Seasonally smoothing risk control methods, equipment, and storage media for e-commerce businesses

Abstract

This invention relates to the field of risk control monitoring, and discloses a method, device, and storage medium for seasonal smoothing of e-commerce business risk control. The method includes: constructing a business intensity index based on business indicators of the e-commerce business; performing distribution calculations on the business indicators according to a preset PSI formula to generate original PSI values; performing seasonal smoothing mapping processing on the business intensity index according to a preset smoothing formula to generate a seasonal smoothing factor; performing adaptive calibration processing on the original PSI values ​​according to the seasonal smoothing factor to generate seasonally smoothed PSI values; and performing risk control analysis on the e-commerce business based on the seasonally smoothed PSI values ​​to generate risk control results. In this embodiment of the invention, the result drift caused by seasonal fluctuations in existing risk control models is overcome, meeting the risk control requirements in e-commerce scenarios.

Description

Technical Field

[0001] This invention relates to the field of risk control and monitoring, and in particular to a risk control method, device and storage medium for seasonal smoothing of e-commerce business. Background Technology

[0002] The group stability index is a core indicator in the field of risk control for measuring the distribution drift of features and model outputs. In traditional risk control, the group stability index is based on the core assumption that the data distribution is stable. However, in e-commerce scenarios, the group stability index may fail due to significant distribution differences caused by actual scenarios such as peak promotional seasons.

[0003] During peak promotional seasons, sellers' operational metrics such as cash inflows, sales growth rates, and inventory ratios significantly improve. Risk control levels shift from a previously stable high-risk level to a low-risk level, causing the group stability index to far exceed the daily stable-state threshold. However, the actual operational status of sellers remains largely unchanged, resulting in distorted risk control data. Current solutions primarily rely on manual verification of false alarms, which consumes significant risk control resources and can even trigger automated credit granting suspensions, severely impacting normal business operations. Therefore, a new technology is needed to address the technical issue of existing risk control metrics becoming ineffective under seasonal data changes in e-commerce applications. Summary of the Invention

[0004] The main objective of this invention is to solve the technical problem that existing risk control indicators in e-commerce application scenarios become ineffective under seasonal data changes.

[0005] The first aspect of this invention provides a risk control method for seasonal smoothing in e-commerce business, comprising the following steps: Construct a business strength index based on e-commerce business metrics; Based on the preset PSI formula, the business indicators are distributed and calculated to generate the original PSI value; According to the preset smoothing formula, the business intensity index is subjected to seasonal smoothing mapping to generate a seasonal smoothing factor. Based on the seasonal smoothing factor, the original PSI value is adaptively calibrated to generate a seasonally smoothed PSI value; Based on the seasonally smoothed PSI value, risk control analysis is performed on the e-commerce business to generate risk control results.

[0006] Optionally, in a first implementation of the first aspect of the present invention, the step of performing distribution calculations on the business indicators according to a preset PSI formula to generate original PSI values ​​includes: ; E j ∈[10-6 [1], A j ∈[10 -6 ,1]; Among them, E j Let A be the expected distribution proportion of the j-th distribution rating. j The actual distribution proportion of the j-th distribution rating, m is the total number of distribution ratings, and PSI org This is the original PSI value.

[0007] Optionally, in a second implementation of the first aspect of the present invention, the step of performing seasonal smoothing mapping on the business intensity index according to a preset smoothing formula to generate a seasonal smoothing factor includes: ; Where k is the response coefficient, S is the seasonal smoothing factor, and λ is the business intensity index. base This serves as the benchmark strength for normal business operations.

[0008] Optionally, in a third implementation of the first aspect of the present invention, the step of constructing a business intensity index based on e-commerce business metrics includes: Normalize the business metrics of e-commerce operations to generate normalized business metrics. Based on the preset index construction formula, the normalized business indicators are weighted and synthesized to generate a business intensity index.

