Factor decomposition machine regression model construction method, device and readable storage medium

Through the factorization machine regression model construction method, using secret sharing technology to perform data processing in vertical federated learning, solving the problem of inability to protect data privacy in the existing technology, and achieving efficient and secure data transmission and model construction.

CN112000988BActive Publication Date: 2024-11-05WEBANK (CHINA)
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202010893497.X
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2020-08-28
Publication Date
2024-11-05
Estimated Expiration
2040-08-28

AI Technical Summary

Technical Problem

Existing vertical federated learning modeling cannot effectively protect the data privacy of participants when building regression models. Especially, there is a risk of data leakage based on two-party federated learning methods based on unencrypted two-party federated learning methods, while homomorphic encryption methods require third parties to generate keys. If third parties are not trustworthy, data privacy will still not be protected.

Method used

The factor decomposition machine regression model construction method is adopted, secretly sharing with the second device, secretly sharing model parameters and training data are obtained, vertical federated learning is performed based on secretly sharing data, secretly sharing regression error is calculated, and model parameters are updated, and all data is transmitted. All are carried out between participants without third-party encryption, ensuring data privacy.

Benefits of technology

It effectively protects the data privacy of all participants in vertical federated learning modeling, avoids the risk of data leakage, improves the security and computing efficiency of the model, and enhances the privacy of data transmission.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN112000988B_ABST
    Figure CN112000988B_ABST
Patent Text Reader

Abstract

The present application discloses a factor decomposition machine regression model construction method, device and readable storage medium, the factor decomposition machine regression model construction method includes: performing secret sharing with a second device, obtaining secret sharing model parameters and secret sharing training data, performing longitudinal federated learning modeling with the second device based on the secret sharing training data and the secret sharing model parameters, calculating the secret sharing regression error, determining the first target regression model parameters based on the secret sharing regression error, and assisting the second device in determining the second target regression model parameters to construct a longitudinal federated factor decomposition machine regression model. The present application solves the technical problem that the data privacy of each participant cannot be protected when constructing a regression model based on longitudinal federated learning modeling.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence in financial technology (Fintech), and in particular to a factor decomposition machine regression model construction method, device and readable storage medium. Background Art

[0002] With the continuous development of financial technology, especially Internet technology finance, more and more technologies (such as distributed, blockchain, artificial intelligence, etc.) are applied in the financial field, but the financial industry also puts forward higher requirements on technology, such as higher requirements on the distribution of corresponding to-do items in the financial industry.

[0003] With the continuous development of computer software and artificial intelligence, the application field of federated learning is becoming more and more extensive. At present, vertical federated learning modeling usually adopts an unencrypted two-party federated learning method or a homomorphically encrypted two-party longitudinal federated learning modeling method to construct a regression model. However, the unencrypted two-party federated learning method has the risk of data leakage and cannot protect the data privacy of the participants in the longitudinal federated learning modeling. For the homomorphically encrypted two-party longitudinal federated learning modeling method, a third party is required to generate a key pair and provide encryption and decryption services. The third party must be trustworthy. If the third party is not trustworthy or has low credibility, there is still a risk of data leakage, and the data privacy of the participants in the longitudinal federated learning modeling is still not protected. Summary of the invention

[0004] The main purpose of this application is to provide a factor decomposition machine regression model construction method, device and readable storage medium, aiming to solve the technical problem in the prior art that the data privacy of each participant cannot be protected when constructing a regression model based on vertical federated learning modeling.

[0005] To achieve the above-mentioned purpose, the present application provides a factor decomposition machine regression model construction method, which is applied to a factor decomposition machine regression model construction device, and the factor decomposition machine regression model construction method includes:

[0006] Perform secret sharing with the second device to obtain secret sharing model parameters and secret sharing training data;

[0007] Based on the secret sharing training data and the secret sharing model parameters, perform longitudinal federated learning modeling with the second device to calculate the secret sharing regression error;

[0008] Based on the secret shared regression error, first target regression model parameters are determined, and the second device is assisted in determining second target regression model parameters to construct a longitudinal federated factorization machine regression model.

[0009] The present application also provides a personalized recommendation method, which is applied to a personalized recommendation device, and the personalized recommendation method includes:

[0010] Perform secret sharing with the second device to obtain secret sharing to-be-recommended user data and secret sharing model parameters;

[0011] Inputting the secret sharing user data to be recommended into a preset scoring model to score the recommended items corresponding to the secret sharing user data to be recommended based on the secret sharing model parameters to obtain a first secret sharing scoring result;

[0012] Based on the first secret sharing scoring result, perform a federation interaction with the second device to calculate a target score in conjunction with the second secret sharing scoring result determined by the second device;

[0013] Based on the target scores, a target recommendation list corresponding to the items to be recommended is generated.

[0014] The present application also provides a factor decomposition machine regression model construction device, the factor decomposition machine regression model construction device is a virtual device, and the factor decomposition machine regression model construction device is applied to a factor decomposition machine regression model construction device, and the factor decomposition machine regression model construction device includes:

[0015] A secret sharing module, used to perform secret sharing with the second device and obtain secret sharing model parameters and secret sharing training data;

[0016] A vertical federation module, configured to perform vertical federated learning modeling with the second device based on the secret sharing training data and the secret sharing model parameters, and calculate a secret sharing regression error;

[0017] A determination module is used to determine the first target regression model parameters based on the secret shared regression error, and assist the second device in determining the second target regression model parameters to construct a longitudinal federated factor decomposition machine regression model.

[0018] The present application also provides a personalized recommendation device, which is a virtual device and is applied to a personalized recommendation device. The personalized recommendation device includes:

[0019] A secret sharing module, used to perform secret sharing with the second device, and obtain secret sharing recommended user data and secret sharing model parameters;

[0020] A scoring module, configured to input the secret sharing user data to be recommended into a preset scoring model, so as to score the recommended items corresponding to the secret sharing user data to be recommended based on the secret sharing model parameters, and obtain a first secret sharing scoring result;

[0021] a calculation module, configured to perform a federation interaction with the second device based on the first secret sharing score result, so as to calculate a target score in conjunction with a second secret sharing score result determined by the second device;

[0022] A generation module is used to generate a target recommendation list corresponding to the to-be-recommended items based on the target scores.

