WO2022048107A1

WO2022048107A1 - Multi-dimensional statistical analysis system and method for sales amounts of seller users on e-commerce platform

Info

Publication number: WO2022048107A1
Application number: PCT/CN2021/074760
Authority: WO
Inventors: 沈华; 王胜男; 张明武; 张�浩
Original assignee: 湖北工业大学
Priority date: 2020-09-02
Filing date: 2021-02-02
Publication date: 2022-03-10
Also published as: CN112163227A; CN112163227B

Abstract

Disclosed in the present invention are a multi-dimensional statistical analysis system and method for sales amounts of seller users on an e-commerce platform. The system comprises one e-commerce platform service party (ECS), n cloud servers (CSs), and m sellers. The method comprises the steps of system initialization, block compression processing for the sales amounts of the sellers, and generation of corresponding secret shares; the CSs aggregate the secret shares; and the ECS obtains multidimensional statistical analysis results. The present invention avoids using encryption and decryption operations having high calculation cost to obtain the statistical analysis results of the sales amounts of various commodities under the condition of not revealing the sales amounts of the sellers, and implements lightweight multi-dimensional statistical analysis for the sales amounts of the sellers having privacy protection. According to the obtained statistical values, the ECS can analyze and discover the distribution of the sales amounts of the various commodities on the platform. The present invention effectively solves the problem of how to efficiently perform statistical analysis on the sales conditions of the various commodities under the condition of not revealing the privacy of the seller users.

Description

A multi-dimensional statistical analysis system and method for sales of sellers and users of e-commerce platforms

This application claims the priority of the Chinese patent application filed on September 2, 2020, with the application number of 202010910937.8 and the invention titled "A system and method for multi-dimensional statistical analysis of sales of sellers and users of e-commerce platforms", all of which are The contents are incorporated herein by reference.

technical field

The invention belongs to the technical field of information security, and in particular relates to a system and method for multi-dimensional statistical analysis of sales of sellers and users of a lightweight e-commerce platform with privacy protection.

Background technique

In the field of e-commerce platforms, in order to keep abreast of the latest market situation, e-commerce platforms need to know the sales situation of various commodities of platform sellers and users. Through the statistical analysis of the sales of various commodities of the platform sellers and users, the platform can understand the overall sales of the platform sellers and users, understand the seasonal and regional laws of the sales of various commodities, and predict the future trends of the market, which is helpful for It formulates relevant platform policies to guide sellers and users to reasonably configure the inventory ratio of various commodities. However, the sales of various products of sellers and users belong to their private data, and they do not want to leak it to any other person or institution. Therefore, it is an important topic to study how to perform statistical analysis on these data without revealing the sales of various commodities of sellers.

SUMMARY OF THE INVENTION

In order to solve the above technical problems, the present invention provides a lightweight e-commerce platform seller user sales multi-dimensional statistical analysis system and method with privacy protection, which solves the problem of privacy leakage in the current commodity buying and selling process.

For achieving the above object, the present invention provides the following scheme:

A method for multi-dimensional statistical analysis of sales of sellers and users on an e-commerce platform, which is applied to a multi-dimensional statistical analysis system for sales of sellers and users on an e-commerce platform; characterized in that the system includes a platform service party ECS, n cloud servers CS and m The i-th cloud server is represented as CS _i , i=1,2,…,n; the j-th seller is represented as U _j , j=1,2,…,m; the seller U _j sells K species Different types of commodities; record the sales data of seller U _j as A _j = ( _aj,1 , _aj,2 ,..., _aj,K ), where a _j,k is a component of A _j , it represents The sales of the k-th commodity of seller U _j , where k=1,2,...,K;

The method includes the following steps:

Step 1: System initialization;

Step 1.1: The e-commerce platform service provider ECS selects the appropriate integers T and R as system parameters, where the system parameter T is used in the block processing of the sales of sellers and users, and the system parameter R is used to compress the sales data in the block. In the processing, the parameter T determines the size of each block, and the block size will affect the system performance. The larger the block size, the smaller the communication overhead but the higher the computational overhead. Therefore, the selection of the parameter T needs to weigh the performance overhead of various aspects. Selected, the selection of parameter R needs to satisfy that its value is greater than the sum of the sales of each commodity in each block. Assuming that the sales of each commodity in a period of time does not exceed W, then R needs to satisfy R>TW;

Step 1.2: The e-commerce platform service provider ECS sends the system parameters T and R to the seller;

Step 2: Compress the seller's sales data in blocks and generate corresponding secret shares;

Step 3: The cloud server aggregates the secret share;

Step 4: The ECS, the service provider of the e-commerce platform, obtains multi-dimensional statistical analysis results.

According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

(1) The present invention solves the problem of how to perform statistical analysis on these data without revealing the sales of various commodities of the sellers and users of the e-commerce platform, and has very good practical application value.