[0009] Optionally, in the fourth implementation of the first aspect of the present invention, the normalized business indicators include: month-on-month change rate of cash collection, month-on-month change rate of order volume, inventory-to-sales ratio, and n auxiliary parameters, wherein n is a non-negative integer, and the index construction formula includes: ; Where λ is the business intensity index, and α, β, γ, θ i R is the weight parameter. std O represents the month-on-month change rate of cash collection. std I represents the month-on-month change rate of order volume. std As the reciprocal of the inventory-to-sales ratio, A i，std Let i be the i-th auxiliary parameter, i = 1, 2, ..., n.

[0010] Optionally, in the fifth implementation of the first aspect of the present invention, the step of normalizing the business indicators of e-commerce business to generate normalized business indicators includes: ; Among them, X std For normalized business metrics, X is a business metric. min X is the historical minimum value of the business metric. maxThis represents the historical maximum value of the business indicator.

[0011] Optionally, in a sixth implementation of the first aspect of the present invention, the step of adaptively calibrating the original PSI value according to the seasonal smoothing factor to generate a seasonally smoothed PSI value includes: PSI cal =PSI org *(1-S*w) Among them, PSI cal For seasonally smoothed PSI values, PSI org is the original PSI value, S is the seasonal smoothing factor, and w is the feature sensitivity coefficient.

[0012] Optionally, in the seventh implementation of the first aspect of the present invention, the step of performing risk control analysis on the e-commerce business based on the seasonally smoothed PSI value and generating risk control results includes: Based on preset dual judgment thresholds, the range of the seasonally smoothed PSI value is analyzed to generate a judgment range; Based on the aforementioned judgment interval, a risk control result is generated.

[0013] A second aspect of the present invention provides a risk control device for seasonal smoothing of e-commerce business, comprising: a memory and at least one processor, wherein the memory stores instructions, and the memory and the at least one processor are interconnected via a line; the at least one processor invokes the instructions in the memory to cause the risk control device for seasonal smoothing of e-commerce business to execute the aforementioned risk control method for seasonal smoothing of e-commerce business.

[0014] A third aspect of the present invention provides a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the aforementioned risk control method for seasonal smoothing of e-commerce business.

[0015] In this embodiment of the invention, a business intensity index is constructed based on business indicators of e-commerce business. The business intensity index is then subjected to seasonal smoothing mapping to generate a seasonal smoothing factor. The seasonal smoothing factor is used as a parameter to balance the seasonal fluctuations of e-commerce. The original PSI value is adaptively calibrated to generate a seasonally smoothed PSI value. The seasonally smoothed PSI value is used as the basis for risk control analysis, which overcomes the result drift caused by seasonal fluctuations in existing risk control models, meets the risk control requirements in e-commerce scenarios, and solves the technical problem that existing risk control indicators in e-commerce application scenarios will fail under seasonal data changes. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of an embodiment of the risk control method for seasonal smoothing of e-commerce business in this invention. Figure 2 This is a schematic diagram of a specific embodiment of the 101 steps of the risk control method for seasonal smoothing of e-commerce business in this invention. Figure 3 This is a schematic diagram of an embodiment of a risk control device for seasonal smoothing of e-commerce business in this invention. Detailed Implementation

[0017] This invention provides a method, device, and storage medium for seasonal smoothing of risk control in e-commerce business.

[0018] The embodiments of the present invention will now be described in more detail with reference to the accompanying drawings. While some embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of the present disclosure. It should be understood that the accompanying drawings and embodiments are for illustrative purposes only and are not intended to limit the scope of protection of the present invention.

[0019] In the description of the embodiments disclosed in this invention, the term "comprising" and similar terms should be understood as open-ended inclusion, i.e., "including but not limited to". The term "based on" should be understood as "at least partially based on". The term "one embodiment" or "the embodiment" should be understood as "at least one embodiment". The terms "first", "second", etc., may refer to different or the same objects. Other explicit and implicit definitions may also be included below.

[0020] For ease of understanding, the specific process of the embodiments of the present invention is described below. Please refer to [link / reference]. Figure 1 One embodiment of the risk control method for seasonal smoothing of e-commerce business in this invention includes: 101. Construct a business intensity index based on e-commerce business metrics; In this embodiment, the business indicators of e-commerce business are mainly operational data related to risk control. Based on the operational data, indicators such as collection ability, sales growth, and inventory health constitute the business strength indicator. The business strength indicator reflects the parameters of actual e-commerce operation. By smoothing the actual business data, the parameters of the risk control model can be corrected, and the risk control results can better match the actual e-commerce operation.