[0023] The present application also provides a factor decomposition machine regression model construction device, which is a physical device. The factor decomposition machine regression model construction device includes: a memory, a processor, and a program of the factor decomposition machine regression model construction method stored in the memory and executable on the processor. When the program of the factor decomposition machine regression model construction method is executed by the processor, the steps of the factor decomposition machine regression model construction method as described above can be implemented.

[0024] The present application also provides a personalized recommendation device, which is a physical device, comprising: a memory, a processor, and a program of the personalized recommendation method stored in the memory and executable on the processor. When the program of the personalized recommendation method is executed by the processor, the steps of the personalized recommendation method as described above can be implemented.

[0025] The present application also provides a readable storage medium, on which is stored a program for implementing a factor decomposition machine regression model construction method. When the program for implementing a factor decomposition machine regression model construction method is executed by a processor, the steps of the factor decomposition machine regression model construction method as described above are implemented.

[0026] The present application also provides a readable storage medium, on which is stored a program for implementing the personalized recommendation method. When the program of the personalized recommendation method is executed by a processor, the steps of the personalized recommendation method as described above are implemented.

[0027] The present application provides a factorization machine regression model construction method, device and readable storage medium. Compared with the technical means of constructing a regression model based on an unencrypted two-party federated learning method or a homomorphically encrypted two-party longitudinal federated learning modeling method adopted in the prior art, the present application obtains secret sharing model parameters and secret sharing training data by secret sharing with a second device, and then performs longitudinal federated learning modeling with the second device based on the secret sharing training data and the secret sharing model parameters, calculates the secret sharing regression error, and then updates the secret sharing model parameters based on the secret sharing regression error to obtain the secret sharing regression model update parameters, wherein, when interacting with the second device, the data sent or received are all secret sharing data, and no third party generation is required. The public-private key is used to encrypt the data, and all data transmission processes are carried out between the two parties participating in the vertical federated learning modeling, which protects the privacy of the data. Then, the parameters of the secret shared regression model are updated based on the secret shared regression model. The first target regression model parameters can be determined by decrypting interaction with the second device, and the second device is assisted in determining the second target regression model parameters, so as to complete the construction of the vertical federated factor decomposition machine regression model. This overcomes the technical defect of the prior art that the regression model is constructed based on an unencrypted two-party federated learning method or a homomorphic encrypted two-party vertical federated learning modeling method, which makes it impossible to protect the data privacy of each participant in the vertical federated learning modeling. Therefore, the technical problem that the data privacy of each participant cannot be protected when the regression model is constructed based on the vertical federated learning modeling is solved. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0029] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0030] Figure 1 This is a flow chart of the first embodiment of the factor decomposition machine regression model construction method of the present application;

[0031] Figure 2 This is a flow chart of the second embodiment of the factor decomposition machine regression model construction method of the present application;

[0032] Figure 3 This is a flowchart of the third embodiment of the personalized recommendation method of this application;

[0033] Figure 4A schematic diagram of the device structure of the hardware operating environment involved in the factor decomposition machine regression model construction method of the embodiment of the present application;

[0034] Figure 5 A schematic diagram of the device structure of the hardware operating environment involved in the personalized recommendation method of the embodiment of the present application.

[0035] The purpose, features and advantages of this application will be further described in conjunction with the embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0036] It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0037] The present application embodiment provides a factor decomposition machine regression model construction method. In the first embodiment of the factor decomposition machine regression model construction method of the present application, refer to Figure 1 The factor decomposition machine regression model construction method is applied to a first device, and the factor decomposition machine regression model construction method includes:

[0038] Step S10, performing secret sharing with the second device to obtain secret sharing model parameters and secret sharing training data;

[0039] In this embodiment, it should be noted that the first device and the second device are both participants in vertical federated learning. The first device has first-party training label data with sample labels. The first-party training label data can be represented by a first-party training data matrix and sample labels. For example, assuming that the first-party training label data is (X A , Y), X A is the first party training data matrix, Y is the sample label, and in addition, the second device has second party training data without sample labels, and the second party training data can be represented by the second party training data matrix. For example, assuming that the second party training data matrix is ​​X B .

[0040] In addition, in this embodiment, the factor decomposition machine regression model is a machine learning model constructed based on longitudinal federated learning, and the model parameters of the factor decomposition machine regression model are jointly held by the first device and the second device, wherein the factor decomposition machine regression model includes a first type model parameter and a second type model parameter, the first type model parameter includes a first-party first type model parameter and a second-party first type model parameter, and the second type model parameter includes a first-party second type model parameter and a second-party second type model parameter. For example, assuming that the first type model parameter is w and the second type model parameter is V, the first-party first type model parameter is w A , the second party first type model parameter is wB , the first party second type model parameter is V A , the second party second type model parameter is V B .

[0041] In addition, it should be noted that the process of secret sharing of data is a process of splitting the data into two sub-data, and the two sub-data are held by two parties of secret sharing respectively. For example, assuming that the two parties of secret sharing are A and B, then data X is secret shared, and A holds the first share of data X [[X]] A , B holds the second share of data X [[X]] B , and X=[[X]] A +[[X]] B .

[0042] In addition, it should be noted that the model expression of the factor decomposition machine regression model is as follows:

[0043] z(x)=<w,x> +∑ i<j <V i , V j >x i x j

[0044] Wherein, x is the data matrix corresponding to the model input data, wherein the model input data includes the first-party training label data (X A , Y) and the second-party training data X B , where Y is the sample label, X A With d A feature dimensions, X B With d B feature dimensions, the first type model parameter is w, where w is a d-dimensional vector, and the second type model parameter is V, where V is a d*d matrix, and w=[w A , w B ], that is, w is the first-party first-type model parameter w A and the second party first type model parameter w B Composition, among which w A is d A dimensional vector, w B is d B dimensional vector, otherwise, V = [V A , V B ], where V is the first-party second-type model parameter V A and the second party second type model parameter V B Composition, among which, V A is d A *d Xis a matrix, V B is d B *d X dimensional matrix,<w,x> is the inner product of w and x, V i is the column vector of the i-th column of V, V j is the column vector of the jth column of V, x i is the column vector of the i-th column of x, x j is the column vector of the j-th column of x.

