A limited commodity anti-brush method and system based on regional weight

Abstract

The present application relates to the technical field of e-commerce transaction risk control, and particularly relates to a limited quantity commodity anti-brushing method and system based on regional weight. The method first configures sales weight for each region based on historical order data; within a preset time window after the start of the activity, regional quantity control is enabled; the current order quantity of each region is obtained in real time, and the expected sales quantity of each region is determined based on the regional sales weight; when the current order quantity of a region exceeds the expected sales quantity, the order pass rate of the region is reduced in a gradient decreasing manner to flexibly intercept orders exceeding the expected sales quantity. The present application effectively identifies and intercepts regional batch brushing, dynamically regulates regional sales quantity, protects the fair purchase rights of users nationwide, and improves inventory utilization.

Description

Technical Field

[0001] This invention relates to the field of e-commerce transaction risk control technology, and in particular to a method and system for preventing fraudulent purchases of limited-quantity goods based on regional weighting. Background Technology

[0002] E-commerce transaction risk control, as a crucial technology to ensure fair transactions on the platform, is widely used in scenarios involving limited-quantity flash sales. Among related technologies, a multi-dimensional anti-fraud system is constructed through the collaborative operation of human-machine verification, account restrictions, and frequency limits. Specifically, this system covers the entire process from user identity verification to order behavior analysis, including key aspects such as graphic verification codes, SMS verification, and purchase frequency limits.

[0003] However, existing anti-fraud methods directly rely on operational behavior identification and static quota allocation, without implementing dynamic regional adjustments at the order result level. This may lead to artificial fraudulent orders being highly similar to normal user behavior, making them difficult to distinguish, or fixed quotas causing imbalances in inventory allocation and damage to regional fairness, thereby affecting the market order and user rights of limited-edition goods sales. Summary of the Invention

[0004] The present invention aims to at least partially solve one of the technical problems in the related art.

[0005] Therefore, the first objective of this invention is to propose a method for preventing the fraudulent sale of limited-edition goods based on regional weights, comprising the following steps: S1 assigns sales volume weights to each region based on historical order data; S2, within a preset time window after the start of the activity, enables regional quantity control restrictions; S3, obtain the current order volume of each region in real time, and determine the expected sales volume of each region based on the sales volume weight of the region; S4: When the current order volume in a region exceeds its expected sales volume, the order pass rate in that region is reduced in a gradient manner to flexibly intercept orders that exceed the expected sales volume.

[0006] In one embodiment of the present invention, S1 includes: Historical order data from the previous year and the year before that were collected. The order volume percentage of each region was calculated by administrative division of the delivery address, and the sales volume weight of each region was calculated using a weighted average method.

[0007] In one embodiment of the present invention, S2 includes: Set the time window length and enable regional quantity control limits within the set time window after the activity starts.

[0008] In one embodiment of the present invention, the method further includes: after the preset time window ends, releasing the regional quantity control restriction and restoring the order pass rate of each region to 100%.

[0009] In one embodiment of the present invention, S3 includes: Multiply the total inventory of goods by the sales volume weight of the region to obtain the expected sales volume of the region.

[0010] In one embodiment of the present invention, S4 includes: The success rate of the current order is calculated based on the subscription success rate formula, which is as follows: , in, To increase the success rate of the subscription, To maximize the probability of winning the lottery, The subscription ratio for this region, This represents the sales volume weight for that region.

[0011] In one embodiment of the present invention, the maximum probability of winning the lottery. The value is 1.

[0012] In one embodiment of the present invention, the scale of the administrative division can be dynamically adjusted through a management interface, and the scale includes provincial, municipal, or district / county levels.

