier of the potential user is established in the relationship graph. The connection relationship between the first node and the target node is established in the relationship graph.

[0049] It should be noted that other users who access the marketing campaign through the shared link are potential users. Potential users enter the marketing campaign page for login verification. By associating the province and city in the potential user's mobile phone number with a unique identifier, a unique identifier is generated for the potential user, and the first node in the relationship graph is established. In actual implementation, the user authorization and authentication for the marketing campaign can be quickly completed through one-click login with a China Telecom account.

[0050] After establishing the first node of a potential user, the target node (first target user) corresponding to the shared link received by the first node (potential user) is determined based on the unique identifier in the shared link, and the connection relationship between the target node and the first node is established.

[0051] S400: Calculate the importance of each node based on the connection relationships between nodes in the relationship graph, and determine the second target user based on the importance of the nodes;

[0052] It's important to note that in a relationship graph, the factors influencing node importance include the number of nodes and the importance of the connected nodes. The more nodes a node connects to, the more other nodes can receive the marketing campaigns shared by that node, resulting in a larger user base and a greater impact from those users on the campaign's promotion. Therefore, the more nodes a node connects to, the higher its importance. Similarly, the higher the importance of the other nodes a node connects to, the more users those other nodes can potentially reach. Thus, the higher the importance of the nodes a node connects to, the higher the importance of that node itself.

[0053] For example, when two nodes are connected to the same number of other nodes, the importance of these other connected nodes can lead to a difference in the importance of the two nodes. Suppose two nodes are A and B. Node A connects to nodes A1, A2, and A3, while node B connects to nodes B1, B2, and B3. However, A1, A2, and A3 are connected to A11, A21, and A31 respectively, while B1, B2, and B3 are not connected to any other nodes besides node B. In this case, although node A and node B have the same number of connected nodes, the nodes connected to node A are of higher importance than those connected to node B. Therefore, node A is more important than node B. In this embodiment, to achieve a wider propagation range, the determination of node importance is more influenced by the number of connected nodes than by the quality of the connected nodes.

[0054] Since most companies have limited marketing budgets, it is difficult to reward users at every user node on the relationship graph for sharing marketing activities. Furthermore, full-coverage marketing can lead to duplicate marketing and information bombardment during the dissemination and sharing process, resulting in low marketing efficiency. Therefore, after establishing a complete relationship graph, each node can be screened according to its importance, and only users at high-importance nodes can be selected as secondary target users for the next marketing campaign information distribution. This can reduce marketing costs and improve marketing efficiency.

[0055] When distributing marketing information to the second target user, it is equivalent to returning to step S200, treating the second target user as the first target user, and developing and building a relationship graph through the spread of sharing links. Through multiple rounds of marketing campaign iterations, a rich and complete relationship graph can be established, making the target users selected based on the importance of nodes more accurate, and making the marketing efficiency higher as the relationship graph develops.

[0056] Furthermore, this indirect marketing approach hands over some of the marketing initiative to users, reducing direct interaction between marketing companies and a large number of users. It avoids user resistance and complaints caused by repeated promotional information. Compared to information bombardment marketing, this solution is very user-friendly and indirectly maintains the company's good image.

[0057] In this embodiment, a unique identifier for the first target user is established, and a corresponding target node is created in the relationship graph. Marketing activity information is distributed to the first target user, and a unique identifier corresponding to the first target user is added to the marketing activity sharing link created by the first target user. When a potential user accesses the marketing activity through the sharing link, a unique identifier for the potential user is established, a corresponding first node is created in the relationship graph, and a connection relationship between the first node and the target node is established in the relationship graph based on the unique identifier in the link. The marketing activity is disseminated to more users through sharing among users, avoiding direct contact between the enterprise and users, reducing user resistance and aversion, and establishing a desensitized relationship network with user information during the sharing process. The importance of each node is calculated based on the connection relationship between each node in the relationship graph, and a second target user is determined based on the node importance. Users who can play a greater role in the promotion of the marketing activity are accurately selected for the distribution of marketing activity information, achieving the effect of full-user promotion with less marketing cost and improving marketing efficiency.

[0058] The above is a detailed description of the first embodiment of a precision marketing method based on user relationship graphs provided in this application. The following is a detailed description of the second embodiment of a precision marketing method based on user relationship graphs provided in this application.