[0021] For details, please refer to Figure 2 , Figure 2 This is a schematic diagram of a specific embodiment of step 101 of the risk control method for seasonal smoothing of e-commerce business in this invention. Step 101 includes the following specific implementation methods: 1011. Normalize the business metrics of e-commerce operations to generate normalized business metrics. 1012. Based on the preset index construction formula, the normalized business indicators are weighted and synthesized to generate a business intensity index.

[0022] In steps 1011-1012, the original data dimensions of all e-commerce businesses are different, so the original business indicators need to be normalized first to generate normalized business indicators.

[0023] Based on the importance of different types of business data, different weight parameters are assigned to normalized business indicators. To meet risk control requirements, the weights are generally designed from largest to smallest based on cash collection ability, sales growth, and inventory health. After calculating the normalized business indicators using the index construction formula, a business strength index is generated.

[0024] Furthermore, step 1011 includes the following specific implementation methods: ; Among them, X std For normalized business metrics, X is a business metric. min X is the historical minimum value of the business metric. max This represents the historical maximum value of the business indicator.

[0025] It should be noted that X represents the raw data of the business metrics. min X is the historical minimum value of the business metric. max This represents the historical maximum value of the business metric. The historical range here can be all historical data or data from the same historical period, and can be set according to requirements.

[0026] Furthermore, the normalized business indicators include: month-on-month change rate of cash collection, month-on-month change rate of order volume, inventory-to-sales ratio, and n auxiliary parameters, where n is a non-negative integer. The index construction formula includes: ; Where λ is the business intensity index, and α, β, γ, θ i R is the weight parameter. std O represents the month-on-month change rate of cash collection. std I represents the month-on-month change rate of order volume. std As the reciprocal of the inventory-to-sales ratio, A i，std Let i be the i-th auxiliary parameter, i = 1, 2, ..., n.

[0027] It should be noted that λ is the business intensity index, representing the business climate; the larger the value of λ, the higher the business climate. R std The α parameter represents the month-on-month change in cash collection. Cash collection is the top priority for operational stability, and the corresponding α parameter can be set to 0.4. stdThis represents the month-on-month change in order volume. Order volume growth is a leading indicator for seasonal promotions, and the corresponding β parameter can be set to 0.25. std The inverse of the inventory turnover ratio can be set to a 0-30 day inventory turnover ratio. A lower inventory turnover ratio indicates faster inventory turnover and higher store operating efficiency. The inverse of the inventory turnover ratio is I... std The γ parameter can be set to 0.15. A i，std The i-th auxiliary parameter, such as on-time delivery rate, month-on-month change rate of average order value, net operating cash flow, store opening duration, store complaint rate, and store's ranking in similar recommendations, is used to supplement and verify the authenticity of the business operations, corresponding to ∑θ. i The parameter can be set to 0.2.

[0028] 102. Based on the preset PSI formula, perform distribution calculations on the business indicators to generate the original PSI values; In this embodiment, based on the traditional PSI formula, the distribution level of business indicators is calculated to generate the original PSI value. By comparing the actual operation with the expected distribution, the operational stability of various types of merchants is calculated.

[0029] Specifically, step 102 includes the following specific implementation methods: ; E j ∈[10 -6 [1], A j ∈[10 -6 ,1]; Among them, E j Let A be the expected distribution proportion of the j-th distribution rating. j The actual distribution proportion of the j-th distribution rating, m is the total number of distribution ratings, and PSI org This is the original PSI value.

[0030] It should be noted that, regarding the evaluation of e-commerce merchants' operational capabilities, E... j The expected distribution proportion for the j-th distribution rating is the sample proportion for rating during the model development phase, representing the basis for steady-state operation. A j The actual distribution proportion of the j-th distribution rating is the current rating sample proportion of the monitor, representing the actual distribution of the current operating status. m is the total number of distribution ratings, which is designed into 11 levels in this scheme. The specific level design can be found in Table 1.

