Promoter fission method, system, electronic device and medium in a recycling service scenario

Abstract

The application relates to the technical field of data processing and intelligent recommendation, and discloses a promoter fission method and system in a recycling business scenario, an electronic device and a medium. The promoter fission method comprises the following steps: acquiring invitation relationship data of users in a waste recycling platform, taking each user as a node and taking an invitation relationship as a directed edge, and adopting a graph database to construct and store a multi-level fission relationship graph between promoters and users; based on the multi-level fission relationship graph, a multi-level online-offline relationship between the promoters and the users is defined, and the multi-level online-offline relationship at least comprises a first-level offline user and a second-level offline user; when a waste recycling order is completed by any offline user, a sharing income obtained by the promoter from the order is calculated according to the level of the offline user relative to the promoter and the activity of the promoter; the application significantly improves the fission efficiency, improves the user quality, and enhances the continuity of promotion.

Description

Technical Field

[0001] This invention relates to the field of data processing and intelligent recommendation technology, specifically applied to the recycling business scenario, and specifically relates to a method, system, electronic device and medium for promoter fission in the recycling business scenario. Background Technology

[0002] In the waste recycling industry, acquiring new users is one of the core challenges for the platform's sustainable development. Traditional customer acquisition methods, including on-the-ground promotion and advertising, suffer from high costs and low efficiency. In recent years, a referral-based referral model based on social relationships has gradually emerged, encouraging existing users to become "promoters" and earn rewards by inviting friends to register and place orders. However, existing referral technologies have the following problems in the waste recycling scenario:

[0003] (1) A fission model based on simple invitation codes; In this model, each user has an invitation code, which is shared with friends who register, and both parties receive a fixed amount of reward, such as 5 yuan for each of them. The disadvantages of this fission model are: the incentive is singular, it cannot distinguish users with different promotion capabilities, high-value users (such as group leaders of community owners' groups, community leaders) receive the same reward as ordinary users, and there is a lack of motivation for continuous promotion; the reward is decoupled from subsequent transaction behavior, which can easily breed "wool party" (only register to receive rewards but do not place orders); it is difficult to identify and incentivize core promoters, and it is difficult to establish a stable promoter system.

[0004] (2) Viral marketing model based on social fission; In this model, users share posters with exclusive links to their Moments or WeChat groups, and are rewarded based on the number of clicks or registrations by their friends. The disadvantages of this model are: it emphasizes dissemination but not conversion; the registration rate is high but the order rate is low; friends click to register out of politeness, but may not actually have any unwanted items to sell; it is impossible to track the subsequent repurchase behavior of friends; the promoter's one-time effort cannot obtain continuous benefits, resulting in a rapid decline in promotion enthusiasm; there is a lack of hierarchical management and empowerment for promoters, and promoters do not know how to promote effectively, resulting in inconsistent conversion effects.

[0005] (3) Distribution model based on simple commission; In this model, the promoter receives a fixed percentage of commission for each order placed by the downline users developed by the promoter. The disadvantages of this model are: the commission rate is fixed, and it is impossible to differentiate incentives based on the promoter's contribution and user quality; there is a lack of multi-level fission mechanism, and promoters can only develop direct downlines and cannot obtain income from the downlines of their downlines (i.e., second-level and third-level), resulting in insufficient fission motivation; the commission settlement cycle is long (e.g., monthly settlement), and feedback is not timely, making it difficult to maintain the promoter's continuous enthusiasm.

[0006] (4) A crude promotion model that ignores community attributes; In this model, all promoters use the same set of promotional copy and reward rules without considering the characteristics of their community. The disadvantages of this model are: the user profiles and waste disposal habits of different communities are very different. Users in old communities may mainly use cardboard and plastic, while users in high-end communities may have more home appliances and old clothes. There is a lack of targeted promotion strategies and the conversion rate is low; it is impossible to identify key opinion leaders in the community and miss the opportunity to drive the entire community to grow through key figures.

[0007] (5) Vulnerable fission model lacking risk control mechanism; the reward rules of this model are not rigorously designed and can be exploited by black market operators or "wool party" members to obtain rewards through fake registration and fake orders. The disadvantages of this model are: the platform invests a lot of promotion funds but gets a lot of fake users and invalid orders in return, and the growth of real active users is limited; the rewards for normal users are diluted and their enthusiasm is dampened; the platform needs to invest a lot of manpower for manual review, which is costly.

[0008] In summary, existing technologies in the waste recycling scenario generally suffer from problems such as single incentives, shallow fission levels, low conversion rates, lack of community targeting, and weak risk control, resulting in high customer acquisition costs, poor user quality, and difficulty in scaling the promoter system. Summary of the Invention

[0009] The present invention aims to at least partially solve one of the aforementioned technical problems in the prior art.