[0059] This embodiment further provides a specific implementation of step S400 in a precision marketing method based on user relationship graphs. Please refer to [link to relevant documentation]. Figure 2 Step S400 specifically includes steps S401-S402, as detailed below:

[0060] S401. Calculate the importance of each node based on the connection relationships between nodes in the relationship diagram;

[0061] It's important to note that, according to the six degrees of separation theory of social networks, everyone can be connected to anyone within a six-level social network. In the increasingly developed information network era, connections between people have become easier, making users who can spread information to more people via shorter paths even more important in a social network. Based on this theory, using graph computation algorithms, in the eigenvector centrality theory, calculating the eigenvector centrality of node i depends not only on the number of nodes connected to node i (i.e., equivalent to the degree of node i; the more nodes connected, the higher the degree centrality of node i), but also on the importance of the nodes connected to node i (i.e., the importance of neighboring nodes is distributed to node i). Calculating the eigenvector in a stable graph can be achieved by multiplying the adjacency matrix A of the graph by the matrix X containing the sum of the degrees of each node, resulting in the eigenvector of the first iteration. Multiplying the adjacency matrix A by the degree matrix X redistributes the importance of each node. If we iterate the eigenvector from the first iteration by left-multiplying it by matrix A multiple times, we can observe that although the value of this eigenvector increases, the proportion of each point in the graph remains relatively stable (i.e., we obtain an eigenvector M, where each element represents the eigenvector centrality of that point). When calculating the eigenvector centrality of node i in the graph, we can use a node importance model:

[0062]

[0063] In the formula, C is a proportionality constant, and x ij For the importance (i.e., degree sum) of node j connected to node i, a ij This indicates that there is an adjacency relationship between node i and node j connected to node i (in different marketing campaigns, the same two users may not necessarily have a relationship set). When node i and node j do not have a relationship, a ij The value is 0), denoted as

[0064] x = [x1, x2, x3, ..., x n ]T,

[0065] When the steady state is reached after multiple iterations, it can be written in the following matrix form:

[0066] Ax = λx.

[0067] After a user purchases goods during a marketing campaign, data such as product type, campaign type, and purchase amount are stored in the node. The number of purchases, purchase intervals, and purchase levels are also recorded to complete the basic data in the relationship graph. The weight of each user node is comprehensively evaluated based on their spending power, and this weight is used to further calculate the node's importance, increasing the accuracy of the importance assessment. In practice, users are categorized into potential users, activated users, and mature users. Potential users have zero purchases, while activated and mature users have non-zero purchases. Mature users have more purchases and purchase intervals than activated users, and there is a certain threshold; activated users can only become mature users if they reach a certain level of purchase frequency and purchase interval. The weight is evaluated using a score system from 1 to 9 points. The weight range for potential users is set to [1-3] points, for activated users to [4-6] points, and for activated users to [7-9] points. Assigning weighted labels to users facilitates the evaluation and selection of user nodes based on factors such as marketing cost budget and marketing campaign type, forming a marketing promotion user relationship graph. Then, a node importance model is used to calculate the importance of each user in the graph to the marketing campaign.

[0068] S402. Sort the nodes from highest to lowest importance, and select the number of nodes that are at the top of the importance ranking.

[0069] It should be noted that the nodes are sorted from highest to lowest importance, with higher rankings indicating greater importance. Depending on the marketing budget, the preset number or proportion of nodes can be adjusted. Generally, small and medium-sized enterprises can achieve good marketing results by selecting users corresponding to the top 20% of nodes in the importance ranking as their second target users.

[0070] The relationship graph nodes also store basic data such as user type and spending level. When selecting nodes, further filtering can be performed based on the marketing effect requirements. For example, to achieve targeted marketing to certain cities, users in cities within that province can be filtered, or to increase marketing revenue, users with high spending levels can be prioritized. User types also include the user's source channel, such as the China Telecom account platform, SMS, and WeChat official accounts. To achieve marketing effects across the entire platform, multiple user types should be considered when selecting nodes to enrich the channels for sharing marketing activities.

[0071] Furthermore, to avoid duplicate marketing and expand the marketing scope, the relationship network of nodes can be compared to reduce the number of nodes with similar connections.