[0031] Table 1. Rating Distribution Table Rating <![CDATA[Expected distribution proportion E j > <![CDATA[Actual distribution proportion A j > <![CDATA[Percentage difference (A j -E j )]]> S 1 2.33 +1.23 A1 3 4.69 +1.69 A 4 3.79 -0.21 B1 6 6.70 +0.70 B 6 6.03 +0.03 C1 13 13.39 +0.39 C 16 16.07 +0.07 D1 17 16.74 -0.26 D 15 14.51 -0.49 E 10 8.71 -1.29 F 10 7.14 -2.86 total 100 100 0.00 103. Based on the preset smoothing formula, perform seasonal smoothing mapping on the business intensity index to generate a seasonal smoothing factor; In this embodiment, the business intensity index is subjected to a seasonal smoothing process under the constraint of a smoothing formula, which achieves a gradual change in the peak season and a gradual recovery in the off-season, thereby generating a seasonal smoothing factor.

[0032] Specifically, step 103 includes the following specific implementation methods: ; Where k is the response coefficient, S is the seasonal smoothing factor, and λ is the business intensity index. base This serves as the benchmark strength for normal business operations.

[0033] It should be noted that k is the response coefficient, with a value ranging from 8 to 15, and a default value of 12. λ base The baseline strength for normal business operations is set based on the average value during non-promotional periods. When S approaches 1, the smoothing mechanism is fully activated to smooth out fluctuations; when S approaches 0.5, only moderate smoothing is performed; when S approaches 0, the smoothing mechanism is turned off, and the difference between the original PSI value and the smoothed PSI value is not significant.

[0034] 104. Based on the seasonal smoothing factor, the original PSI value is adaptively calibrated to generate a seasonally smoothed PSI value; In this embodiment, the seasonal smoothing factor is corrected with the original PSI value and adjusted according to seasonal fluctuations to solve the problem of seasonal offset results in existing risk control models.

[0035] Specifically, step 104 includes the following specific implementation methods: PSI cal =PSI org *(1-S*w) Among them, PSI cal For seasonally smoothed PSI values, PSI org is the original PSI value, S is the seasonal smoothing factor, and w is the feature sensitivity coefficient.

[0036] It should be noted that w is the feature sensitivity coefficient, ranging from 0 to 1, representing the degree to which the risk status is affected by business seasonality. When w is 0.8, the risk level is greatly affected by operating indicators; when w approaches 1, the parameter is completely determined by operating indicators, and seasonality maximizes the suppression of the PSI value; when w approaches 0, the risk status is unrelated to seasonality and no calibration is required.

[0037] 105. Based on the seasonally smoothed PSI value, perform risk control analysis on the e-commerce business and generate risk control results.

[0038] In this embodiment, risk control analysis is performed on the e-commerce business based on the seasonally smoothed PSI value, and risk control results are generated.

[0039] The range of seasonally smoothed PSI values ​​determines the risk status. When the seasonally smoothed PSI value is less than 0.1, the distribution is considered stable and the risk is controllable. When the seasonal PSI value is not less than 0.1 and less than 0.25, the distribution is considered to be slightly off and the risk is increasing, requiring continuous monitoring. When the seasonal PSI value is not less than 0.25, the distribution is considered to be severely off, triggering a risk control alarm.

[0040] To better illustrate the advantages of smooth seasonality, the following example is provided: The business's month-on-month growth rate is 80%, the inventory-to-sales ratio is 0.5, the business intensity λ is 2.0, and the smoothing parameter k is set to 12. base Set to 1, feature sensitivity coefficient w is 0.8, and the distribution rating is referenced in Table 1.

[0041] The original PSI value is 0.285 after calculation. This 0.285 is greater than 0.25, which is considered a serious deviation and will trigger a warning alarm. During peak season, a large number of abnormal risk warnings will be triggered.

[0042] By adding a seasonal smoothing factor and substituting the parameters into the smoothing formula, the seasonal smoothing factor S can be calculated to be 0.9999.