[0010] To achieve the above objectives, this invention provides a method for promoter viral marketing in a business scenario, comprising the following steps:

[0011] Obtain user invitation relationship data from the waste recycling platform, and construct and store a multi-level fission relationship graph between promoters and users using a graph database, with each user as a node and the invitation relationship as a directed edge;

[0012] Based on the multi-level fission relationship graph, a multi-level upline and downline relationship between promoters and users is defined, wherein the multi-level upline and downline relationship includes at least a first-level downline and a second-level downline.

[0013] When any of the downstream users completes a waste recycling order, the commission earned by the promoter from that order is calculated based on the downstream user's level relative to the promoter and the promoter's own activity level.

[0014] Preferably, the multi-level upline / downline relationship also includes a third-level downline, wherein the third-level downline consists of users who registered using the invitation codes of the second-level downline, and the promoter and the downline users form a three-level upline / downline relationship.

[0015] Preferably, the profit sharing is calculated according to the following formula:

[0016] Reward = Amount ⋅ Rate level ⋅ K j

[0017] Where Reward is the commission earned; Amount is the transaction amount of a single order placed by a referred user; Rate level The profit sharing ratio is differentiated by level; K j This represents the promoter's activity level.

[0018] The activity coefficient is dynamically adjusted based on the promoter's own login frequency and the frequency of developing new downlines; when the promoter's activity level is higher than the threshold, K... j > 1; When the promoter's activity level is below the threshold, K j < 1.

[0019] Preferably, the promoter fission method further includes: clustering users into community grids based on their geographical location information, constructing a community profile for each community grid including community type, main waste product types, and community penetration rate, and pushing personalized promotion suggestions or targeted incentive tasks to promoters based on the community profile.

[0020] Preferably, the promoter fission method further includes: when a new user registers through a sharing link, using the multi-level fission relationship graph to trace the complete sharing and propagation path of the new user before registration, and allocating corresponding level of revenue sharing rights to each promoter node on the propagation path according to the hierarchical distance between each node in the propagation path and the new user.

[0021] Preferably, the promoter fission method further includes: identifying and intercepting cheating behavior based on at least one of device fingerprint recognition, relationship network anomaly detection, behavioral feature analysis, and LBS anomaly detection.

[0022] Preferably, the device fingerprint recognition includes: collecting the unique device identifier when the user registers; if the same device identifier is associated with more than a preset number of accounts, it is marked as a risky account.

[0023] The anomaly detection of the relationship network includes: based on the multi-level fission relationship graph, using a community detection algorithm to detect abnormal subgraphs of star-shaped or chain-shaped structures, and marking the corresponding nodes;

[0024] The behavioral feature analysis includes: monitoring the order behavior of newly registered users; if the order amount is lower than a preset threshold, orders are frequently canceled, or no order is placed after registration for more than a preset period of time, the corresponding promotional rewards will be temporarily suspended.

[0025] The LBS anomaly detection includes: verifying the consistency between the user's GPS location and IP address location during registration; if they are inconsistent, risk control is triggered.

[0026] Preferably, the graph database is the Neo4j graph database; the multi-level fission relationship graph uses directed edges to store invitation relationships, supporting fast querying and path tracing of multi-level online and offline relationships.

[0027] This invention also provides a promoter referral system for recycling business scenarios, comprising:

[0028] The multi-level fission relationship graph construction module is used to obtain the invitation relationship data of users in the waste recycling platform. With each user as a node and the invitation relationship as a directed edge, a graph database is used to construct and store the multi-level fission relationship graph between promoters and users.

[0029] The hierarchical relationship definition module is used to define multi-level upline and downline relationships between promoters and users based on the multi-level fission relationship graph, wherein the multi-level upline and downline relationships include at least first-level downlines and second-level downlines; and

[0030] The revenue calculation module is used to calculate the commission earned by the promoter from a waste recycling order when any of the downstream users complete the order, based on the downstream user's level relative to the promoter and the promoter's own activity level.

[0031] Preferably, the revenue calculation module calculates the promoter's commission according to the following formula:

[0032] Reward = Amount ⋅ Rate level ⋅ K j

[0033] Where Reward is the commission earned; Amount is the transaction amount of a single order placed by a referred user; Rate level The profit sharing ratio is differentiated by level; K j This represents the promoter's activity level.

[0034] The activity coefficient is dynamically adjusted based on the promoter's own login frequency and the frequency of developing new downlines; when the promoter's activity level is higher than the threshold, K... j > 1; When the promoter's activity level is below the threshold, K j < 1;

[0035] Preferably, the promoter fission system further includes a community profile module, which is used to cluster users into community grids based on their geographical location information, construct a community profile for each community grid including community type, main waste product types, and community penetration rate, and push personalized promotion suggestions or targeted incentive tasks to promoters based on the community profile;

[0036] Preferably, the promoter fission system further includes an attribution analysis module, which is used to trace the complete sharing and propagation path of the new user before registration using the multi-level fission relationship graph when a new user registers through a sharing link, and allocate corresponding level of revenue sharing rights to each promoter node on the propagation path according to the hierarchical distance between each node in the propagation path and the new user.