[0013] To achieve the above objectives, a second aspect of the present invention provides a limited-quantity goods anti-fraud system based on regional weights, comprising: The weight configuration module is used to configure sales volume weights for each region based on historical order data; The timeliness control module is used to enable regional quantity control restrictions within a preset time window after the start of the activity; The sales volume calculation module obtains the current order volume of each region in real time and determines the expected sales volume of each region based on the sales volume weight of the region. The gradient rate limiting module reduces the order pass rate for a region when the current order volume exceeds its expected sales volume, thus flexibly blocking orders that exceed the expected sales volume.

[0014] To achieve the above objectives, a third aspect of the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method described in the first aspect.

[0015] The method, system, and storage medium of this invention can effectively identify and intercept regional bulk order-brushing, realize dynamic control of regional sales volume, protect the fair purchasing rights of users nationwide, and improve inventory utilization.

[0016] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0017] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 This is a flowchart of a method for preventing the fraudulent sale of limited-edition goods based on regional weights according to an embodiment of the present invention; Figure 2 This is a structural diagram of a limited-quantity goods anti-fraud system based on regional weights according to an embodiment of the present invention; Figure 3 This is an overall technical flowchart according to an embodiment of the present invention; Figure 4 This is a subscription success rate decay curve according to an embodiment of the present invention. Detailed Implementation

[0018] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0019] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0020] The following description, with reference to the accompanying drawings, describes a method and system for preventing the fraudulent sale of limited-edition goods based on regional weights, according to an embodiment of the present invention.

[0021] Example 1 Figure 1 This is a flowchart of a method for preventing fraudulent sales of limited-edition goods based on regional weights, according to an embodiment of the present invention.

[0022] like Figure 1 As shown, the method for preventing fraudulent sales of limited-edition products based on regional weights includes the following steps: S1 assigns sales volume weights to each region based on historical order data.

[0023] Based on historical order data, sales volume weights are assigned to each region. These weights represent the expected proportion of each region in total product sales, serving as a benchmark for subsequent regional volume control decisions. Specifically, by statistically analyzing historical order data, the actual purchasing performance of each region in past sales activities can be obtained. Based on the regional purchasing power distribution patterns reflected in this historical data, a quantified sales volume weight is assigned to each region. This weight is not fixed but can be dynamically adjusted according to actual business needs, such as by adjusting the statistical time range, data source, or calculation model to optimize the accuracy of the weight.

[0024] One implementation method is to use a weighted average to comprehensively calculate order data from multiple historical periods. For example, different weight ratios can be assigned to order data from the previous year and the year before that to obtain the sales volume weight for each region. Furthermore, the weights can be manually adjusted based on product characteristics or business strategies to more accurately match the regional demand distribution of specific products. This configuration provides an objective and quantifiable regional sales benchmark for subsequent order flow control, making the sales allocation between regions more fair and reasonable.

[0025] This step, by assigning sales volume weights to each region based on historical data, achieves a quantitative assessment of regional purchasing power, providing a scientific and dynamic benchmark for subsequent regional volume control. This avoids the inventory allocation imbalance caused by traditional fixed quota allocation and improves the fairness and flexibility of sales allocation between regions.

[0026] S2 enables regional quantity control limits within a preset time window after the activity begins.

[0027] Within a preset time window after the event begins, regional volume control is activated. This step aims to implement targeted traffic management during the initial phase of a limited-quantity product flash sale, where order-faking behavior is highly concentrated. Specifically, the system pre-sets a time window length, typically corresponding to a short period after the event starts, such as a few minutes. During this window, the system activates the regional volume control function, dynamically adjusting orders from different regions based on regional sales weights. This time window is based on the understanding of the temporal distribution characteristics of order-faking behavior: automated scripts or manual order-faking often launch a large number of orders in a very short time after the event starts to seize inventory, while normal users need a certain amount of time to browse, select, and place orders. Therefore, by activating regional volume control at the beginning of the event, the system can effectively suppress the instantaneous surge in orders, preventing an abnormal increase in order volume in a particular region due to order-faking behavior, thus providing a relatively fair starting point for users in all regions. The length of the time window can be flexibly configured based on factors such as the popularity of the product, inventory, and historical risks of fraudulent orders. For example, a longer window period can be set for popular limited-edition products to strengthen protection, while a shorter window period can be set for ordinary limited-edition products to balance user experience. For regular products, this restriction may not even be enabled.