(2) In order to perform statistical analysis on these data without leaking the sales of various commodities of sellers and users, the present invention adopts the idea of block compression and the method of (t,n) threshold secret sharing, After the sales of similar commodities are compressed in blocks, the secret sharing technology is used to generate a corresponding secret share for each cloud server for the value obtained by the compression process, and each cloud server aggregates the secret share corresponding to the compressed value of the same block of all users. Finally, after receiving the aggregated secret share sent by any t cloud servers, the e-commerce platform service party can recover and extract the total sales of various commodities of all sellers and users.

(3) The present invention avoids using an encryption scheme with large computational overhead, storage overhead and communication overhead to achieve privacy protection, adopts a secret sharing technology with small computational overhead, and simultaneously uses block compression and aggregation technology to reduce storage and communication overhead, A lightweight privacy-preserving e-commerce platform seller user sales multi-dimensional statistical analysis method is implemented.

Instruction drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the accompanying drawings required in the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some of the present invention. In the embodiments, for those of ordinary skill in the art, other drawings can also be obtained according to these drawings without any creative effort.

FIG. 1 is a system frame diagram of a multi-dimensional statistical analysis method for sales of sellers and users of e-commerce platforms provided by Embodiment 1 of the present invention;

FIG. 2 is a control flow chart of a method for multi-dimensional statistical analysis of sales of sellers and users of an e-commerce platform according to Embodiment 1 of the present invention.

detailed description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

The invention provides a multi-dimensional statistical analysis system and method for the sales of sellers and users of a lightweight e-commerce platform with privacy protection, and solves the problem of privacy leakage in the current commodity buying and selling process.

In order to make the above objects, features and advantages of the present invention more clearly understood, the present invention will be described in further detail below with reference to the accompanying drawings and specific embodiments.

Example 1:

Referring to FIG. 1, an embodiment of the present invention provides a lightweight e-commerce platform seller user sales multi-dimensional statistical analysis system with privacy protection, the system includes an e-commerce platform server ECS, n cloud servers and m seller users;

Denote the i-th cloud server as CS _i , i=1,2,...,n;

Denote the jth seller as U _j , j=1,2,...,m; seller U _j sells K kinds of commodities; denote the sales data of seller Uj as A _j =(a _j,1 ,a _{j ,2} ,...,a _j,K ), where a _j,k is a component of A _j , which represents the sales of the kth commodity of seller U _j , where k=1,2,...,K.

Aiming at the above system, the embodiment of the present invention also provides a method for multi-dimensional statistical analysis of sales of sellers and users on a lightweight e-commerce platform with privacy protection. As shown in FIG. 2 , the method provided by the embodiment of the present invention includes the following steps:

Step 1: System initialization.

See Figure 2, the system initialization includes the following sub-steps:

Step 1.1: The e-commerce platform service provider ECS selects appropriate integers T and R as system parameters, where the system parameter T is used in the block processing of the sales of sellers and users, and the system parameter R is used in the compression of the sales data in the block Processing;

The parameter T is the block parameter that needs to be used in step 2, which determines the size of each block. The block size will affect the system performance. The larger the block size, the smaller the communication overhead but the higher the computational overhead. Therefore, the choice of parameter T It is necessary to weigh the performance costs of various aspects to select. The parameter R is the compression parameter that needs to be used in step 2. Its selection needs to satisfy that its value is greater than the sum of the sales of each commodity in each block. If the sales of goods do not exceed W, then R needs to satisfy R>TW;

Step 1.2: The e-commerce platform service provider ECS sends the system parameters T and R to the seller.

Step 2: Block compression processing of seller sales data and generation of corresponding secret shares. Its specific implementation includes the following sub-steps:

Step 2.1: The seller U _j divides the sales data A _j = (a _j,1 ,a _j,2 ,...,a _j,K ) of its K kinds of commodities according to the rule of T components as a group, Whether the number of commodity types K is an integer multiple of the block size T does not affect the entire processing process; the only difference is that if the number of commodity types K is not an integer multiple of the block size T, the last block is less than T components, otherwise the last block is also Contains T components; for the convenience of description, without loss of generality, it is assumed that the number of commodity types K is an integer multiple of the block size T; seller U _j will A _j = (a _j,1 ,a _j,2 ,...,a _{j, K} ) is divided into

groups, where the first group is SA _j,1 =(a _j,1 ,a _j,2 ,...,a _j,T ), and the second group is SA _j,2 =(a _j,T+1 ,a _{j ,T+2} ,…,a _j,T+T ), the third group is SA _j,3 =(a _j,2T+1 ,a _j,2T+2 ,…,a _j,2T+T ), in turn analogy.