[0037] Preferably, the promoter referral system further includes a risk control module, which is used to identify and intercept fraudulent behavior based on at least one of the following methods: device fingerprint recognition, relationship network anomaly detection, behavioral feature analysis, and LBS anomaly detection.

[0038] The present invention also provides an electronic device, including a processor and a memory, wherein the memory stores machine-executable instructions that can be executed by the processor, and the processor can execute the machine-executable instructions to implement the above-described promoter fission method.

[0039] The present invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the promoter fission method as described above.

[0040] Compared with existing technologies, the promoter fission method for recycling business scenarios provided by this invention constructs a multi-level fission relationship graph based on a graph database, defines the upline and downline relationships including at least first-level and second-level downlines, and calculates revenue sharing based on the downline level and the promoter's own activity level, achieving the following technical effects:

[0041] (1) Compared to traditional fission models that only support single-level downlines, promoters cannot earn income from indirect downlines, resulting in insufficient motivation for fission. This invention enables promoters to earn commissions not only from orders placed by their direct downlines (level 1) but also from orders placed by their indirect downlines (level 2 and higher). This mechanism incentivizes promoters to actively cultivate and develop downline promoters, forming a positive mentor-apprentice relationship where "teaching apprentices benefits the master as well." The fission network expands from a single chain to a tree / network structure, driving exponential expansion of the platform's user base and significantly improving customer acquisition efficiency.

[0042] (2) When calculating revenue sharing, this invention incorporates two key factors: the hierarchical level of the downline user and the activity level of the promoter. On the one hand, promoters only receive revenue when their downline users complete actual orders, and each repeat purchase by a downline user generates revenue sharing, forcing promoters to focus on the quality of new user acquisition rather than the quantity, thus suppressing "coupon hunter" behavior at the source and improving user retention rate and lifetime value. On the other hand, the higher the promoter's activity level (frequent logins, actively developing new downlines), the higher their revenue sharing coefficient, forming a positive incentive loop of "the more active, the higher the revenue," effectively solving the problem of "one-time effort, one-time benefit" and rapid decline in enthusiasm of promoters in existing technologies, making the promoter system sustainable.

[0043] (3) Existing technologies mostly use relational databases to store invitation relationships. When the fission level reaches three levels or deeper, multiple table join queries are required to obtain the complete upstream and downstream chain, resulting in low query efficiency and difficulty in meeting real-time requirements. This invention uses a graph database (such as Neo4j graph database) to construct a multi-level fission relationship graph. By leveraging the natural path tracing advantage of graph databases, it can quickly query the complete upstream and downstream relationship chain of any user, supporting real-time revenue calculation and propagation path tracing in large-scale user scenarios. This provides a technical foundation for the large-scale application of multi-level fission mechanisms and improves the system's response speed and scalability.

[0044] This invention represents a substantial technological advancement and has significant commercial value compared to existing technologies. Attached Figure Description

[0045] Figure 1 This is a flowchart of a promoter fission method in a recycling business scenario provided by an embodiment of the present invention. Detailed Implementation

[0046] To make the technical means, creative features, objectives, and effects of this invention readily understandable, the invention is further illustrated below with reference to specific embodiments. It should be understood that the specific embodiments described herein are for illustrative and explanatory purposes only and are not intended to limit the invention.

[0047] As mentioned above, combined with Figure 1 As shown, this embodiment of the invention provides a promoter-driven growth method in a business scenario involving data recycling, comprising the following steps:

[0048] S1: Obtain the invitation relationship data of users in the waste recycling platform, take each user as a node and the invitation relationship as a directed edge, and use a graph database to construct and store the multi-level fission relationship graph between promoters and users.

[0049] In this step, the invitation relationship data includes each user's unique identifier, the invitation code they used, and the parent user corresponding to that invitation code. To efficiently store and query these complex invitation relationships, this invention uses a graph database to construct a multi-level fission relationship graph, in which each user is defined as a node. For example, when user A registers using promoter B's invitation code, a directed edge is established between node A and node B, pointing from A to B, indicating that B is A's parent promoter. In this way, the multi-level invitation relationships between promoters and users across the entire platform are completely stored in the graph database.