[0028] As one implementation method, the system records the start time of the activity and continuously monitors the difference between the current time and the start time. When the difference is less than the preset time window length, it is determined that the activity is within the window period, and regional quantity control is enabled. When the window period ends, the system automatically removes the restriction and restores the normal order pass rate to avoid unnecessary interference with subsequent normal purchasing behavior.

[0029] This step, by introducing a time window control mechanism, achieves precise focus and flexible control over peak periods of fraudulent orders. Without affecting the overall smoothness of the event, it significantly improves the anti-fraud effect and regional fairness in limited-quantity product flash sales, while avoiding the negative impact of long-term restrictions on user experience.

[0030] S3: Obtain the current order volume of each region in real time, and determine the expected sales volume of each region based on the sales volume weight of the region.

[0031] During the limited-time sales of goods, the system collects order data from each region in real time to obtain the current cumulative order volume for each region. This current order volume reflects the total number of orders successfully submitted in that region since the start of the event, serving as the basis for judging the sales progress and degree of abnormal concentration in a region. Simultaneously, based on the pre-configured regional sales volume weights for each region and the total inventory of the product, the system determines the expected sales volume that each region should achieve in this sales event. The expected sales volume is the core benchmark for measuring whether regional orders exceed the normal range. It is calculated by multiplying the total inventory of the product by the corresponding regional sales volume weight, thus obtaining the reasonable sales share that the region should receive under the principle of fair allocation.

[0032] This step achieves accurate quantification and real-time perception of regional sales progress by monitoring the cumulative order situation in each region in real time and dynamically determining the expected sales volume based on scientifically configured regional sales volume weights. This provides an objective and fair judgment benchmark for the subsequent gradient response mechanism, thereby effectively ensuring the balance of purchase opportunities for users in each region and avoiding misjudgments and unfairness caused by information lag or inaccurate benchmarks.

[0033] S4: When the current order volume in a region exceeds its expected sales volume, the order pass rate in that region is reduced in a gradient manner to flexibly intercept orders that exceed the expected sales volume.

[0034] When the current order volume in a region exceeds its expected sales volume, the system triggers a gradient-decreasing dynamic adjustment to the order approval rate for that region, thus flexibly intercepting orders exceeding the expected sales volume. Specifically, based on the degree to which the current order volume in the region exceeds its expected sales volume, the system calculates and dynamically adjusts the approval probability of subsequent orders in the region in real time. The higher the degree of exceeding the limit, the lower the approval rate, and the rate of decline in the approval rate exhibits a non-linear, accelerating decay trend as the degree of exceeding the limit increases. Through this gradient-decreasing control method, the system can maintain a relatively high approval rate when the regional order volume slightly exceeds expectations, only mildly intercepting a small number of orders, thereby tolerating reasonable fluctuations in normal purchasing demand; while when the regional order volume significantly exceeds expectations, the system drastically reduces the approval rate, strongly intercepting a large number of orders, thereby effectively curbing potential bulk order-brushing behavior.

[0035] As one implementation method, the system can calculate the subscription success rate based on the ratio of the subscription ratio to the regional sales volume weight (i.e., the oversubscription multiple) using an exponential decay function, for example, by using the formula:

[0036] To determine the probability of the current order being approved, where This represents the preset maximum probability of winning the lottery. This represents the current subscription ratio for the region. The weighting factor is the sales volume of the region. This formula ensures that the pass rate remains high when the over-limit multiple is close to 1, and rapidly approaches zero as the over-limit multiple increases, thus achieving a smooth transition from flexible early warning to strong blocking.