[0045] The method comprises the steps of: performing secret sharing with a second device, obtaining secret sharing model parameters and secret sharing training data, specifically, obtaining an initialization model and first-party training label data corresponding to the factor decomposition machine regression model, and obtaining first-party first-type model parameters and corresponding first-party second-type model parameters corresponding to the initialization model. Similarly, the second device obtains the second-party training data, the second-party first-type model parameters and the corresponding second-party second-type model parameters corresponding to the initialization model before performing secret sharing, and then the first device and the second device perform secret sharing, wherein when performing secret sharing, the first device provides the first-party training label data, the first-party first-type model parameters and the first-party second-type model parameters, and the second device provides the second-party training data, the second-party first-type model parameters and the second-party second-type model parameters, and then the first device obtains the secret sharing. The second device obtains the second-party secret shared model parameters and secret shared training data owned by the second party, wherein the secret shared model parameters include the first share of the first-type model parameters of the first party, the first share of the second-type model parameters of the first party, the second share of the first-type model parameters of the second party, and the second share of the second-type model parameters of the second party, the second-party secret shared model parameters include the second share of the first-type model parameters of the first party, the second share of the second-type model parameters of the first party, the first share of the first-type model parameters of the second party, and the first share of the second-type model parameters of the second party, the secret shared training data include the first share of the first-party training label data and the second share of the second-party training data, and the second-party secret shared training data include the second share of the first-party training label data and the first share of the second-party training data.

[0046] The secret sharing model parameters include a first sharing parameter and a second sharing parameter, and the secret sharing training data include a first sharing training data and a second sharing training data.

[0047] The step of performing secret sharing with the second device and obtaining secret sharing model parameters and secret sharing training data comprises:

[0048] Step S11, obtaining a first-party model parameter and first-party training label data, and using a first share of the first-party model parameter as the first shared parameter;

[0049] In this embodiment, it should be noted that the first-party model parameters include first-party first-type model parameters and first-party second-type model parameters, and the second-party secret sharing model parameters include third sharing parameters and fourth sharing parameters.

[0050] Acquire first-party model parameters and first-party training label data, and use the first share of the first-party model parameters as the first shared parameters. Specifically, split the first-party first-type model parameters, the first-party second-type model parameters and the first-party training label data into two parts, and use the first share of the first-party first-type model parameters and the first share of the first-party second-type model parameters together as the first shared parameters.

[0051] Step S12, sending the second share of the first party model parameter to the second device, so that the second device can determine a third shared parameter;

[0052] In this embodiment, the second share of the first-party model parameters is sent to the second device so that the second device can determine the third shared parameters. Specifically, the second share of the first-party first-type model parameters and the second share of the first-party second-type model parameters are both sent to the second device, and then the second device uses the second share of the first-type model parameters and the second share of the first-party second-type model parameters together as the third shared parameters.

[0053] Step S13, receiving a second shared parameter sent by the second device, wherein the second shared parameter is a second share of a second-party model parameter acquired by the second device, and the first share of the second-party model parameter is a fourth shared parameter of the second device;

[0054] In this embodiment, a second shared parameter sent by the second device is received, wherein the second shared parameter is the second share of the second-party model parameter obtained by the second device, and the first share of the second-party model parameter is the fourth shared parameter of the second device. Specifically, the second device splits the second-party first-type model parameter and the second-party second-type model parameter into two parts respectively, and then uses the first share of the second-party first-type model parameter and the first share of the second-party second-type model parameter together as the fourth shared parameter, and sends the second share of the second-party first-type model parameter and the second share of the second-party second-type model parameter to the first device, and then the first device receives the second share of the second-party first-type model parameter and the second share of the second-party second-type model parameter, and uses the second share of the second-party first-type model parameter and the second share of the second-party second-type model parameter together as the second shared parameter.

[0055] Step S14, using the first share of the first-party training label data as the first shared training data, and sending the second share of the first-party training label data to the second device, so that the second device can determine the third shared training data;

[0056] In this embodiment, it should be noted that the second-party secret shared training data includes third shared training data and fourth shared training data.

[0057] The first share of the first-party training label data is used as the first shared training data, and the second share of the first-party training label data is sent to the second device, so that the second device determines the third shared training data. Specifically, the first-party training label data is split into two parts, and then the first share of the first-party training label data is used as the first shared training data, and the second share of the first-party training label data is sent to the second device, and then the second device uses the second share of the first-party training label data as the third shared training data.

[0058] Step S15: receiving second shared training data sent by the second device, wherein the second shared training data is a second share of second-party training data acquired by the second device, and the first share of the second-party training data is fourth shared training data of the second device.

[0059] In this embodiment, second shared training data sent by a second device is received, wherein the second shared training data is the second share of the second-party training data acquired by the second device, and the first share of the second-party training data is the fourth shared training data of the second device. Specifically, the second device splits the second-party training data into two parts, and uses the first share of the second-party training data as the fourth shared training data, sends the second share of the second-party training data to the first device, and the first device uses the first share of the second-party training data as the second shared training data.

[0060] Step S20, based on the secret sharing training data and the secret sharing model parameters, perform longitudinal federated learning modeling with the second device to calculate the secret sharing regression error;

[0061] In this embodiment, it should be noted that the secret shared training data includes first shared training data and second shared training data, wherein the first shared training data is the first share of the first party's training label data, and the second shared training data is the second share of the second party's training data, and the secret shared model parameters include first shared parameters and second shared parameters, wherein the first shared parameters include the first share of the first party's first type model parameters and then the first share of the first party's second type model parameters, and the second shared parameters include the second share of the second party's first type model parameters and the second share of the second party's second type model parameters.

[0062] Additionally, the second device provides second-party secret shared training data and second-party secret shared model parameters when performing vertical federated learning modeling, wherein the second-party secret shared training data includes third shared training data and fourth shared training data, wherein the third shared training data is the second share of the first-party training label data, and the fourth shared training data is the first share of the second-party training data, and the secret shared model parameters include third shared parameters and fourth shared parameters, wherein the third shared parameters include the second share of the first-party first-type model parameters and the second share of the first-party second-type model parameters, and the fourth shared parameters include the first share of the second-party first-type model parameters and the first share of the second-party second-type model parameters.