[0043] Based on the seasonal smoothing factor S, the original PSI value is corrected and smoothed. It is known that the seasonal smoothed PSI value is 0.057. Since 0.057 is less than 0.1, the offset is considered to be a normal business fluctuation. No alarm is needed and the business can proceed normally, which is compatible with the risk control requirements of seasonal fluctuations in e-commerce scenarios.

[0044] Specifically, step 105 includes the following specific implementation methods: 1051. Based on preset dual judgment thresholds, analyze the range of the seasonally smoothed PSI value and generate a judgment range; 1052. Based on the determination interval, generate risk control results.

[0045] In steps 1051-1052, the dual judgment thresholds can be set to 0.1 and 0.25 respectively. When the seasonal smoothed PSI value is less than 0.1, the distribution is considered stable and the risk is controllable. When the seasonal PSI value is not less than 0.1 and less than 0.25, the distribution is considered to have a slight deviation and the risk is increasing, requiring continuous monitoring. When the seasonal PSI value is not less than 0.25, the distribution is considered to have a serious deviation, triggering a risk control alarm.

[0046] By using two judgment thresholds, the range of seasonally smoothed PSI values ​​is determined, and then the risk control result is finally generated based on the judgment range.

[0047] In this embodiment of the invention, a business intensity index is constructed based on business indicators of e-commerce business. The business intensity index is then subjected to seasonal smoothing mapping to generate a seasonal smoothing factor. The seasonal smoothing factor is used as a parameter to balance the seasonal fluctuations of e-commerce. The original PSI value is adaptively calibrated to generate a seasonally smoothed PSI value. The seasonally smoothed PSI value is used as the basis for risk control analysis, which overcomes the result drift caused by seasonal fluctuations in existing risk control models, meets the risk control requirements in e-commerce scenarios, and solves the technical problem that existing risk control indicators in e-commerce application scenarios will fail under seasonal data changes.

[0048] Figure 3 This is a schematic diagram of the structure of a seasonal smoothing risk control device for e-commerce business provided by an embodiment of the present invention. The seasonal smoothing risk control device 300 can vary significantly due to different configurations or performance. It may include one or more central processing units (CPUs) 310 and memory 320, and one or more storage media 330 storing application programs 333 or data 332. The memory 320 and storage media 330 can be temporary or persistent storage. The program stored in the storage media 330 may include one or more modules (not shown in the diagram), each module may include a series of instruction operations on the seasonal smoothing risk control device 300. Furthermore, the processor 310 may be configured to communicate with the storage media 330 and execute the series of instruction operations in the storage media 330 on the seasonal smoothing risk control device 300.

[0049] The risk control device 300 based on seasonal smoothing of e-commerce business may also include one or more power supplies 340, one or more wired or wireless network interfaces 350, one or more input / output interfaces 360, and / or one or more operating systems 331, such as Windows Server, Mac OS X, Unix, Linux, Free BSD, etc. Those skilled in the art will understand that... Figure 3 The illustrated risk control device structure for seasonal smoothing of e-commerce business does not constitute a limitation on risk control devices based on seasonal smoothing of e-commerce business. It may include more or fewer components than illustrated, or combine certain components, or have different component arrangements.

[0050] The present invention also provides a computer-readable storage medium, which can be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium, wherein the computer-readable storage medium stores instructions that, when the instructions are executed on a computer, cause the computer to perform the steps of the risk control method for seasonal smoothing of e-commerce business.

[0051] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0052] Furthermore, although the operations are described in a specific order, this should be understood as requiring that such operations be performed in the specific order shown or in sequential order, or requiring that all illustrated operations be performed to achieve the desired result. In certain environments, multitasking and parallel processing may be advantageous. Similarly, although several specific implementation details are included in the above discussion, these should not be construed as limiting the scope of this disclosure. Certain features described in the context of individual embodiments may also be implemented in combination in a single implementation. Conversely, various features described in the context of a single implementation may also be implemented individually or in any suitable sub-combination in multiple implementations.

[0053] Although the subject matter has been described using language specific to structural features and / or methodological logic, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or actions described above. Rather, the specific features and actions described above are merely illustrative examples of implementing the claims.