[0037] This technology employs a gradient-decreasing flexible interception mechanism, avoiding the one-size-fits-all strategy of "blocking anyone who exceeds the limit" under the traditional fixed quota. While effectively suppressing regional bulk order-brushing behavior, it significantly reduces the false blocking rate for normal users, thereby achieving a balance between efficient inventory utilization and optimized user experience while ensuring fair purchasing rights for users across the country.

[0038] Example 2 like Figure 3 As shown in the illustration, as a specific embodiment, the present invention provides a method and system for preventing the fraudulent sale of limited-edition goods based on regional weights. Its technical architecture includes the following three core modules: Module 1: Regional Weight Configuration Module Responsible for assigning sales volume weights to each region based on historical data. The weight assignment is based on: Weighted average of booking volume from the previous year and the year before last (60%:40%). Supports dynamic adjustment of administrative division scale (provincial / municipal / district / county level); Supports custom numerical allocation methods.

[0039] Specifically, the regional sales volume weighting is based on historical order data and uses a weighted average method to calculate the purchase expectation of each administrative region.

[0040] Data collection and statistics: Data collection scope: Collect complete order data from the previous one and two years of historical booking business. Statistical dimensions: The percentage of orders in each region is calculated based on the administrative division where the delivery address is located, accurate to the district / county level. Data processing: Filtering abnormal orders (such as canceled orders and refunded orders). As a specific implementation method, the weighted average method for calculating the purchase expectations of each administrative region is as follows: A weighted average method is used, with the previous year's data having a weight of 60% and the data from the year before that having a weight of 40%. Weight calculation formula: Let W(r) be the expected purchase amount in region r, then: W(r) = (Previous year's order volume(r) × 60% + Previous year's order volume(r) × 40%) / (Previous year's total national order volume × 60% + Previous year's total national order volume × 40%).

[0041] Administrative division scale configuration: The system supports dynamically adjusting the administrative division scale and numerical allocation through the management interface:

[0042] By using weighted average calculations and flexible administrative division scales, the actual distribution ratio and purchasing habits of product enthusiasts in each region can be accurately reflected, providing an objective and fair basis for regional allocation of limited-edition products.

[0043] Characteristics of weight calculation: Timeliness: Data from the previous year has higher weighting and better reflects the latest purchasing trends. Accuracy: Precise down to the district / county level to avoid masking differences within the region. Flexibility: Supports dynamic adjustment of administrative division scale and numerical allocation Adjustability: Manual adjustments can be made based on product characteristics (see "Manual Weight Adjustment Based on Product Characteristics" for details). The following is an explanation of the weighting allocation using a block of four stamps from the "Chu Ci" collection as an example: Step 1: Data Statistics

[0044] Step 2: Weighted average calculation Weighted order volume in Haidian District = 28,000 × 60% + 22,000 × 40% = 16,800 + 8,800 = 25,600 orders; Total weighted orders nationwide = 2.8 million × 60% + 2.2 million × 40% = 1.68 million + 0.88 million = 2.56 million orders; Purchase expectation in Haidian District W (Haidian District) = 25,600 / 2,560,000 = 1.0%.

[0045] Step 3: Calculation of Expected Sales Volume

[0046] Furthermore, the regional sales volume weight is calculated based on historical sales volume data for each region over the previous two years, and is a fixed calculation result. However, in practical applications, it can be manually adjusted according to the characteristics of each product.

[0047] Manual scene adjustment:

[0048] Manual adjustment principles: Relevance principle: When a product feature is highly correlated with a certain region, the weight of that region can be increased; Principle of fairness: When increasing the weight of a certain region, it is necessary to ensure that the weights of other regions are not excessively compressed; Traceability principle: Manual adjustments must be recorded, including the reason for the adjustment, the adjustment range, and the operator.

[0049] Example of manual adjustment: Take the "Soviet Premier League Commemorative Envelope" as an example: Product characteristics: Related to the Jiangsu Provincial Football Super League; Basic weighting: Orders from Jiangsu accounted for 3.5% in the first two years; Adjustment plan: The weight of Jiangsu region will be increased to 35%; Reason for adjustment: The product theme is highly relevant to Jiangsu, and purchasing demand in Jiangsu is expected to be significantly higher than the historical average.