[0063] Based on the secret sharing training data and the secret sharing model parameters, longitudinal federated learning modeling is performed with the second device to calculate the secret sharing regression error. Specifically, based on the first shared parameters, the second shared model parameters, the first shared training data and the second shared training, a federal interaction is performed with the second device, wherein the second device provides a third shared parameter, a fourth shared model parameter, the third shared training data and the fourth shared training data when performing the federal interaction, calculates the secret sharing intermediate parameters, and then based on the secret sharing intermediate parameters, the secret sharing regression error is calculated by using a preset secret sharing regression error calculation formula.

[0064] The secret sharing model parameters include first type sharing parameters and second type sharing parameters, the secret sharing training data include secret sharing label data,

[0065] The step of performing longitudinal federated learning modeling with the second device based on the secret sharing training data and the secret sharing model parameters and calculating the secret sharing regression error comprises:

[0066] Step S21, based on a preset secret sharing mechanism, by performing federation interaction with the second device, calculating the inner product of the secret sharing cross-feature term corresponding to the second type of sharing parameter and the secret sharing training data;

[0067] In this embodiment, it should be noted that the preset secret sharing mechanism includes secret sharing addition and secret sharing multiplication, the first type of shared parameters include the first share of the first party's first type model parameters and the second share of the second party's first type model parameters, the second type of shared parameters include the first share of the first party's second type model parameters and the second share of the second party's second type model parameters, and the second device has the second party's first type shared parameters and the second party's second type shared parameters, wherein the second party's first type shared parameters include the second share of the first party's first type model parameters and the first share of the second party's first type model parameters, the second party's second type shared parameters include the second share of the first party's second type model parameters and the first share of the second party's second type model parameters, and the secret sharing label data is a sample label for secret sharing.

[0068] Based on the preset secret sharing mechanism, by performing federal interaction with the second device, the secret sharing cross-characteristic inner product corresponding to the second type of shared parameters and the secret sharing training data is calculated. Specifically, based on secret sharing multiplication, by performing federal interaction with the second device, the cross inner product between each parameter element in the second type of shared parameters and each training data element of the secret sharing training data is calculated, wherein there is a cross inner product between a parameter element and a training data element, and then each cross inner product is accumulated to obtain the secret sharing cross-characteristic inner product. In addition, when performing federal interaction, based on secret sharing multiplication, the second device will calculate the second-party cross inner product between each second-party parameter element in the second type of shared parameters of the second party and each second training data element in the second-party secret sharing training data to obtain the second-party secret sharing cross-characteristic inner product.

[0069] The second type of shared parameters includes a first shared second type model parameter and a second shared second type model parameter, the secret shared training data includes a first shared training data and a second shared training data, the secret shared cross-feature item inner product includes a first cross-feature item inner product and a second cross-feature item inner product, and the preset secret sharing mechanism includes a secret shared multiplication.

[0070] The step of calculating the inner product of the secret sharing cross-feature term corresponding to the second type of sharing parameter and the secret sharing training data based on the preset secret sharing mechanism by performing federated interaction with the second device comprises:

[0071] Step S211, based on the secret sharing multiplication, by performing federation interaction with the second device, calculating the cross inner product between each element in the first shared second type model parameter and each element in the first shared training data, to obtain the cross inner product of each first element;

[0072] In this embodiment, it should be noted that the first shared second-type model parameter may be a parameter in matrix form, wherein the first shared second-type model parameter is the second share of the second-party second-type model parameter, and each column of the first shared second-type model parameter is a first parameter element. The first shared training data may be training data in matrix form, wherein the first shared training data is the second share of the second-party training data, and each column of the first shared training data is a first training data element.

[0073] Based on the secret sharing multiplication, by performing federal interaction with the second device, the cross inner product between each element in the first shared second type model parameter and each element in the first shared training data is calculated to obtain each first element cross inner product. Specifically, a first secret sharing multiplication triplet corresponding to the secret sharing multiplication is obtained, and then based on the first secret sharing multiplication triplet, by performing federal interaction with the second device through secret sharing multiplication, the inner product between each first parameter element and each first training data element is calculated to obtain each first element cross inner product, wherein the second device calculates the second party first element cross inner product corresponding to each first element cross inner product when performing federal interaction with the first device.

[0074] In one feasible solution, it is assumed that the first device has a secret shared multiplication triple ([[a]] A ,[[b]] A , [[c]] A ), the second device possesses the secret shared multiplication triplet ([[a]] B ,[[b]] B , [[c]] B ), where [[a]] A +[[a]] B =a,[[b]] A +[[b]] B =b,[[c]] A +[[c]] B =c, c=a*b, and the first parameter element is the secret shared [[x]] A , the first training data element is [[y]] A , the parameter element corresponding to the first parameter element in the second device is [[x]] B , the training data element corresponding to the first training data element is [[y]] B , where [[x]] A +[[x]] B =x,[[y]] A +[[y]] B =y, then the cross inner product of the first element calculated by the first device is the secret shared [[x*y]] A , the cross inner product of the first element of the second side calculated by the second device is [[x*y]] B , and [[x*y]] A +[[x*y]] B =x*y, and then, specifically, the calculation process is as follows:

[0075] First, the first device calculates [[e]] A=[[x]] A -[[a]] A and [[f]] A =[[y]] A -[[b]] A , the second device calculates [[e]] B =[[x]] B -[[a]] B and [[f]] B =[[y]] B -[[b]] B , and then the first device will [[e]] A and [[f]] A Send to the second device, the second device will [[e]] B and [[f]] B Send to the second device, so that the first device and the second device both obtain e=xa and f=yb, and then the first device calculates [[x*y]] A =f*[[a]] A +e*[[b]] A +[[c]] A , the second device calculates [[x*y]] B =e*f+f*[[a]] B +e*[[b]] B +[[c]] B , and then [[x*y]] A +[[x*y]] B =e*f+f*a+e*b+c, and then substitute e=xa and f=yb into the calculation expression to obtain [[x*y]] A +[[x*y]] B =x*y, that is, the cross inner product of the first elements and the cross inner product of the second first elements are calculated.