Claims

1. A risk control method for seasonal smoothing in e-commerce business, characterized in that, Including the following steps: Construct a business strength index based on e-commerce business metrics; Based on the preset PSI formula, the business indicators are distributed and calculated to generate the original PSI value; According to the preset smoothing formula, the business intensity index is subjected to seasonal smoothing mapping to generate a seasonal smoothing factor. Based on the seasonal smoothing factor, the original PSI value is adaptively calibrated to generate a seasonally smoothed PSI value; Based on the seasonally smoothed PSI value, risk control analysis is performed on the e-commerce business to generate risk control results.

2. The risk control method for seasonal smoothing of e-commerce business according to claim 1, characterized in that, The step of performing distribution calculations on the business indicators according to a preset PSI formula to generate the original PSI value includes: ； AND j ∈[10 -6 ,1],A j ∈[10 -6 ,1]; Among them, E j Let A be the expected distribution proportion of the j-th distribution rating. j The actual distribution proportion of the j-th distribution rating, m is the total number of distribution ratings, and PSI org This is the original PSI value.

3. The risk control method for seasonal smoothing of e-commerce business according to claim 1, characterized in that, The step of performing seasonal smoothing mapping on the business intensity index according to a preset smoothing formula to generate a seasonal smoothing factor includes: ； Where k is the response coefficient, S is the seasonal smoothing factor, and λ is the business intensity index. base This serves as the benchmark strength for normal business operations.

4. The risk control method for seasonal smoothing of e-commerce business according to claim 1, characterized in that, The steps for constructing the business intensity index based on e-commerce business metrics include: Normalize the business metrics of e-commerce operations to generate normalized business metrics. Based on the preset index construction formula, the normalized business indicators are weighted and synthesized to generate a business intensity index.

5. The risk control method for seasonal smoothing of e-commerce business according to claim 4, characterized in that, The normalized business indicators include: month-on-month change rate of cash collection, month-on-month change rate of order volume, inventory-to-sales ratio, and n auxiliary parameters, where n is a non-negative integer. The index construction formula includes: ； Where λ is the business intensity index, and α, β, γ, θ i R is the weight parameter. std O represents the month-on-month change rate of cash collection. std I represents the month-on-month change rate of order volume. std As the reciprocal of the inventory-to-sales ratio, A i，std Let i be the i-th auxiliary parameter, i = 1, 2, ..., n.

6. The risk control method for seasonal smoothing of e-commerce business according to claim 4, characterized in that, The steps for normalizing e-commerce business metrics to generate normalized business metrics include: ； Among them, X std For normalized business metrics, X is a business metric. min X is the historical minimum value of the business metric. max This represents the historical maximum value of the business indicator.

7. The risk control method for seasonal smoothing of e-commerce business according to claim 1, characterized in that, The step of adaptively calibrating the original PSI value according to the seasonal smoothing factor to generate a seasonally smoothed PSI value includes: PSI cal =PSI org *(1-S*w); Among them, PSI cal For seasonally smoothed PSI values, PSI org is the original PSI value, S is the seasonal smoothing factor, and w is the feature sensitivity coefficient.

8. The risk control method for seasonal smoothing of e-commerce business according to claim 1, characterized in that, The steps for performing risk control analysis on the e-commerce business based on the seasonally smoothed PSI value and generating risk control results include: Based on preset dual judgment thresholds, the range of the seasonally smoothed PSI value is analyzed to generate a judgment range; Based on the aforementioned judgment interval, a risk control result is generated.

9. A risk control device for seasonal smoothing of e-commerce business, characterized in that, The risk control device for seasonal smoothing of e-commerce business includes: a memory and at least one processor, wherein the memory stores instructions and the memory and the at least one processor are interconnected via a line; The at least one processor invokes the instructions in the memory to cause the e-commerce business seasonal smoothing risk control device to execute the e-commerce business seasonal smoothing risk control method as described in any one of claims 1-8.

10. A computer-readable storage medium storing a computer program thereon, characterized in that, When the computer program is executed by the processor, it implements the risk control method for seasonal smoothing of e-commerce business as described in any one of claims 1-8.

Images