[0050] Module 2: Time Window Control Module Responsible for controlling when to enable and disable regional quantity control restrictions. Core functions include: Activity start time record; Time window timing (e.g., the first 3 minutes); Enable regional quantity control limits within the time window; The restrictions will be lifted after the time window ends, and the normal pass rate will be restored.

[0051] Module 3: Gradient Response Decision Module Responsible for determining whether to trigger a tiered response based on real-time order data, and calculating the subscription success rate. The decision-making logic includes: Receive order parameters (third-level administrative division code, SKUId, quantity to be purchased); Check the status of the main switch for interception; Check if the product has a blocking policy configured; Calculate the subscription ratio (districtSaleRatio = current sales volume in the region / total inventory of goods); Calculate the subscription success rate: Success_Rate = maxDefenseProbability / 2^((districtSaleRatio / weight) - 1)^2, generate a random number, and determine the order result.

[0052] If the random number is less than the subscription success rate, the order is successful; otherwise, the order fails.

[0053] Specifically, this invention introduces a time window control mechanism, which enables regional quantity control restrictions within a specific time window after the start of the activity (such as the first 3 minutes), and releases the restrictions after the time window ends.

[0054] Time window setting principles:

[0055] Time window control logic: Activity start time = T0; Time window length = 3 minutes; Current time = T.

[0056] if (T - T0 < time window length): # Within the time window: Enable regional quantity control restrictions Enable gradient response mechanism = True else: # After the time window ends: Remove restrictions Enable gradient response mechanism = False Pass rate = 100% (returned to normal) The function of time window control: Prevent automated scripts from instantly buying up items: Automated scripts for popular items often place a large number of orders within seconds of the event starting. Enabling regional quantity control within the time window can limit the speed at which scripts buy up items.

[0057] To ensure that human users have the opportunity to purchase: Human users need a certain amount of time to complete the order placement process, and the time window limits the instantaneous burst of power of the script.

[0058] Non-hot items can be purchased freely: Users can purchase freely after the time window ends to avoid accidentally blocking normal users.

[0059] Specifically, the gradient response mechanism is explained below: Order parameters: When a user places an order, the system receives the following key parameters:

[0060] Subscription success rate calculation: When order volume exceeds the region's expected sales volume, the system calculates the subscription success rate. A random number is generated when a user places an order; if the random number is less than the subscription success rate, the order is successful; otherwise, the order fails. The specific subscription success rate decay curve is shown below. Figure 4 As shown.

[0061] Formula for calculating subscription success rate: Success_Rate = maxDefenseProbability / 2^((districtSaleRatio / weight) - 1)^2 in: Success_Rate ∈ (0, maxDefenseProbability) The parameters in the above formula are explained as follows:

[0062] Characteristics of the success rate curve: The curve showing the change in subscription success rate with the oversubscription multiple exhibits the following characteristics: Initial stage: When the subscription ratio does not reach the regional sales weight, the subscription success rate remains 100%. Slow decline phase: As sales in the region increase and the subscription ratio exceeds the regional sales weight by a small amount, the subscription success rate declines at a relatively slow rate.

[0063] Rapid Decline Phase: As sales in the region continue to increase, and the subscription ratio exceeds the regional sales weight by a large margin, the subscription success rate declines at an accelerated rate.

[0064] Stable phase: When the subscription ratio exceeds three times the regional sales weight, the subscription success rate gradually approaches 0.

[0065] Explanation of attenuation effect: Taking maxDefenseProbability = 1 and regional sales weight of 0.01 as an example:

[0066] Order placement logic: Generate random numbers: random_value = random(0, 1) Judgment result: if random_value <Success_Rate: Order placed successfully, order released. else: Order failed to place, order blocked. Example 3 The following example, using the sale of a block of four stamps containing the "Chu Ci" (Songs of Chu) as a specific implementation, illustrates the method proposed in this application.