[0076] Step S212: Based on the secret sharing multiplication, by performing federation interaction with the second device, calculate the cross inner product between each element in the second shared second type model parameter and each element in the second shared training data to obtain the cross inner product of each second element;

[0077] In this embodiment, it should be noted that the second shared second type model parameter may be a parameter in matrix form, wherein the second shared second type model parameter is the first share of the first party's second type model parameter, and each column of the second shared second type model parameter is a second parameter element, and the second shared training data may be training data in matrix form, wherein the second shared training data is the first share of the first party's training label data, and each column of the second shared training data is a second training data element.

[0078] Based on the secret sharing multiplication, by performing federal interaction with the second device, the cross inner product between each element in the second shared second type model parameter and each element in the second shared training data is calculated to obtain each cross inner product of the second elements. Specifically, a second secret sharing multiplication triplet corresponding to the secret sharing multiplication is obtained, and then based on the second secret sharing multiplication triplet, by performing federal interaction with the second device through secret sharing multiplication, the inner product between each second parameter element and each second training data element is calculated to obtain each cross inner product of the second elements.

[0079] Step S213, respectively accumulating the first element cross inner products and the second element cross inner products to obtain the first cross characteristic term inner products corresponding to the first element cross inner products and the second cross characteristic term inner products corresponding to the second element cross inner products.

[0080] In this embodiment, each of the first element cross inner products and each of the second element cross inner products are accumulated respectively to obtain the first cross feature item inner product corresponding to each of the first element cross inner products and the second cross feature item inner product corresponding to each of the second element cross inner products. Specifically, each of the first element cross inner products is accumulated to obtain the first cross feature item inner product, and each of the second element cross inner products is accumulated to obtain the second cross feature item inner product, wherein the calculation expression of the first cross feature item inner product is as follows:

[0081]

[0082] in, is a column vector of the first parameter element of the secret share owned by the first device, that is, the second share of the second type model parameter of the second party, The first training data element of the secret share owned by the first device is the column vector in the second share of the second party's training data. In addition, the calculation formula of the inner product of the second feature cross feature term is as follows:

[0083]

[0084] in, is a column vector of the second parameter element of the secret share owned by the first device, that is, the first share of the second type model parameter of the first party, The second training data element of the secret share owned by the first device is the column vector in the first share of the first-party training label data.

[0085] Additionally, the second device calculates the inner product of the second cross-feature term of the second party using the following formula:

[0086]

[0087] in, is a second party first parameter element of the secret share owned by the second device, that is, a column vector in the first share of the second party second type model parameter, The second training data element of the secret share owned by the second device is the column vector in the second share of the first training label data. In addition, the second device calculates the inner product of the second cross-feature term of the second party as follows:

[0088]

[0089] in, is a column vector of the second party second parameter element of the secret share owned by the second device, that is, the second share of the first party second type model parameter, The second party second training data element of the secret share possessed by the second device is also a column vector in the second share of the first party training label data.

[0090] Step S22, based on the preset secret sharing mechanism, by performing federated interaction with the second device, calculating the secret sharing intermediate parameter corresponding to the inner product of the secret sharing cross-feature term, the secret sharing training data, the first type of sharing parameter and the second type of sharing parameter;

[0091] In this embodiment, based on the preset secret sharing mechanism, by performing federal interaction with the second device, the secret sharing intermediate parameter corresponding to the inner product of the secret sharing cross-feature terms, the secret sharing training data, the first type of shared parameters and the second type of shared parameters is calculated. Specifically, based on the third secret sharing multiplication triplet corresponding to the secret sharing multiplication, by performing federal interaction with the second device, the first intermediate parameter item is calculated based on the first type of shared parameters and the secret sharing training data, and the second intermediate parameter item is calculated based on the second type of shared parameters, the inner product of the secret sharing cross-feature terms and the secret sharing training data, and then the sum of the first intermediate parameter item and the second intermediate parameter item is calculated to obtain the secret sharing intermediate parameter.

[0092] The preset secret sharing mechanism includes secret sharing multiplication and secret sharing addition.

[0093] The step of calculating the secret sharing intermediate parameter corresponding to the secret sharing cross-feature term inner product, the secret sharing training data, the first type of sharing parameter and the second type of sharing parameter by performing federated interaction with the second device based on the preset secret sharing mechanism comprises:

[0094] Step S221, based on the secret sharing multiplication, by performing federation interaction with the second device, calculating a first intermediate parameter item corresponding to the first type of shared parameter and the secret sharing training data;

[0095] In this embodiment, it should be noted that the first intermediate parameter item includes a first shared intermediate parameter item and a second shared intermediate parameter item, the first type of shared parameter includes a first shared first type model parameter and a second shared first type model parameter, wherein the first shared first type model parameter is the second share of the second party's first type model parameter, and the second shared first type model parameter is the first share of the first party's first type model parameter, and the secret shared training data includes first shared training data and second shared training data, wherein the first shared training data is the second share of the second party's training data, and the second shared training data is the first share of the first party's training data.

[0096] Based on the secret sharing multiplication, by performing federal interaction with the second device, a first intermediate parameter item corresponding to the first type of shared parameter and the secret shared training data is calculated. Specifically, based on the third secret sharing multiplication triplet, by performing federal interaction with the second device, the inner products of the first shared first type model parameter and each column vector of the first shared training data are respectively calculated to obtain each first intermediate parameter inner product, and each first intermediate parameter inner product is accumulated to obtain the first shared intermediate parameter item, and the inner products of the second shared first type model parameter and each column vector of the second shared training data are respectively calculated to obtain each second intermediate parameter inner product, and each second intermediate parameter inner product is accumulated to obtain the second shared intermediate parameter item, wherein the calculation expression of the first shared intermediate parameter item is as follows:

[0097]

[0098] Wherein, M1 is the first shared intermediate parameter item, d B Represents X B The feature dimension is d B ,[[w B ]]A The first shared first type model parameter for the secret sharing of the first device, that is, the second share of the first type model parameter for the second party, [[X B ]] A is an element in the first shared training data of the secret share of the first device, that is, a column vector of the second share of the second party's training data, X B is the first shared training data. Additionally, the calculation expression of the second shared intermediate parameter item is as follows:

[0099]

[0100] Wherein, M2 is the second shared intermediate parameter item, d A Represents X A The feature dimension is d A ,[[w A ]] A The second shared first type model parameter of the secret shared by the first device, that is, the first share of the first type model parameter of the first party, [[X A ]] A is an element in the second shared training data of the secret share of the first device, that is, a column vector of the first share of the first party training label data, X A The second shared training data.