[0050] S2: Based on the multi-level fission relationship graph, define a multi-level upline / downline relationship between promoters and users. The multi-level upline / downline relationship includes at least a first-level downline and a second-level downline. Taking a three-level upline / downline relationship as an example, users who register directly through the promoter's invitation code are defined as first-level downlines, users who register through a first-level downline's invitation code are defined as second-level downlines, and users who register through a second-level downline's invitation code are defined as third-level downlines. Through this definition, the promoter's fission network can be extended to three levels, breaking through the limitations of traditional single-level fission.

[0051] S3: When any of a downstream user completes a waste recycling order, the promoter's commission from that order is calculated based on the downstream user's level relative to the promoter and the promoter's own activity level. Specifically, the promoter can earn commission from orders placed by direct downstream users (Level 1) and indirect downstream users (Level 2 and Level 3), with the commission decreasing progressively with increasing levels. Simultaneously, the promoter's own activity level is included in the commission calculation; higher activity levels result in higher commissions, while lower activity levels lead to lower commissions.

[0052] In this invention, the "promoter" refers to a user who has registered on a waste recycling platform and obtained an invitation code. Through the promoter fission method provided by this invention, promoters focus not only on the quantity but also on the quality of new users they attract. Furthermore, because they can receive continuous revenue sharing from their indirect downlines, promoters actively cultivate and develop their subordinate promoters, forming a positive mentor-apprentice relationship where "teaching a disciple benefits the master." Simultaneously, an activity mechanism encourages promoters to maintain their enthusiasm for promotion, preventing the phenomenon of "passive income." The use of a graph database to store the relationship graph supports rapid querying and path tracing of multi-level upline and downline relationships, providing a technical foundation for real-time revenue calculation. Ultimately, the promoter fission method provided by this invention achieves exponential expansion of the fission network, a significant improvement in user quality, and the sustainable development of the promoter system.

[0053] In embodiments of the present invention, the total number of levels in the multi-level referral relationship can be two, three, or four. In a specific embodiment of the present invention, the multi-level referral relationship further includes a third-level referral, wherein the third-level referral is a user registered through the invitation code of the second-level referral, and a three-level referral relationship is formed between the promoter and the referral user.

[0054] In this invention, when a user applies to become a promoter, the system collects their basic information and constructs an initial profile: Promoter j =[U j L j Social j InviteCode j ]; Among them, U j This includes basic user information such as name, mobile phone number, registration time, and historical transaction data; L j For geolocation information, a permanent address can be obtained via GPS to associate it with the community to which it belongs; Social j For social media influence, you can optionally fill in social media account information, such as the number of WeChat groups and the number of friends on WeChat Moments, to initially assess the promotion potential; InviteCode j This is a unique invitation code generated by the system and used to track down referral relationships.

[0055] The system can initially screen users who apply to become promoters based on historical transaction data and social influence, and give them an initial rating. The specific rating is ordinary promoter, potential promoter or core promoter, and different ratings correspond to different basic reward coefficients.

[0056] In this embodiment of the invention, the revenue sharing is calculated according to the following formula:

[0057] Reward = Amount ⋅ Rate level ⋅ K j

[0058] Among them, Reward is the commission earned; Amount is the transaction amount of a single order from a referred user, which can be understood as being automatically calculated by the system based on factors such as the type, weight, and market price of the waste products in the order; Rate level The revenue sharing ratio is differentiated by level. This ratio can be pre-configured according to the platform's operation strategy. Taking a three-level referral system as an example, the revenue sharing ratio for the first-level referral can be set to 5%, the second-level referral can be set to 2%, and the third-level referral can be set to 1%. jK represents the promoter's activity coefficient; this activity coefficient is dynamically adjusted based on the promoter's login frequency and the frequency of developing new downlines; when the promoter's activity level exceeds a threshold, K... j > 1; When the promoter's activity level is below the threshold, K j < 1. For example, the system tracks the promoter's activity every 30 days: if a promoter frequently logs into the platform and actively recruits new downlines, their activity level exceeds a preset threshold, then K j > 1, meaning the earnings are amplified; conversely, if the promoter does not log in for a long time and does not develop new downlines, their activity level will be lower than the preset threshold, then K j < 1, meaning returns are compressed; when activity is at a moderate level, K j = 1. This invention links promoters' earnings to order amount, referral level, and their own activity level. The larger the order amount, the higher the promoter's earnings, incentivizing them to attract high-value users. While the percentage decreases with deeper referral levels, the cumulative earnings across multiple levels can still generate substantial profits. The activity level coefficient creates a positive incentive loop of "the more active, the higher the earnings," effectively maintaining promoters' continued enthusiasm. Simultaneously, the real-time calculation and withdrawal mechanism provides immediate positive feedback, solving the problems of long settlement cycles and untimely feedback in existing technologies.

