Mall point distribution management system

Abstract

The invention discloses a mall point distribution management system, which relates to the technical field of computers and comprises a user interaction interface module, a rule engine module, a behavior data acquisition and analysis module, a point calculation and distribution execution module and an account management module. The method comprises the steps of receiving a user integral operation request; obtaining user portrait data related to the request; matching and determining an applicable integral distribution rule through a rule engine module; according to an integral distribution rule, performing integral distribution operation in combination with the user portrait data; updating a user point account; the record comprises request information and a rule matching result, and precise personalized incentive is realized through second-level hot update of a dynamic rule engine and differentiated distribution driven by a user portrait; and meanwhile, a data intelligent closed loop is constructed, the user retention rate is remarkably improved by 23%, the integral cost conversion efficiency is remarkably improved by 50%, and the problems that a traditional system is rigid in rule, low in efficiency and mismatched in resources are thoroughly solved.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a shopping mall points distribution and management system. Background Art

[0002] Traditional shopping mall points systems generally have four major defects: rigid rules (unable to dynamically adjust according to user value and scenarios), low efficiency (relying on manual review or simple triggering mechanisms, which are difficult to cope with high concurrency), data silos (behavior, transaction, and activity data are separated), and lack of intelligent analysis (allocation strategies are not combined with user portraits to quantify value). These defects lead to extensive points distribution, lagging risk control, and resource mismatch. For example, when users consume high-priced goods, static rules cannot identify wool-pulling behaviors (such as sudden large transactions by low-frequency users), and points are still issued at a fixed rate. The present invention fundamentally solves the above-mentioned technical bottlenecks through a dynamic rule engine, multi-source data fusion portraits, and intelligent quantitative models. Summary of the Invention

[0003] In order to solve the technical problems existing in the background technology, the present invention proposes a shopping mall points distribution management system.

[0004] In a first aspect, the present invention provides a shopping mall points distribution and management system, comprising:

[0005] User interaction interface module, used to receive user points operation requests;

[0006] Behavioral data collection and analysis module, used to collect user behavior data and generate user portrait data;

[0007] The rule engine module is used to store, manage and execute pre-configured points allocation rules;

[0008] a points calculation and allocation execution module, which, in response to a user points operation request received by the user interaction interface module, calls the rule engine module to determine an applicable points allocation rule, and performs points allocation calculation in combination with the user profile data generated by the behavior data collection and analysis module;

[0009] The account management module is used to maintain user point account information and update the user point account according to the allocation results of the point calculation and allocation execution module.

[0010] Furthermore, the rule engine module supports dynamic configuration and hot update of points allocation rules.

[0011] Furthermore, the behavior data collection and analysis module is configured to collect user behavior data in real time or near real time.

[0012] Furthermore, the integral calculation formula used by the integral calculation and distribution execution module to perform the integral distribution calculation is: D=0.3*A2+0.5*B1+0.2*C;

[0013] in:

[0014] D represents the integration coefficient;

[0015] A2 represents the numerical coefficient corresponding to the user registration region;

[0016] B1 represents the numerical coefficient corresponding to the user's average monthly consumption;

[0017] C represents the numerical coefficient corresponding to the user value stratification.

[0018] Furthermore, the mall points distribution management system further includes:

[0019] Real-time risk control and early warning module, used to build an abnormal behavior recognition model based on user behavior sequence analysis and dynamically adjust the points allocation coefficient;

[0020] The machine learning optimization module is used to analyze the relationship between historical points data and user repurchase rate, and automatically optimize the weight of the points calculation formula.

[0021] Furthermore, the real-time risk control and warning module uses an LSTM neural network to perform sequence modeling on the user's behavior trajectory over the past N days, where N is a configurable parameter.

[0022] In a second aspect, the present invention provides a method for managing mall points distribution, comprising the following steps:

[0023] S1. Receive user points operation request;

[0024] S2. Obtain user portrait data related to the user points operation request;

[0025] S3. Matching and determining, through a rule engine module, a points allocation rule applicable to the user's points operation request;

[0026] S4. Execute a points allocation operation based on the points allocation rule determined in step S3 and the user portrait data obtained in step S2;

[0027] S5. Update the user's points account information based on the result of the points allocation operation;

[0028] S6. Record the information of the user's points operation request, the matched points allocation rules, and the results of the points allocation operation.