[0101] In addition, if necessary, the second device will calculate the second party's first shared intermediate parameter item based on the first share of the second party's first type model parameters and the first share of the second party's training data, wherein the second party's first shared intermediate parameter item is calculated in the same manner as the first shared intermediate parameter item, and the second device will calculate the second party's second shared intermediate parameter item based on the first party's second share of the first type model parameters and the second share of the first party's training label data, wherein the second party's second shared intermediate parameter item is calculated in the same manner as the second shared intermediate parameter item.

[0102] Step S222, calculating, based on the secret sharing addition and the secret sharing multiplication, a second intermediate parameter item corresponding to the inner product of the secret sharing cross-feature item, the secret sharing training data and the second type of shared parameter;

[0103] In this embodiment, it should be noted that the second intermediate parameter item includes a third shared intermediate parameter item and a fourth shared intermediate parameter item.

[0104] Based on the secret sharing addition and the secret sharing multiplication, the inner product of the secret sharing cross-characteristic term, the second intermediate parameter term corresponding to the secret sharing training data and the second type of shared parameters is calculated. Specifically, a first transposed matrix corresponding to the first shared second type model parameter and a second transposed matrix corresponding to the first shared training data are obtained. Then, based on the secret sharing multiplication, the inner product of the first shared second type model parameter, the first transposed matrix, the first shared training data and the second transposed matrix is ​​calculated by federal interaction with the second device to obtain a first inner product term, and a third shared intermediate parameter term is calculated based on the first cross-characteristic term inner product and the first inner product term. Similarly, a third transposed matrix corresponding to the second shared second type model parameter and a fourth transposed matrix corresponding to the second shared training data are obtained. Then, based on the secret sharing multiplication, the inner product of the second shared second type model parameter, the third transposed matrix, the second shared training data and the fourth transposed matrix is ​​calculated by federal interaction with the second device to obtain a second inner product term, and a fourth shared intermediate parameter term is calculated based on the second cross-characteristic term inner product and the second inner product term, wherein the expression of the third shared intermediate parameter term is as follows:

[0105]

[0106] in,[[]] A represents the partial share of data owned by the first device after secret sharing, V B The second type of shared parameter for the second party, X B is the second-party training data, V B Column vector of V B With d x column vector, For X B A column vector of B With d B column vector, and in addition, the expression of the fourth shared intermediate parameter term is as follows:

[0107]

[0108] in,[[]] A represents the partial share of data owned by the first device after secret sharing, V A is the first-party second-type model parameter, X A For first-party training label data, V A Column vector of V A With d x column vector, For X AA column vector of A With d A A column vector.

[0109] In addition, it should be noted that the second device will calculate the second party third shared intermediate parameter item and the second party fourth shared intermediate parameter item based on the partial share of data owned by the second device after secret sharing, and the calculation method is consistent with the calculation method in the first device. The second party third shared intermediate parameter item and the second party fourth shared intermediate parameter item are as follows:

[0110]

[0111]

[0112] in,[[]] B Data representing the portion of the share owned by the second device after secret sharing.

[0113] Step S223: Calculate the secret sharing intermediate parameter based on the first intermediate parameter item and the second intermediate parameter item.

[0114] In this embodiment, the secret sharing intermediate parameters include a first secret sharing intermediate parameter and a second secret sharing intermediate parameter.

[0115] Based on the first intermediate parameter item and the second intermediate parameter item, the secret sharing intermediate parameter is calculated. Specifically, the sum of the first shared intermediate parameter item and the third shared intermediate parameter item is calculated to obtain a first secret sharing intermediate parameter, and the sum of the second shared intermediate parameter item and the fourth shared intermediate parameter item is calculated to obtain a second secret sharing intermediate parameter, wherein the calculation expression of the first secret sharing intermediate parameter is as follows:

[0116]

[0117] in, is the first secret sharing intermediate parameter. In addition, the calculation expression of the second secret sharing intermediate parameter is as follows:

[0118]

[0119] Among them, [[f(X A )]] is the first secret shared intermediate parameter. In addition, the second device will calculate the sum of the second party's first shared intermediate parameter item and the second party's third shared intermediate parameter item to obtain the second party's first secret shared intermediate parameter, and calculate the sum of the second party's second shared intermediate parameter item and the second party's fourth shared intermediate parameter item to obtain the second party's second secret shared intermediate parameter.

[0120] Step S23, substituting the secret sharing intermediate parameter and the secret sharing label data into a preset regression error calculation formula to calculate the secret sharing regression error.

[0121] In this embodiment, it should be noted that the secret shared label data is a portion of the sample labels owned by the first device after secret sharing, and the second device owns the second party secret shared label data.

[0122] Substituting the secret sharing intermediate parameter and the secret sharing label data into a preset regression error calculation formula to calculate the secret sharing regression error, specifically, substituting the first secret sharing intermediate parameter, the second secret sharing intermediate parameter and the secret sharing label data into a preset regression error calculation formula to calculate the secret sharing regression error, wherein the preset regression error calculation formula is as follows:

[0123]

[0124] Among them, Y is the sample label. Similarly, the second device substitutes the second party first secret sharing intermediate parameter, the second party second secret sharing intermediate parameter and the second party secret sharing label data into the preset regression error calculation formula to calculate the second party secret sharing regression error.

[0125] Step S30, based on the secret shared regression error, determine the first target regression model parameters, and assist the second device in determining the second target regression model parameters to construct a longitudinal federated factorization machine regression model.