[0059] In this embodiment of the invention, the promoter fission method further includes: based on the user's geographical location information, such as obtaining the user's permanent address through GPS, clustering the user into community grids, constructing a community profile for each community grid including community type, main waste product types, and community penetration rate, and pushing personalized promotion suggestions or targeted incentive tasks to promoters based on the community profile.

[0060] For example, regarding community type, the type can be inferred based on POI data, such as classifying it as an older community, a high-end community, or a mixed-use community based on housing prices or surrounding facilities. The main types of waste can be identified by statistically analyzing the proportion of each type of waste based on historical order data for the community, thus revealing the main waste-selling habits of users in that community, such as the proportion of cardboard, plastic, home appliances, and used clothing. Community penetration rate is the proportion of registered users in the community to the total number of households, primarily reflecting the platform's development level. Furthermore, in building community profiles, the system can also identify users with the highest order volume within the community and mark them as potential core promoters; if a user is not yet a promoter, the system can automatically send them an invitation to join the promoter program.

[0061] Based on the aforementioned community profiles, the system provides promoters with personalized promotional suggestions or targeted incentive tasks. For example, if a community profile shows that the main waste products in the community are cardboard and plastic, the system will push cardboard recycling promotional activities and corresponding promotional copy to promoters in that community. As another example, if a community has low penetration but high potential value (such as a high-end residential area), the system will push targeted incentive tasks to promoters in that community, such as "Develop 5 new users this month and receive an extra reward of 100 yuan."

[0062] By profiling communities and enabling precise targeting, promotion has shifted from a broad, indiscriminate approach to a more targeted one. Promoters gain tailored promotional tools, resulting in a significant increase in conversion rates. The platform can identify high-value communities and concentrate resources on tackling them, achieving refined community-based operations and effectively solving the low conversion rate problem caused by neglecting community attributes in existing technologies.

[0063] It is understood that, in embodiments of the present invention, the construction of the community profile can also be based on user interest tags, which can be generated based on the user's historical search or browsing behavior.

[0064] In this embodiment of the invention, the promoter fission method further includes: when a new user registers through a sharing link, the multi-level fission relationship graph is used to trace the complete sharing propagation path of the new user before registration, and according to the hierarchical distance between each node in the propagation path and the new user, the corresponding level of revenue sharing rights are allocated to each promoter node in the propagation path.

[0065] For example, user A shares a link with user B, who then shares the link with user C. User C clicks the link and completes registration. The propagation path recorded by the system is: C←B←A. The system allocates corresponding levels of revenue sharing rights to each promoter node along the propagation path based on the hierarchical distance between each node and the new user. Specifically, the node closest to the new user (the direct sharer, i.e., B) is considered a first-level promoter and receives first-level revenue sharing; the next higher-level node (i.e., A) is considered a second-level promoter and receives second-level revenue sharing; if a longer path exists, this continues, supporting a maximum of three levels. In this way, even if a promoter is not the direct inviter, as long as they are upstream in the propagation chain, they can still earn revenue from indirect conversions.

[0066] The technical solution provided by this invention enables propagation path tracking to ensure that every contribution is visible, quantified, and rewarded. For example, user A shares a link to a large group of 500 homeowners, and many people in the group register directly, earning A a large first-level commission. Simultaneously, someone in the group forwards the link to their own smaller groups, leading to even more registrations, and A, as an upstream member of the propagation path, receives a second-level commission. This motivates promoters to share more actively on public channels (such as large groups and public accounts), thereby cultivating "super propagators" and further improving the efficiency of viral marketing. Based on the technical solution provided by this invention, the unfairness of traditional attribution models that only reward the last link in the chain is solved, ensuring that every contributor in the propagation chain receives a reasonable reward.

[0067] In embodiments of the present invention, when allocating revenue sharing rights at corresponding levels, a time decay attribution model can also be used, wherein sharing behaviors closer to the registration time are assigned a higher reward weight.

[0068] In this embodiment of the invention, the promoter fission method further includes: identifying and intercepting cheating behavior based on at least one of device fingerprint recognition, relationship network anomaly detection, behavioral feature analysis, and LBS anomaly detection;

[0069] The device fingerprint recognition includes: collecting the unique device identifier when the user registers, such as IMEI, MAC address, or device model. If the same device identifier is associated with more than a preset number of accounts, it is marked as a risky account. For example, if the same device identifier is associated with more than 3 accounts, it is considered as a batch registration fraud.

[0070] The anomaly detection of the relationship network includes: based on the multi-level fission relationship graph, using a community detection algorithm to detect abnormal subgraphs with star-shaped or chain-shaped structures, and marking the corresponding nodes; in this invention, typical abnormal structures include: star-shaped structure: a promoter develops a large number of downlines, but these downlines have no connection with each other, suspected to be machine-generated batch registration; chain-shaped structure: accounts are linearly connected, lacking the randomness of real social relationships. After detecting these abnormal subgraphs, the system marks the corresponding nodes as risky accounts.