[0067] Event Background

[0068] Regional weight configuration Based on a weighted average of historical booking data from the previous year and the year before last, regional sales weights are configured (accurate to the district / county level):

[0069] Configuration parameters: maxDefenseProbability = 1.0 Time window = 3 minutes III. Activity Execution Process This activity has a 3-minute time window. During this window, regional volume control restrictions are implemented; after 3 minutes, the restrictions are lifted. Taking Haidian District as an example: Within the time window (0-3 minutes): Enable regional quantity control; Phase 1: Normal Sales (0-85 sets) Order quantity: 0-85 sets; Exceeding limit multiple: ≤1.0; Subscription success rate: 100%; Action taken: Orders were released normally and all orders were accepted. User experience: Smooth purchasing process, guaranteed success.

[0070] Phase Two: Mild Warning (86-102 sets) Order quantity: 86-102 sets (over-limit multiple 1.0~1.2); Subscription success rate: 100%~97.3%; Handling: Mild blocking, approximately 0% to 3% of orders were rejected; Judgment logic: Exceeding expectations may be normal buying enthusiasm, and there is no need to strongly block it; User experience: Most users can still make purchases normally.

[0071] Phase 3: Alert Status (103-127 sets) Order quantity: 103-127 sets (over-limit multiple 1.2~1.5); Subscription success rate: 97.3%~84.1%; Response: Stricter measures were implemented, with approximately 3% to 16% of orders being rejected; Judgment logic: If the price exceeds expectations by more than 20%, it may indicate that scalpers have started placing bulk orders; User experience: Some normal users were affected, but the system effectively curbed fraudulent transactions.

[0072] Phase Four: Strong Intervention (>127 sets) Order quantity: >127 sets (exceeding the limit by more than 1.5); Subscription success rate: <84.1%, rapidly declining; Action taken: Strong measures were taken to block orders, resulting in over 16% of orders being rejected; Judgment logic: If the number exceeds expectations by more than 50%, it is highly likely to be a case of mass order fraud. User experience: Normal purchases are difficult, but the rights of users in other regions are protected; After the time window ends (3 minutes later): Restrictions are lifted; Subscription success rate: restored to 100%; Regional volume control: Remove restrictions; Action taken: Orders from all regions have been released normally. Objective: To allow ordinary users to freely purchase non-hot-selling products and avoid mistakenly blocking normal users.

[0073] IV. Order Interception Example Example 1: User A makes a purchase (normal phase) User A's order parameters: provinceCode: 110000 (Beijing) cityCode: 110100 (Beijing Municipal District) districtCode: 110108 (Haidian District) SKUId: SKU2024001 Quantity: 1 System processing flow: Query the main interception switch: On Query product SKU2024001 blocking policy: Configured Current order volume in Haidian District: 80 units Calculate the excess factor: 80 / 85 = 0.94 < 1.0 Calculating the subscription success rate: If the number of units sold is less than 85 (expected sales volume), the subscription success rate = 100%. Generate random number: 0.32 Judgment: 0.32 < 1.0, order placed successfully. Example 2: User B makes a purchase (mild warning stage) User B's order parameters: provinceCode: 110000 (Beijing) cityCode: 110100 (Beijing Municipal District) districtCode: 110108 (Haidian District) SKUId: SKU2024001 Quantity: 1 System processing flow: Query the main interception switch: On Query product SKU2024001 blocking policy: Configured Current order volume in Haidian District: 100 sets Calculate the excess multiple: 100 / 85 ≈ 1.18 Calculate the subscription success rate: Success_Rate = 1.0 / 2^((1.18-1)^2) ≈ 1.0 / 2^0.0324 ≈ 97.9% Generate random numbers: 0.85 Judgment: 0.85 < 0.979, order placed successfully. Example 3: User C makes a purchase (alert phase) User C's order parameters: provinceCode: 110000 (Beijing) cityCode: 110100 (Beijing Municipal District) districtCode: 110108 (Haidian District) SKUId: SKU2024001 Quantity: 1 System processing flow: Query the main interception switch: On Query product SKU2024001 blocking policy: Configured Current order volume in Haidian District: 120 units Calculate the excess multiple: 120 / 85 ≈ 1.41 Calculate the subscription success rate: Success_Rate = 1.0 / 2^((1.41-1)^2) ≈ 1.0 / 2^0.168 ≈ 88.9% Generate random numbers: 0.92 Judgment: 0.92 > 0.889, order failed (intercepted).