[0029] Furthermore, before executing the points allocation operation in step S4, the following steps are further included:

[0030] S3.1. Dynamically adjust the parameters in the points allocation rule determined in step S3 based on the user portrait data obtained in step S2.

[0031] In a third aspect, the present invention proposes a computer-readable storage medium on which a computer program is stored, characterized in that when the computer program is executed by a processor, the steps of the above-mentioned mall points allocation management method are implemented.

[0032] In a fourth aspect, the present invention proposes an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above-mentioned mall points allocation management method when executing the computer program.

[0033] Beneficial effects of the present invention:

[0034] 1. Real-time configuration and hot updates of points distribution rules are achieved through a dynamic rule engine (e.g., anti-cheating parameters are loaded in seconds during promotions), completely resolving the rigidity of rules in traditional solutions. Combined with a fully automated execution chain (≤100ms from request trigger to account update), processing efficiency is significantly improved, and it can handle millions of concurrent requests, avoiding manual intervention errors and delays. At the same time, deep integration of multi-source data (behavior, transaction, and activity data) builds accurate user portraits, providing real-time decision-making basis for the rule engine. For example, the quantitative model D=0.3*A2+0.5*B1+0.2*C is used to automatically identify wool-pulling behaviors (e.g., sudden large-scale consumption by low-frequency users), significantly enhancing risk control accuracy and control.

[0035] 2. A differentiated allocation mechanism based on user profiles (e.g., a VIP user points multiplier of ×1.5) delivers personalized incentives, increasing the retention rate of high-value users by over 23% (measured data). Intelligent parameter dynamic adjustment capabilities (e.g., automatically balancing regional consumption differences based on the regional coefficient A2) ensure fairness in points distribution and reduce resource misallocation by 35%. Furthermore, data-driven rule optimization (e.g., analyzing the correlation between average monthly spending (B1) and repurchase rate) empowers operational strategy iteration, increasing point cost conversion efficiency by 50%, truly achieving an intelligent closed loop of "precision delivery - user growth - data feedback." BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 This is a diagram of the overall system architecture of the present invention;

[0037] Figure 2 This is a flow chart of the points allocation process of the present invention;

[0038] Figure 3 Schematic diagram of the user portrait label system of the present invention. DETAILED DESCRIPTION

[0039] Reference Figure 1-3 The present invention proposes a shopping mall points distribution management system, which consists of five modules: user interaction interface module, behavior data collection and analysis module, rule engine module, points calculation and distribution execution module, and account management module. Specifically:

[0040] 1. User interaction interface module: Receive user-triggered point operation requests (such as consumption payment, daily check-in, task completion);

[0041] 2. Behavioral data collection and analysis module: collects raw data such as user transaction flow and activity participation records in real time, and builds dynamic user profiles (including value stratification C, activity, and loyalty tags) after cleaning and integration;

[0042] 3. Rule engine module: provides a graphical configuration interface ( Figure 1 ), which allows administrators to define complex rules:

[0043] 3.1. Rule elements: conditional judgment (user attributes / behavior / time / scenario), variable reference (such as points multiplier), and calculation logic (such as points formula).

[0044] 3.2. Dynamic capabilities: support hot update of rules (such as adjusting anti-cheating parameters during promotion period, Figure 2 ).

[0045] 4. Points calculation and distribution execution module:

[0046] 4.1. Call the rule engine to match the rule corresponding to the request;

[0047] 4.2. Differentiated Calculation: Combine user profiles to achieve value stratification (e.g., increase VIP user rate);

[0048] 4.3、Execute the formula: D=0.3*A2+0.5*B1+0.2*C( Figure 3 ),in:

[0049] A2: Registration region coefficient (first-tier cities = 1.2, second-tier cities = 1.0);

[0050] B1: Average monthly consumption coefficient (tiered conversion, e.g., consumption > 5,000 yuan → coefficient 1.5);

[0051] C: Value stratification (VIP1=0.8, VIP3=1.5).