[0071] In this invention, the community detection algorithm is a technique used to identify "small groups" in complex networks, grouping closely connected nodes together and separating sparsely connected ones. The community detection algorithm can be, for example, the Louvain algorithm.

[0072] In embodiments of the present invention, the anomaly detection of the relationship network can also employ a graph neural network to perform end-to-end learning on the multi-level fission relationship graph, automatically identifying anomalous subgraphs, thereby replacing the aforementioned community detection algorithm.

[0073] The behavioral feature analysis includes: monitoring the order behavior of newly registered users; if the order amount is lower than a preset threshold, orders are frequently canceled, or no order is placed after registration for more than a preset period of time, the corresponding promotional rewards will be temporarily suspended.

[0074] The LBS anomaly detection includes: verifying the consistency between the user's GPS location and IP address location during registration; if they are inconsistent, risk control is triggered.

[0075] For example, if the GPS shows Beijing, but the IP address is located in Guangdong; or if the registration location differs significantly from the usual login location; or if cross-city registration occurs within a short period of time, risk control will be triggered, and the account will be marked as a risky account.

[0076] Users flagged as high-risk will be subject to manual review or automatic blocking, and their rewards will not be settled temporarily. If cheating is confirmed, the account will be frozen and any rewards already issued will be recovered.

[0077] The risk control mechanism provided by this invention effectively intercepts "coupon hunters" and black market activities. For example, when a group attempted to register fake accounts in bulk using scripts and simulate orders to claim rewards, the system detected that a large number of accounts originated from the same device sequence through device fingerprinting. Simultaneously, relationship network analysis revealed that these accounts exhibited a typical star-shaped structure. All related accounts and pending rewards were immediately frozen, preventing losses of hundreds of thousands of yuan. This mechanism ensures that promotional funds are truly used to acquire high-quality users, maintaining platform fairness and the legitimate interests of promoters.

[0078] In this embodiment of the invention, the graph database is the Neo4j graph database; Neo4j is a high-performance graph database with native graph storage and processing capabilities. In the multi-level fission relationship graph, the system uses directed edges to store invitation relationships: each edge points from a downstream user to its direct upline promoter. This directed edge structure allows for rapid upward tracing of the parent chain from any node, and also allows for reverse querying of all downstream nodes. Because the underlying storage and query algorithms of the graph database are optimized for graph structures, queries of multi-level upline / downline relationships (e.g., querying all three levels of downlines of a promoter) and path tracing (e.g., finding the complete sharing and propagation chain starting from a new user) can be completed in milliseconds, far faster than multi-table join queries in traditional relational databases. This provides technical assurance for real-time revenue calculation and the scalability of large-scale fission networks.

[0079] By employing the Neo4j graph database, this invention enables the system to support rapid querying and path tracing of multi-level fission relationships in massive user scenarios, ensuring real-time response capabilities under complex operations such as multi-level incentives and multi-touchpoint attribution, thus laying the foundation for the large-scale application of fission mechanisms.

[0080] This invention also provides a promoter referral system for a recycling business scenario, used to execute the aforementioned promoter referral method. The promoter referral system includes a multi-level referral relationship graph construction module, a hierarchical relationship definition module, and a revenue calculation module. The multi-level referral relationship graph construction module is used to acquire user invitation relationship data from the waste recycling platform, using each user as a node and invitation relationships as directed edges, and constructing and storing the multi-level referral relationship graph between promoters and users using a graph database. The hierarchical relationship definition module is used to define multi-level upline / downline relationships between promoters and users based on the multi-level referral relationship graph, wherein the multi-level upline / downline relationships include at least first-level and second-level downlines. The revenue calculation module is used to calculate the promoter's share of the revenue from a waste recycling order when any downline user completes the order, based on the downline user's level relative to the promoter and the promoter's own activity level.

[0081] The three core modules of this invention work together to achieve a complete closed loop from storing fission relationships and defining hierarchies to calculating revenue. By introducing a graph database and configuring multi-level incentive rules, the system can support real-time revenue distribution in complex fission networks, significantly improving the operational efficiency of the promoter system.

[0082] In this invention, the revenue calculation module calculates the promoter's commission according to the following formula:

[0083] Reward = Amount ⋅ Rate level ⋅ K j

[0084] Among them, Reward is the commission earned; Amount is the transaction amount of a single order from a referred user, which can be understood as being automatically calculated by the system based on factors such as the type, weight, and market price of the waste products in the order; Rate level The revenue sharing ratio is differentiated by level. This ratio can be pre-configured according to the platform's operation strategy. Taking a three-level referral system as an example, the revenue sharing ratio for the first-level referral can be set to 5%, the second-level referral can be set to 2%, and the third-level referral can be set to 1%. j K represents the promoter's activity coefficient; this activity coefficient is dynamically adjusted based on the promoter's login frequency and the frequency of developing new downlines; when the promoter's activity level exceeds a threshold, K... j > 1; When the promoter's activity level is below the threshold, K j < 1.