[0126] In this embodiment, based on the secret shared regression error, the first target regression model parameters are determined, and the second device is assisted in determining the second target regression model parameters to construct a longitudinal federated factor decomposition machine regression model. Specifically, the calculation of the secret shared regression error is repeated to iteratively update the secret shared model parameters until the preset model training end condition is reached to obtain the first secret shared target parameters. Similarly, the second device will repeat the calculation of the second party secret shared regression error to iteratively update the second party secret shared model parameters until the preset model training end condition is reached to obtain the second secret shared target parameters, and then receive the second secret shared target parameters sent by the second device. The second shared first-party target parameter in the first secret shared target parameter is obtained, and the first shared first-party target parameter in the first secret shared target parameter is obtained, the sum of the first shared first-party target parameter and the second shared first-party target parameter is calculated, and the first target regression model parameter is obtained, and the second shared second-party target parameter in the first secret shared target parameter is sent to the second device, so that the second device can calculate the second shared second-party target parameter and the sum of the first shared second-party target parameter in the second secret shared target parameter to obtain the second target regression model parameter, that is, the first type model parameters and the second type model parameters after the model training are determined, and then the vertical federated factor decomposition machine regression model can be determined.

[0127] In addition, it should be noted that the longitudinal federated factor decomposition machine regression model includes a recommendation model for personalized recommendation. Compared with the existing longitudinal federated learning method, the present application does not require a homomorphic encryption and decryption process when constructing a longitudinal federated factor decomposition machine regression model based on longitudinal federated learning, thereby reducing the amount of computation during longitudinal federated learning modeling, thereby improving the computational efficiency when constructing the longitudinal federated factor decomposition machine regression model. Moreover, since the longitudinal federated factor decomposition machine regression model is constructed based on longitudinal federated learning modeling, the feature richness of the training samples when constructing the longitudinal federated factor decomposition machine regression model is higher, and thus the model performance of the longitudinal federated factor decomposition machine regression model will be better, and thus the personalized recommendation effect of the longitudinal federated factor decomposition machine regression model as a recommendation model will be better.

[0128] The present embodiment provides a factorization machine regression model construction method. Compared with the technical means of constructing a regression model based on an unencrypted two-party federated learning method or a homomorphically encrypted two-party longitudinal federated learning modeling method adopted in the prior art, the present embodiment obtains secret sharing model parameters and secret sharing training data by performing secret sharing with a second device, and then performs longitudinal federated learning modeling with the second device based on the secret sharing training data and the secret sharing model parameters, calculates the secret sharing regression error, and then updates the secret sharing model parameters based on the secret sharing regression error to obtain the secret sharing regression model update parameters, wherein, when interacting with the second device, the data sent or received are all secret sharing data, and there is no need for public-private key pairs generated by a third party. The data is encrypted with a key, and all data transmission processes are carried out between the two parties participating in the vertical federated learning modeling, which protects the privacy of the data. Then, the parameters of the secret sharing regression model are updated based on the secret sharing regression model. The first target regression model parameters can be determined by decrypting interaction with the second device, and the second device is assisted in determining the second target regression model parameters, so as to complete the construction of the vertical federated factor decomposition machine regression model. This overcomes the technical defect of the prior art that the regression model is constructed based on an unencrypted two-party federated learning method or a homomorphic encrypted two-party vertical federated learning modeling method, which makes it impossible to protect the data privacy of each participant in the vertical federated learning modeling. Therefore, the technical problem that the data privacy of each participant cannot be protected when the regression model is constructed based on the vertical federated learning modeling is solved.

[0129] Further, refer to Figure 2 Based on the first embodiment of the present application, in another embodiment of the present application, the step of determining the first target regression model parameters based on the secret shared regression error and assisting the second device in determining the second target regression model parameters includes:

[0130] Step S31, based on the secret sharing regression error, updating the secret sharing model parameters to obtain the secret sharing regression model update parameters;

[0131] In this embodiment, based on the secret shared regression error, the secret shared model parameters are updated to obtain the secret shared regression model update parameters. Specifically, based on the secret shared regression error, the model gradient information corresponding to the secret shared model parameters is calculated, and then based on the model gradient information, the secret shared model parameters are updated to obtain the secret shared regression model update parameters.

[0132] The secret sharing model parameters include a first secret sharing parameter and a second secret sharing model parameter, and the secret sharing regression model update parameters include a first shared regression model parameter and a second shared regression model parameter.

[0133] The step of updating the secret sharing model parameters based on the secret sharing regression error to obtain the updated parameters of the secret sharing regression model comprises:

[0134] Step S311, calculating first gradient information of the secret shared regression error with respect to the first secret shared model parameter, and calculating second gradient information of the secret shared regression error with respect to the first shared second type model parameter;

[0135] In this embodiment, it should be noted that the first secret sharing model parameter includes the first share of the first party's first type model parameter and the first share of the first party's second type model parameter, and the first gradient information includes the first type gradient and the second type gradient, wherein the first type gradient is the gradient of the secret sharing corresponding to the first share of the first party's first type model parameter, and the second type gradient is the gradient set of the secret sharing of each column vector in the first share of the first party's second type model parameter.

[0136] In addition, it should be noted that the second secret sharing model parameters include the second share of the first type model parameters of the second party and the second share of the second type model parameters of the second party, and the second gradient information includes a third type gradient and a fourth type gradient, wherein the third type gradient is the gradient of the secret sharing corresponding to the second share of the first type model parameters of the second party, and the fourth type gradient is the set of gradients of the secret sharing of each column vector in the second share of the second type model parameters of the second party.

[0137] The partial derivative of the secret shared regression error with respect to the first secret shared model parameter is calculated to obtain the first gradient information. Specifically, the partial derivative of the secret shared regression error with respect to the first share of the first type model parameter of the first party is calculated to obtain the first type gradient, and the partial derivative of each column vector in the first share of the second type model parameter of the first party is calculated to obtain the second type gradient, wherein the calculation expression of the first type gradient is as follows:

[0138]

[0139] Where T1 is the first type of gradient, α is a hyperparameter whose size can be set by yourself and is used to control the value range of the gradient, and w A is the first-party first-type model parameter, [[w A ]] A is the first share of the first-type model parameter of the first party. Additionally, the calculation expression of the second-type gradient is as follows:

[0140]

[0141] Where T2 is the second type of gradient, α is a hyperparameter whose size can be set by yourself and is used to control the value range of the gradient, V A is the first party second type model parameter, [[V A ]] A is a first share of the second type model parameter of the first party, is a column vector of the first share of the second type model parameters of the first party, further, the partial derivative of the secret shared regression error with respect to the second share of the first type model parameters of the second party is calculated to obtain the third type gradient, and the partial derivative of the secret shared regression error with respect to each column vector in the second share of the second type model parameters of the second party is calculated to obtain the fourth type gradient, wherein the calculation expression of the third type gradient is as follows:

[0142]

[0143] Wherein, T3 is the third type of gradient, α is a hyperparameter whose size can be set by oneself and is used to control the value range of the gradient, and w B is the first type model parameter of the second party, [[w B ]] A is the second share of the first type model parameter of the second party. Additionally, the calculation expression of the fourth type gradient is as follows:

[0144]

[0145] Wherein, T4 is the fourth type of gradient, α is a hyperparameter whose size can be set by oneself and is used to control the value range of the gradient, V B is the second type model parameter of the second party, [[V B ]] A a second share of the second type model parameter of the second party, A column vector of second shares of the second type model parameters of the second party.