[0072] S403 will use the users corresponding to the selected nodes as the second target users to send marketing campaign information.

[0073] It should be noted that after selecting the second target users, the marketing staff should be shown the importance, weight value and basic data of the second target users in the corresponding nodes. The staff can then manually adjust the list of second target users.

[0074] Based on the information stored in the second target user node, marketing activities can be further planned. Marketing strategies can be customized according to marketing costs, and marketing coupons or points can be reasonably provided to second target users for sharing marketing information, thereby promoting the sharing enthusiasm of target users. Similarly, in step S200, this interactive promotion method can also be adopted, and potential users who forward marketing activities can be rewarded based on costs to improve promotion effectiveness. In actual implementation, marketing activity content is planned according to the marketing cost budget, a marketing activity page is created, and marketing strategies are customized with incentive strategies such as coupons and points as supplementary measures. Marketing activity information and notifications of marketing activity rewards received are distributed to users through promotional SMS, APP push notifications, etc.

[0075] Furthermore, marketing copy can be generated when distributing marketing information, increasing the reach and popularity of marketing campaigns.

[0076] In this embodiment, the importance of nodes is calculated and analyzed by the connection relationships between nodes in the relationship network and the information of the nodes themselves, and the user's ability to promote marketing information is judged, which improves the accuracy of selecting target users and further increases the effect of precision marketing.

[0077] The above is a detailed description of the second embodiment of a precision marketing method based on user relationship graphs provided in this application. The following is a detailed description of the third embodiment of a precision marketing method based on user relationship graphs provided in this application.

[0078] This embodiment further provides a specific example of step S300 in a precision marketing method based on user relationship graphs. Please refer to [link to specific example]. Figure 3 Step S300 specifically includes steps S301-S303, as detailed below:

[0079] S301, establish a unique identifier for potential users who open the marketing sharing link, create a corresponding node in the relationship graph, and establish a connection relationship between the nodes of potential users and the first target users in the relationship graph based on the unique identifier in the link;

[0080] It should be noted that each user identifier corresponds to a unique node. If a target user opens the marketing analytics link, the connection between the target user nodes can be established directly without creating a unique identifier or user node.

[0081] S302, embed tracking points in the marketing interface to monitor user behavior data when browsing marketing campaign pages;

[0082] It should be noted that behavioral data specifically includes the time a user spends browsing the product details page, the number of times they browse, the types of products they browse, the number of times they click on product details, and the ordering method.

[0083] S303, incentivize the user based on the behavioral data.

[0084] It should be noted that by monitoring behavioral data to comprehensively assess users' ordering intentions, and incentivizing users to consume based on these intentions, it can be understood that the longer users spend browsing the product details page and the more times they click and browse, the stronger their willingness to purchase. However, this also reflects users' indecisiveness. Based on the specific types of products users browse and the ordering methods, and under the premise of reasonably allocating marketing costs, coupons, points, and other incentives are distributed. The specific content of the consumption incentives is displayed to users through promotional SMS messages, APP push notifications, and other means to encourage users to make a purchase decision.

[0085] Furthermore, statistical analysis of marketing data, such as the revenue and expenditure of marketing activities and changes in user types, is used to evaluate the effectiveness of a marketing campaign based on the ratio of revenue to expenditure, as well as the proportion of users who have transitioned from potential users to active users and from active users to mature users. This data provides support for the development of subsequent marketing campaigns.

[0086] Furthermore, a marketing strategy model is established. After inputting user behavior data and marketing cost budgets into the model, accurate marketing incentive strategies are evaluated. Personalized, unequal incentives are triggered for users with different ratings, such as generating batches of marketing coupons and rules for distributing marketing points. The marketing strategy model is then updated and adjusted based on the marketing data collected each time, improving the accuracy of user incentives. During the planning phase of marketing campaigns, companies can conduct refined designs based on the type of marketing campaign, marketing objectives, and expected revenue.