[0085] In this embodiment of the invention, the promoter fission system further includes a community profile module. The community profile module is used to cluster users into community grids based on their geographical location information, construct a community profile for each community grid including community type, main waste product types, and community penetration rate, and push personalized promotion suggestions or targeted incentive tasks to promoters based on the community profile.

[0086] In this embodiment of the invention, the promoter fission system further includes an attribution analysis module. The attribution analysis module is used to trace the complete sharing propagation path of the new user before registration using the multi-level fission relationship graph when a new user registers through a sharing link. According to the hierarchical distance between each node in the propagation path and the new user, the module allocates corresponding level of revenue sharing rights to each promoter node in the propagation path.

[0087] In this embodiment of the invention, the promoter referral system further includes a risk control module, which is used to identify and intercept fraudulent behavior based on at least one of the following methods: device fingerprint recognition, relationship network anomaly detection, behavioral feature analysis, and LBS anomaly detection.

[0088] This invention, through the combination of the aforementioned preferred modules, enables the promoter viral growth system to not only possess core viral incentive functions but also achieve precise community penetration, fair attribution, and intelligent risk control, forming a complete and self-sustaining promoter viral growth engine. Each module can be deployed independently or in combination, and the platform can be flexibly configured according to actual business needs.

[0089] In this invention, the promoter referral system can also provide promoters with dedicated data dashboards for visually displaying personal earnings, referral networks, user activity, and promotional tasks. Personal earnings display today's earnings, this week's earnings, total earnings, and withdrawable balance. The referral network displays the number of first-level, second-level, and third-level referrals, as well as a network structure diagram. User activity displays the number of orders, total transaction amount, and active days of referred users. Promotional tasks display pending promotional tasks and estimated rewards. Furthermore, the promoter referral system can collect feedback from promoters, such as which promotional copy is effective and which communities are difficult to penetrate, to optimize community profiling models and promotional strategies.

[0090] In summary, a core technical concept of this invention is to construct an intelligent promoter viral growth model that integrates multi-level viral incentives, user value binding, precise community penetration, and intelligent risk control, achieving deep binding and viral growth of people, communities, and revenue. By constructing a multi-level viral incentive and user value binding mechanism, it stimulates promoters' continuous promotional motivation, forming a positive cycle of "promotion-revenue-re-promotion"; through community profiling and precise penetration strategies, it improves the conversion rate and community coverage of promotions; and through an intelligent risk control mechanism, it effectively prevents fraudulent behavior and reduces customer acquisition costs. Ultimately, this invention constructs a self-growing, high-quality, and sustainable promoter viral growth engine for waste recycling platforms.

[0091] This invention also provides an electronic device, including a processor and a memory, wherein the memory stores machine-executable instructions that can be executed by the processor, and the processor can execute the machine-executable instructions to implement the above-described promoter fission method.

[0092] In this invention, the processor contains a kernel, which retrieves the corresponding program units from memory. One or more kernels can be configured.

[0093] The memory may include non-permanent memory in a computer-readable storage medium, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0094] This invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the aforementioned promoter fission method.

[0095] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0096] This invention is described 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 can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0097] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0098] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0099] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0100] Memory may include non-persistent memory in computer-readable storage media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of a computer-readable storage medium.

[0101] Computer-readable storage media include both permanent and non-permanent, removable and non-removable media that can store information by any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable storage media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0102] It should also be noted that the term "comprising," or any other variation thereof, is 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. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0103] The foregoing has shown and described the basic principles, main features, and characteristics of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the invention as claimed. The scope of protection of this invention is defined by the appended claims and their equivalents.

Claims

1. A promoter-driven growth method for a recycling business scenario, characterized in that, Includes the following steps: Obtain user invitation relationship data from the waste recycling platform, and construct and store a multi-level fission relationship graph between promoters and users using a graph database, with each user as a node and the invitation relationship as a directed edge; Based on the multi-level fission relationship graph, a multi-level upline and downline relationship between promoters and users is defined, wherein the multi-level upline and downline relationship includes at least a first-level downline and a second-level downline. When any of the downstream users completes a waste recycling order, the commission earned by the promoter from that order is calculated based on the downstream user's level relative to the promoter and the promoter's own activity level.