Step 2.2: According to the obtained system parameter R, the seller U _{j calculates} the sales of the lth group SA _j,l =( _aj,(l-1)T+1 , _aj,(l-1)T+2 ,..., a _j,(l-1)T+T ) is compressed as

S _j,l is an integer, where

Step 2.3: The seller U _j uses the (t, n) threshold secret sharing scheme to generate n secret shares for each S _{j, l} , respectively expressed as

in

Seller U _j produced a total of

A secret share:

In this embodiment, the seller U _j uses the Shamir(t,n) threshold secret sharing scheme to generate n secret shares for each S _j,l , which are respectively expressed as

in

The specific process is as follows:

For S _j,l , seller U _j constructs a t-1 degree polynomial f _j,l (x)=S _j,l +b _j,l,1 x+b _j,l,2 x ² +...+b _{j ,l,t-1} x ^t-1 , where b _j,l,1 , b _j,l,2 ,...,b _j,l,t-1 are random numbers selected by seller U _j , and then generate S _j, n secret shares of _l

where i=1,2,...,n.

Seller U _j produced a total of

A secret share:

Step 2.4: The seller U _j distributes the generated n secret shares of S _jl to n cloud servers, and the seller U _j distributes the secret shares of S _jl

sent to the cloud server CS _i , where

i=1,2,...,n.

Cloud server CS _i receives from seller U _j

A secret share:

A total of received from m sellers

A secret share:

Step 3, the cloud server CS _i will share every m secret

Aggregate into a secret share, with the symbol

represents the secret share obtained by this aggregate,

Essentially the compressed value of the sum of the sales of all sellers belonging to the T category of products in block l

corresponds to a secret share, where

The above aggregation is realized by using the homomorphism of (t, n) threshold secret sharing, and the specific calculation formula is:

Obviously,

is the polynomial value of the aggregate polynomial f _l (x) when x=i, and the specific form of f _l (x) is as follows:

According to the construction method of the polynomial in the Shamir(t,n) threshold secret sharing scheme, it can be known that,

Yes

a secret share.

Repeat the above aggregation operation

times, CS _i will

secret share

compressed to

secret share

in

Yes

a secret share of

Yes

A secret share of,…,

Yes

a secret share.

In step 4, the ECS, the service provider of the e-commerce platform, obtains multi-dimensional statistical analysis results. Its specific implementation includes the following sub-steps:

Step 4.1: ECS, the e-commerce platform service provider, randomly selects t CSs and asks them to use their own information about

The secret share is sent to the e-commerce platform service provider ECS, assuming that the

The t secret shares of

According to the Lagage interpolation formula, the ESC can recover the polynomial through these t secret shares

Thus, the ESC can obtain

The e-commerce platform service provider ECS repeats the above process for a total of

times, you can restore all sellers'

Compressed value of the sum of sales of T items in blocks

Step 4.2: ECS, the e-commerce platform service provider, can extract the compressed value from the compressed value by performing T remainder operations and T-1 rounding operations for the system parameter R.

extracted from

The specific process is as follows:

Execute the 1st time

Divide by R and take the remainder, we get

Execute the 1st time

Divide by R to round, and assign the result to

have to:

Execute the 2nd time

Divide by R and take the remainder, we get

Execute the 2nd time

Divide by R to round, and assign the result to

have to:

…

Execute the T-1 time

Divide by R and take the remainder, we get

Execute the T-1 time

Divide by R to round, and assign the result to

have to:

Execute the T time

Divide by R and take the remainder, we get

times, the compressed value can be

Extract the sum of all sellers' sales for each item from:

in

Indicates the total sales of the k-th commodity of m sellers, where k=1,2,...,K.

The embodiments of the present invention effectively solve the problem of how to efficiently perform statistical analysis on the sales of various commodities of sellers and users without revealing the privacy of sellers and users. The present invention proposes a solution for the statistical analysis with privacy protection of the sales volume of sellers on the e-commerce platform, and the method can also be used for the statistical analysis of the purchase behavior of buyers on the e-commerce platform with privacy protection, and the import and export of commodities in the field of financial trade. The analysis of the situation with privacy protection and the analysis of the policyholder's insurance habits in the insurance field with privacy protection have high practicability.

It should be understood that the parts not described in detail in this specification belong to the prior art.

It should be understood that the above description of the preferred embodiments is relatively detailed, and therefore should not be considered as a limitation on the protection scope of the patent of the present invention. In the case of the protection scope, substitutions or deformations can also be made, which all fall within the protection scope of the present invention, and the claimed protection scope of the present invention shall be subject to the appended claims.

The above embodiments are provided for the purpose of describing the present invention only, and are not intended to limit the scope of the present invention. The scope of the invention is defined by the appended claims. Various equivalent replacements and modifications made without departing from the spirit and principle of the present invention should be included within the scope of the present invention.