[0074] V. Interception Effectiveness Statistics

[0075] Example 4 like Figure 2 As shown, this invention proposes a limited-quantity goods anti-fraud system 10 based on regional weights, comprising: The weight configuration module 100 is used to configure sales volume weights for each region based on historical order data; The timeliness control module 200 is used to enable regional quantity control restrictions within a preset time window after the start of the activity; The sales volume calculation module 300 is used to obtain the current order volume of each region in real time and determine the expected sales volume of each region based on the sales volume weight of the region. The gradient rate limiting module 400 is used to reduce the order pass rate of a region in a gradient manner when the current order volume of a region exceeds its expected sales volume, so as to flexibly intercept orders that exceed the expected sales volume.

[0076] The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-described method for preventing fraudulent sales of limited-quantity goods based on regional weights.

[0077] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0078] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

Claims

1. A method for preventing fraudulent sales of limited-edition goods based on regional weighting, characterized in that, Includes the following steps: S1 assigns sales volume weights to each region based on historical order data; S2, within a preset time window after the start of the activity, enables regional quantity control restrictions; S3, obtain the current order volume of each region in real time, and determine the expected sales volume of each region based on the sales volume weight of the region; S4: When the current order volume in a region exceeds its expected sales volume, the order pass rate in that region is reduced in a gradient manner to flexibly intercept orders that exceed the expected sales volume.

2. The method as described in claim 1, characterized in that, S1 includes: Historical order data from the previous year and the year before that were collected. The order volume percentage of each region was calculated by administrative division of the delivery address, and the sales volume weight of each region was calculated using a weighted average method.

3. The method as described in claim 1, characterized in that, S2 includes: Set the time window length and enable regional quantity control limits within the set time window after the activity starts.

4. The method as described in claim 1, characterized in that, Also includes: After the preset time window ends, the regional quantity control restriction is lifted, and the order pass rate of each region is restored to 100%.

5. The method as described in claim 1, characterized in that, S3 includes: Multiply the total inventory of goods by the sales volume weight of the region to obtain the expected sales volume of the region.

6. The method as described in claim 1, characterized in that, S4 includes: The success rate of the current order is calculated based on the subscription success rate formula, which is as follows: ， in, To increase the success rate of the subscription, To maximize the probability of winning the lottery, The subscription ratio for this region, This represents the sales volume weight for that region.

7. The method as described in claim 6, characterized in that, The maximum probability of winning the lottery The value is 1.

8. The method as described in claim 2, characterized in that, The scale of the administrative divisions can be dynamically adjusted through the management interface, and the scale includes provincial, municipal, or district / county levels.

9. A limited-edition product anti-fraud system based on regional weights, characterized in that, include: The weight configuration module is used to configure sales volume weights for each region based on historical order data; The timeliness control module is used to enable regional quantity control restrictions within a preset time window after the start of the activity; The sales volume calculation module obtains the current order volume of each region in real time and determines the expected sales volume of each region based on the sales volume weight of the region. The gradient rate limiting module reduces the order pass rate for a region when the current order volume exceeds its expected sales volume, thus flexibly blocking orders that exceed the expected sales volume.

10. A computer-readable storage medium storing a computer program that, when executed by a processor, implements the method as claimed in any one of claims 1-8.

Images