[0052] 5. Account management module: execute points increase and decrease, update account balance and generate blockchain evidence log;

[0053] 6. Real-time risk control and warning module: This module is used to build an abnormal behavior recognition model based on user behavior sequence analysis and dynamically adjust the point allocation coefficient. The module uses an LSTM neural network to perform sequence modeling on the user's behavior trajectory over the past N days, where N is a configurable parameter.

[0054] 7. Machine Learning Optimization Module: Used to analyze the relationship between historical points data and user repurchase rate, and automatically optimize the weight of the points calculation formula.

[0055] For example:

[0056] Scenario: A user purchases electronic products worth RMB 3,000;

[0057] 1. Request trigger: User payment is successful → User interaction interface receives "consumption points" request.

[0058] 2. Data collection: The behavior analysis module captures order data in real time and updates user profiles (monthly consumption amount B1 rises to the "high net worth" level).

[0059] 3. Rule matching: The rule engine activates two rules:

[0060] 3.1. Basic rules: Order amount > 100 yuan → Basic points = amount × 0.5;

[0061] 3.2. Risk control rules for big promotions: If D < 0.6, mark the order as abnormal (dynamic loading parameters to prevent freeloaders).

[0062] 4. Points calculation:

[0063] 4.1. Retrieving profile data: A2 = 1.2 (registered in a first-tier city), B1 = 1.5 (monthly consumption exceeding 10,000 yuan), C = 1.5 (VIP3);

[0064] 4.2. Calculation coefficient: D = 0.3*1.2 + 0.5*1.5 + 0.2*1.5 = 1.41;

[0065] 4.3. Final score: 3000 × 0.5 × 1.5 (VIP multiplier) = 2250 points (since D>0.6, it is considered a legitimate order).

[0066] 5. Execution and Evidence Storage:

[0067] 5.1. The account module adds 2250 points;

[0068] 5.2. Generate a log (including rule ID, calculation parameters, and blockchain hash) and store it in the on-chain database.

[0069] It should be noted that;

[0070] 1) Core differences from traditional systems

[0071] Comparison of the traditional dimensional integration system and the technical solution of the present invention

[0072] Rule flexibility static configuration, need to restart the system to update the rules dynamic hot update, support second-level rule parameter adjustment

[0073] Data integration capabilities: Isolated storage of transaction data; Real-time fusion of multi-source data (behavior / transaction / risk control data)

[0074] Post-audit of risk control mechanisms, manual identification of abnormal real-time behavior sequence analysis + machine learning early warning

[0075] Point efficiency is allocated at a fixed rate, resource mismatch rate is high, and parameters are adjusted dynamically based on user value, increasing cost conversion efficiency by 50%.

[0076] (2) Advantages of innovative technology combinations

[0077] "Dynamic rules + real-time portrait" dual-drive mechanism

[0078] Through the two-way interaction between the rule engine and user profiles, a closed loop of "scenario perception - rule adaptation - precise allocation" is achieved. For example, a high-frequency consumer triggers the "loyalty bonus rule", and the points multiplier is automatically increased by 20%;

[0079] When it is detected that the user is placing an order using the APP for the first time, the "Newcomer Exclusive Points Gift Package" rule is dynamically inserted.

[0080] The self-evolution capability of intelligent quantitative models

[0081] The machine learning module continuously analyzes the effectiveness of point distribution (such as the repurchase rate brought by every 100 points), automatically optimizes the formula weights, and forms a self-optimization cycle of "data feedback-strategy iteration", which is 80% more efficient than traditional manual parameter adjustment.

[0082] Technical integration of blockchain + stream computing

[0083] Blockchain evidence storage is used to ensure transparent and traceable point distribution, and combined with the stream computing framework, sub-second response is achieved under millions of concurrent requests (measured processing delay ≤ 80ms), breaking through the performance bottleneck of traditional systems.

[0084] Specific implementation cases

[0085] User behavior: A user completed five large orders using different devices in Beijing, Shanghai, and Guangzhou within three days.

[0086] System processing flow: The real-time risk control module captures behavioral sequence anomalies (large geographical span + frequent equipment changes) and generates a risk level label of "high risk";

[0087] The rule engine dynamically inserts the risk control rule: "when the risk level is greater than or equal to high risk, the integral coefficient D is multiplied by a 0.5 correction factor";

[0088] Integral calculation: assuming that the original D is 1.2 and the corrected D is 0.6, the final integral is order amount x basic multiplier x 0.6;

[0089] The blockchain storage record risk rule trigger log is provided for subsequent auditing by the operator.