[0087] Furthermore, based on the statistical marketing data, marketing campaign rewards should be distributed to the primary target users. Rewards should be given based on the effectiveness of sharing the marketing campaign link. For example, by analyzing the development of the relationship graph, obtaining statistical marketing data, and the number of access messages returned when the primary target user's shared link is accessed, the specific contribution of each primary target user's sharing to the marketing campaign can be determined, allowing for more accurate reward distribution. In practice, different levels of rewards can be given for different numbers of people sharing or assisting in the sharing activity. If, during the valid period of the activity, more than three potential users visit the activity page to participate in the activity or assist the potential user, it is counted as a valid share, and a coupon of 30 off 300 can be distributed to the primary target user who created the sharing link. If more than seven valid shares are made, a coupon of 50 off 300 can be distributed, thus promoting user interaction.

[0088] Furthermore, in a more specific embodiment, after the first target user and potential user access the marketing campaign page, they can quickly complete user authentication for the marketing campaign by logging in with their Tianyi account on the Tianyi platform with one click. At the same time, the province and city of the mobile phone number used by the first target user and potential user to log in with one click are extracted to generate a de-identified unique identifier for the user and create a user node in the graph database of the relationship graph to store this basic data. The activity page on the Tianyi platform is used to embed tracking points to obtain user behavior data, and the message distribution module in the Tianyi platform is used to distribute the receipt information of the sharing marketing campaign rewards and consumption incentives to the first target user and potential user respectively.

[0089] In this embodiment, user browsing behavior data is obtained by embedding points on the marketing page, user intentions are analyzed, incentive measures are implemented to promote user consumption, and marketing data and relationship network development are statistically analyzed after the relationship network is established. Rewards are issued to the first target user to encourage the first target user to share marketing links, promote user interaction, and further improve the effectiveness of precision marketing.

[0090] The above is a detailed description of the third embodiment of a precision marketing method based on user relationship graphs provided in this application. The following is a detailed description of a precision marketing device based on user relationship graphs provided in the second aspect of this application.

[0091] Please see Figure 4 , Figure 4 This is a diagram of a precision marketing device based on a user relationship graph. This embodiment provides a precision marketing device based on a user relationship graph, including:

[0092] The node creation module 10 is used to create a unique identifier for the first target user and to create a target node corresponding to the unique identifier of each first target user in the relationship graph.

[0093] The marketing delivery module 20 is used to distribute marketing campaign information to the first target user and add a unique identifier corresponding to the first target user to the marketing campaign sharing link created by the first target user.

[0094] The account login module 30 is used to establish a unique identifier for potential users when they access marketing activities through a shared link, establish a corresponding first node in the relationship graph, and establish a connection relationship between the first node and the target node in the relationship graph based on the unique identifier in the link.

[0095] The precision marketing module 40 is used to calculate the importance of each node in the relationship graph and determine the second target user based on the importance of the node.

[0096] Furthermore, the account login module 30 includes a marketing activity page where users can quickly participate in activities by logging in with their mobile phone number. Upon logging into the marketing activity page, the user's province and city information can be extracted from the mobile phone number and stored in the node information. The marketing activity page can also embed tracking points to monitor user behavior data and execute user incentives based on this data. The account login module 30 also includes a message sending module, used to distribute incentive details to users when implementing user incentives, such as through marketing SMS, H5 marketing activity URLs, WeChat official accounts, or push notifications from the Xiaoyi Butler APP.

[0097] Furthermore, it also includes a marketing reward generation module 50, which is connected to the account login module 30 and the precision marketing module 40 respectively. It is used to generate marketing coupons and marketing points according to marketing strategies and preset or marketing strategy models. These points are used to reward the sharing of marketing links by the first target users and to incentivize user consumption behavior within the marketing cost budget.

[0098] A third aspect of this application also provides a precision marketing system based on a user relationship graph, including a processor and a memory: wherein the memory is used to store program code and transmit the program code to the processor; the processor is used to execute the precision marketing method based on the user relationship graph described in the first aspect according to the instructions in the program code.

[0099] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the above-described apparatus and system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0100] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection between apparatuses or units through some interfaces, and may be electrical, mechanical, or other forms.