Claims

A method for multi-dimensional statistical analysis of sales of sellers and users on an e-commerce platform, which is applied to a multi-dimensional statistical analysis system for sales of sellers and users on an e-commerce platform; characterized in that the system includes a platform service party ECS, n cloud servers CS and m The i-th cloud server is represented as CS i , i=1,2,…,n; the j-th seller is represented as U j , j=1,2,…,m; the seller U j sells K species Different types of commodities; record the sales data of seller U j as A j = ( aj,1 , aj,2 ,..., aj,K ), where a j,k is a component of A j , it represents The sales of the k-th commodity of seller U j , where k=1,2,...,K;

The method includes the following steps:

Step 1: System initialization;

Step 1.1: The e-commerce platform service provider ECS selects integers T and R as system parameters;

Step 1.2: The e-commerce platform service provider ECS sends the system parameters T and R to the seller;

Step 2: Compress the seller's sales data in blocks and generate corresponding secret shares;

Step 3: The cloud server aggregates the secret share;

Step 4: The ECS, the service provider of the e-commerce platform, obtains multi-dimensional statistical analysis results.
The method for multi-dimensional statistical analysis of sales of sellers and users on an e-commerce platform according to claim 1, wherein the specific implementation of step 2 includes the following sub-steps:

Step 2.1: The seller U j divides the sales data A j = (a j,1 ,a j,2 ,...,a j,K ) of its K kinds of commodities according to the rule of T components as a group, The seller U j divides A j = (a j,1 ,a j,2 ,...,a j,K ) into
groups, where the first group is SA j,1 =(a j,1 ,a j,2 ,...,a j,T ), and the second group is SA j,2 =(a j,T+1 ,a j ,T+2 ,…,a j,T+T ), the third group is SA j,3 =(a j,2T+1 ,a j,2T+2 ,…,a j,2T+T ), in turn analogy;

Step 2.2: According to the obtained system parameter R, the seller U j calculates the sales of the lth group SA j,l =( aj,(l-1)T+1 , aj,(l-1)T+2 ,..., a j,(l-1)T+T ) is compressed as
S j,l is an integer, where

Step 2.3: The seller U j generates n secret shares for each S j,l , denoted as
in

Seller U j produced a total of
A secret share:

Step 2.4: The seller U j distributes the generated n secret shares of S jl to n cloud servers, and the seller U j distributes the secret shares of S jl
sent to the cloud server CS i , where
i=1,2,...,n; cloud server CS i receives from seller U j
A secret share:
A total of received from m sellers
A secret share:
The method for multi-dimensional statistical analysis of sales of sellers and users on an e-commerce platform according to claim 2, wherein the specific implementation process of step 3 is: the cloud server CS i stores every m secret shares
Aggregate into a secret share, with the symbol
represents the secret share obtained by this aggregate,
Essentially the compressed value of the sum of the sales of all sellers belonging to the T category of products in block l
corresponds to a secret share, where

The above aggregation is realized by the homomorphism of (t,n) threshold secret sharing, and the specific calculation formula is:

Repeat the above aggregation operation
times, CSi will
secret share
compressed to
secret share
The method for multidimensional statistical analysis of sales of sellers and users on an e-commerce platform according to claim 3, wherein the specific implementation of step 4 includes the following sub-steps:

Step 4.1: ECS, the e-commerce platform service provider, randomly selects t CSs and asks them to use their own information about
The secret share of the e-commerce platform is sent to ECS, the service provider of the e-commerce platform, and the ECS recovers the

The e-commerce platform service provider ECS repeats the above process for a total of
times, restore all sellers'
Compressed value of the sum of sales of T items in blocks

Step 4.2: ECS, the e-commerce platform service provider, performs T remainder operations and T-1 round operations for the system parameter R from the compressed value.
extracted from

The e-commerce platform service provider ECS repeats the above process for a total of
times, from the compressed value
Extract the sum of all sellers' sales for each item from:
Indicates the total sales of the k-th commodity of m sellers, where k=1,2,...,K.
The method for multidimensional statistical analysis of sales of sellers and users on an e-commerce platform according to claim 1, wherein the specific implementation of step 1 includes the following sub-steps:

Step 1.1: The e-commerce platform service provider ECS selects the appropriate integers T and R as system parameters, where the system parameter T is used in the block processing of the sales of sellers and users, and the system parameter R is used to compress the sales data in the block. Processing;

The parameter T is the block parameter that needs to be used in step 2, which determines the size of each block. The block size will affect the system performance. The larger the block size, the smaller the communication overhead but the higher the computational overhead. Therefore, the choice of parameter T It is necessary to weigh the performance costs of various aspects to select. The parameter R is the compression parameter that needs to be used in step 2. Its selection needs to satisfy that its value is greater than the sum of the sales of each commodity in each block. If the sales of goods do not exceed W, then R needs to satisfy R>TW;

Step 1.2: The e-commerce platform service provider ECS sends the system parameters T and R to the seller.