[0090] The present application breaks through the technical bottleneck of traditional integral system rule rigidity and data fragmentation through the technical combination of "dynamic rule engine + real-time user portrait + intelligent quantification model".

[0091] Among them: the dynamic hot update mechanism solves the timeliness problem of rule adjustment, which is a significant improvement compared with the traditional system which needs to restart the service;

[0092] Multi-source data fusion portrait realizes the transformation from "transaction data driven" to "behavior-value-risk multi-dimensional data driven", and improves the accuracy of distribution strategy;

[0093] Machine learning self-optimization model first combines integral distribution efficiency with algorithm parameter adjustment to form a data closed loop, which is an innovative technical path in the field.

[0094] The combination of the above technical solutions not only solves the defects of the prior art, but also constructs an ecological closed loop of "intelligent distribution-effect feedback-strategy evolution", which has outstanding substantial features and significant progress.

[0095] The above is only the preferred specific embodiment of the present application, but the protection scope of the present application is not limited thereto, any person skilled in the art can make equivalent replacement or change according to the technical scheme and inventive concept of the present application within the technical range disclosed by the present application, which should be covered within the protection scope of the present application.

Claims

1. A shopping mall points distribution management system, characterized in that: include: User interaction interface module, used to receive user points operation requests; Behavioral data collection and analysis module, used to collect user behavior data and generate user portrait data; The rule engine module is used to store, manage and execute pre-configured points allocation rules; a points calculation and allocation execution module, which, in response to a user points operation request received by the user interaction interface module, calls the rule engine module to determine an applicable points allocation rule, and performs points allocation calculation in combination with the user profile data generated by the behavior data collection and analysis module; The account management module is used to maintain user point account information and update the user point account according to the allocation results of the point calculation and allocation execution module.

2. The mall points distribution management system according to claim 1, characterized in that: The rule engine module supports dynamic configuration and hot update of points allocation rules.

3. The mall points distribution management system according to claim 1, characterized in that: The behavior data collection and analysis module is configured to collect user behavior data in real time or near real time.

4. The mall points distribution management system according to any one of claims 1 to 3, characterized in that: The integral calculation formula used by the integral calculation and distribution execution module to perform the integral distribution calculation is: D=0.3*A2+0.5*B1+0.2*C; in: D represents the integral coefficient; A2 represents the numerical coefficient corresponding to the user registration region; B1 represents the numerical coefficient corresponding to the user's average monthly consumption; C represents the numerical coefficient corresponding to the user value stratification.

5. The mall points distribution management system according to claim 1, characterized in that: Also includes: Real-time risk control and early warning module, used to build an abnormal behavior recognition model based on user behavior sequence analysis and dynamically adjust the points allocation coefficient; The machine learning optimization module is used to analyze the relationship between historical points data and user repurchase rate, and automatically optimize the weight of the points calculation formula.

6. The mall points distribution management system according to claim 5, characterized in that: The real-time risk control and warning module uses an LSTM neural network to perform sequence modeling on the user's behavior trajectory over the past N days, where N is a configurable parameter.

7. A mall points distribution management method, characterized in that: The following steps are involved: S1. Receive user points operation request; S2. Obtain user portrait data related to the user points operation request; S3. Matching and determining, through a rule engine module, a points allocation rule applicable to the user's points operation request; S4. Execute a points allocation operation based on the points allocation rule determined in step S3 and the user portrait data obtained in step S2; S5. Update the user's points account information based on the result of the points allocation operation; S6. Record the information of the user's points operation request, the matched points allocation rules, and the results of the points allocation operation.

8. The mall points distribution management method according to claim 7, characterized in that: Before executing the points allocation operation in step S4, the method further includes the following steps: S3.

1. Dynamically adjust the parameters in the points allocation rule determined in step S3 based on the user portrait data obtained in step S2.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to claims 7-8 are implemented.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the steps of the method according to claims 7-8 are implemented.

Images