2. The promoter fission method according to claim 1, characterized in that, The multi-level upline / downline relationship also includes a third-level downline, where the third-level downline consists of users who registered using the invitation code of the second-level downline, and the promoter and the downline users form a three-level upline / downline relationship. Preferably, the profit sharing is calculated according to the following formula: Reward = Amount ⋅ Rate level ⋅ K j； Where Reward is the commission earned; Amount is the transaction amount of a single order placed by a referred user; Rate level The profit sharing ratio is differentiated by level; K j This represents the promoter's activity level. The activity coefficient is dynamically adjusted based on the promoter's own login frequency and the frequency of developing new downlines; when the promoter's activity level is higher than the threshold, K... j > 1; When the promoter's activity level is below the threshold, K j < 1.

3. The promoter fission method according to claim 1, characterized in that, The promoter fission method also includes: clustering users into community grids based on their geographical location information, constructing a community profile for each community grid including community type, main waste product types, and community penetration rate, and pushing personalized promotion suggestions or targeted incentive tasks to promoters based on the community profile.

4. The promoter fission method according to claim 1, characterized in that, The promoter fission method further includes: when a new user registers through a sharing link, the multi-level fission relationship graph is used to trace the complete sharing and propagation path of the new user before registration, and according to the hierarchical distance between each node in the propagation path and the new user, the corresponding level of revenue sharing rights are allocated to each promoter node in the propagation path.

5. The promoter fission method according to claim 1, characterized in that, The promoter fission method also includes: identifying and intercepting cheating behavior based on at least one of the following methods: device fingerprint recognition, relationship network anomaly detection, behavioral feature analysis, and LBS anomaly detection; Preferably, the device fingerprint recognition includes: collecting the unique device identifier when the user registers; if the same device identifier is associated with more than a preset number of accounts, it is marked as a risky account. The anomaly detection of the relationship network includes: based on the multi-level fission relationship graph, using a community detection algorithm to detect abnormal subgraphs of star-shaped or chain-shaped structures, and marking the corresponding nodes; The behavioral feature analysis includes: monitoring the order behavior of newly registered users; if the order amount is lower than a preset threshold, orders are frequently canceled, or no order is placed after registration for more than a preset period of time, the corresponding promotional rewards will be temporarily suspended. The LBS anomaly detection includes: verifying the consistency between the user's GPS location and IP address location during registration; if they are inconsistent, risk control is triggered.

6. The promoter fission method according to claim 1, characterized in that, The graph database is Neo4j; the multi-level fission relationship graph uses directed edges to store invitation relationships, supporting fast querying and path tracing of multi-level online and offline relationships.

7. A promoter referral system for a recycling business scenario, characterized in that, include: The multi-level fission relationship graph construction module is used to obtain the invitation relationship data of users in the waste recycling platform. With each user as a node and the invitation relationship as a directed edge, a graph database is used to construct and store the multi-level fission relationship graph between promoters and users. The hierarchical relationship definition module is used to define the multi-level upline and downline relationship between promoters and users based on the multi-level fission relationship graph. The multi-level upline and downline relationship includes at least a first-level downline and a second-level downline. as well as The revenue calculation module is used to calculate the commission earned by the promoter from a waste recycling order when any of the downstream users complete the order, based on the downstream user's level relative to the promoter and the promoter's own activity level.

8. The promoter referral system according to claim 7, characterized in that, The revenue calculation module calculates the promoter's commission according to the following formula: Reward = Amount ⋅ Rate level ⋅ K j； Where Reward is the commission earned; Amount is the transaction amount of a single order placed by a referred user; Rate level The profit sharing ratio is differentiated by level; K j This represents the promoter's activity level. The activity coefficient is dynamically adjusted based on the promoter's own login frequency and the frequency of developing new downlines; when the promoter's activity level is higher than the threshold, K... j > 1; When the promoter's activity level is below the threshold, K j < 1; Preferably, the promoter fission system further includes a community profile module, which is used to cluster users into community grids based on their geographical location information, construct a community profile for each community grid including community type, main waste product types, and community penetration rate, and push personalized promotion suggestions or targeted incentive tasks to promoters based on the community profile; Preferably, the promoter fission system further includes an attribution analysis module, which is used to trace the complete sharing and propagation path of the new user before registration using the multi-level fission relationship graph when a new user registers through a sharing link, and allocate corresponding level of revenue sharing rights to each promoter node on the propagation path according to the hierarchical distance between each node in the propagation path and the new user. Preferably, the promoter referral system further includes a risk control module, which is used to identify and intercept fraudulent behavior based on at least one of the following methods: device fingerprint recognition, relationship network anomaly detection, behavioral feature analysis, and LBS anomaly detection.

9. An electronic device, characterized in that, It includes a processor and a memory, wherein the memory stores machine-executable instructions that can be executed by the processor, and the processor can execute the machine-executable instructions to implement the promoter fission method according to any one of claims 1-6.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the promoter fission method as described in any one of claims 1-6.

Images