[0101] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0102] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0103] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0104] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A precision marketing method based on user relationship graphs, characterized in that, include: Establish a unique identifier for the first target user, and create a target node corresponding to the unique identifier of each first target user in the relationship graph; Distribute marketing campaign information to the first-target users and add a unique identifier corresponding to the first-target users to the marketing campaign sharing links created by the first-target users; When a potential user accesses a marketing campaign through a shared link, a unique identifier for the potential user is established, and a corresponding first node is created in the relationship graph. Based on the unique identifier in the link, a connection relationship is established between the first node and the target node in the relationship graph. The importance of each node is calculated based on the connection relationships between nodes in the relationship graph. The weight value of the node is set according to the number of consumptions, consumption intervals and consumption levels of the users corresponding to the node. The importance of the node is further calculated based on the weight value. The second target user is determined based on the importance of the node. Specifically, calculating the importance of each node based on the connection relationships between nodes in the relationship graph involves calculating the node's importance based on the number of other nodes it connects to and the importance of those other nodes.

2. The precision marketing method based on user relationship graphs according to claim 1, characterized in that, The step of calculating the importance of each node based on the connection relationships between nodes in the relationship graph further includes: calculating the node importance using a node importance model, wherein the node importance model is: where x i Let C be the importance of node i, and let x be the scaling constant. ij To determine the importance of node j connected to node i, a ij Let i be the adjacency relationship between i and j, which is connected to i.

3. The precision marketing method based on user relationship graphs according to claim 1, characterized in that, The step of determining the second target users based on node importance includes: sorting nodes from high to low importance, selecting a preset number of nodes in the top importance ranking, and using the users corresponding to the selected nodes as the second target users to receive marketing campaign information; the preset number is determined based on the company's marketing costs, with higher marketing costs resulting in a higher preset number.

4. The precision marketing method based on user relationship graphs according to claim 3, characterized in that, The process of determining the second target user based on node importance also includes: obtaining basic data such as the province, city, user type, and consumption level of the corresponding user in the node, and further filtering the node according to the needs of the marketing campaign.

5. The precision marketing method based on user relationship graphs according to claim 1, characterized in that, After establishing the connection between the first node and the target node in the relationship graph, the method further includes: embedding data points in the marketing interface to monitor user behavior data when browsing marketing activity pages, and incentivizing users based on the behavior data; the behavior data includes the time spent browsing the details page, the number of times the user browses the page, the type of product browsed, the number of times the user clicks on the product details page, and the ordering method.

6. The precision marketing method based on user relationship graphs according to claim 5, characterized in that, The specific steps for incentivizing users based on the behavioral data are as follows: a marketing strategy model is established, user behavior data and marketing cost budget are input into the model to obtain marketing incentive strategies for each user, and the incentive content is displayed to users through promotional SMS, APP message push, and other means.

7. A precision marketing device based on user relationship graphs, characterized in that, include: The node creation module is used to create a unique identifier for the first target user and to create a target node corresponding to the unique identifier of each first target user in the relationship graph; The marketing delivery module is used to distribute marketing campaign information to the first-target users and add a unique identifier corresponding to the first-target user to the marketing campaign sharing link created by the first-target user. The account login module is used to establish a unique identifier for potential users when they access marketing activities through shared links, create a corresponding first node in the relationship graph, and establish a connection relationship between the first node and the target node in the relationship graph based on the unique identifier in the link. The precision marketing module is used to calculate the importance of each node based on the connection relationships between nodes in the relationship graph, set the weight value of the node based on the number of purchases, purchase intervals and purchase levels of the users corresponding to the node, further calculate the importance of the node based on the weight value, and determine the second target user based on the importance of the node; wherein, the calculation of the importance of each node based on the connection relationships between nodes in the relationship graph specifically means: calculating the importance of the node based on the number of other nodes connected to the node and the importance of other nodes.

8. The precision marketing device based on user relationship graphs according to claim 7, characterized in that, In the precision marketing module, determining the second target user based on node importance includes: sorting nodes from high to low importance, selecting the nodes that are at the top of the priority ranking, and using the users corresponding to the selected nodes as the second target users to receive marketing campaign information; the preset number is determined based on the company's marketing costs, with higher marketing costs resulting in a higher preset number.

9. A precision marketing system based on user relationship graphs, characterized in that, The system includes a processor and a memory: the memory is used to store program code and transfer the program code to the processor; The processor is used to execute the precision marketing method based on user relationship graphs as described in any one of claims 1-6 according to the instructions in the program code.

Images