[0146] In addition, it should be noted that the second device can also calculate the partial derivative of the second party's secret shared regression error with respect to the first share of the second party's first type model parameter to obtain the fifth type of gradient, and calculate the partial derivative of the second party's secret shared regression error with respect to each column vector in the first share of the second party's second type model parameter to obtain the sixth type of gradient, and then calculate the partial derivative of the second party's secret shared regression error with respect to the second share of the first party's first type model parameter to obtain the seventh type of gradient, and calculate the partial derivative of the second party's secret shared regression error with respect to each column vector in the second share of the first party's second type model parameter to obtain the eighth type of gradient, wherein the method of calculating the gradient in the second device is consistent with the method of calculating the gradient in the first device.

[0147] Step S312: based on the first gradient information and the preset first learning parameters, updating the first secret sharing model parameters until a preset federated learning end condition is met, thereby obtaining the first shared regression model parameters;

[0148] In this embodiment, it should be noted that the preset federated learning termination conditions include convergence of the loss function, reaching a preset iteration number threshold, etc., and the preset first learning parameters include a first learning rate and a second learning rate.

[0149] Based on the first gradient information and the preset first learning parameter, the first secret sharing model parameter is updated until the preset federated learning end condition is met, and the first shared regression model parameter is obtained. Specifically, the product of the first type of gradient and the first learning rate is calculated to obtain a first gradient descent value, and then the difference between the first share of the first type model parameter of the first party and the first gradient descent value is calculated to obtain a first update parameter, and the product of the second type of gradient and the second learning rate is calculated to obtain a second gradient descent value, and then the difference between the first share of the second type model parameter of the first party and the second gradient descent value is calculated to obtain a second update parameter, and then it is determined whether the first update parameter and the second update parameter meet the preset federated learning end condition. If so, the first update parameter and the second update parameter are used together as the first shared regression model parameter. If not, the gradient information is recalculated to iteratively update the first secret sharing model parameter until the preset federated learning end condition is met. The calculation expression for calculating the first update parameter is as follows:

[0150]

[0151] Among them, δ1 is the first learning rate, is the first update parameter. Additionally, the calculation expression for calculating the second update parameter is as follows:

[0152]

[0153] Among them, δ2 is the second learning rate, is the second update parameter.

[0154] Step S313: Based on the second gradient information and the preset second learning parameters, the second secret sharing model parameters are updated until the preset federated learning end condition is met, thereby obtaining the second shared regression model parameters.

[0155] In this embodiment, it should be noted that the preset second learning parameter includes a third learning rate and a fourth learning rate.

[0156] Based on the second gradient information and the preset second learning parameter, the second secret sharing model parameter is updated until the preset federated learning end condition is met, and the second shared regression model parameter is obtained. Specifically, the product of the third type gradient and the third learning rate is calculated to obtain a third gradient descent value, and then the difference between the second share of the first type model parameter of the second party and the third gradient descent value is calculated to obtain a third update parameter, and the product of the fourth type gradient and the fourth learning rate is calculated to obtain a fourth gradient descent value, and then the difference between the second share of the second type model parameter of the second party and the fourth gradient descent value is calculated to obtain a fourth update parameter, and then it is determined whether the third update parameter and the fourth update parameter meet the preset federated learning end condition. If so, the third update parameter and the fourth update parameter are used together as the second shared regression model parameter. If not, the gradient information is recalculated to iteratively update the second secret sharing model parameter until the preset federated learning end condition is met, wherein the calculation expression for calculating the third update parameter is as follows:

[0157]

[0158] Wherein, δ3 is the third learning rate, is the third update parameter. Additionally, the calculation expression for calculating the fourth update parameter is as follows:

[0159]

[0160] Wherein, δ4 is the fourth learning rate, is the fourth updated parameter.

[0161] In addition, it should be noted that the second device calculates the fifth update parameter based on the fifth type of gradient and the preset fifth learning rate Based on the sixth type of gradient and the preset sixth learning rate, calculate the sixth update parameter Based on the seventh type of gradient and the preset seventh learning rate, calculate the seventh update parameter Based on the eighth type of gradient and the preset eighth learning rate, calculate the eighth update parameter The second device calculates each gradient in the same manner as the first device.

[0162] Step S32: based on the secret shared regression model update parameters, the first target regression model parameters are determined by performing decryption interaction with the second device, so that the second device can determine the second target regression model parameters.

[0163] In this embodiment, based on the secret shared regression model update parameters, the first target regression model parameters are determined by performing decryption interaction with the second device, so that the second device can determine the second target regression model parameters. Specifically, the seventh update parameter and the eighth update parameter sent by the second device are received, and the first target regression model parameters are calculated based on the first update parameter, the second update parameter, the seventh update parameter and the eighth update parameter, and the third update parameter and the fourth update parameter are sent to the second device, so that the second device can calculate the second target regression model parameters based on the three update parameters, the fourth update parameter, the fifth update parameter and the sixth update parameter.

[0164] The secret shared regression model update parameters include a first share of the first party model update parameters and a second share of the second party model update parameters.

[0165] The step of updating the parameters based on the secret shared regression model and determining the first target regression model parameters by performing decryption interaction with the second device so that the second device can determine the second target regression model parameters comprises:

[0166] Step S321, receiving the second share of the first-party model update parameters determined by the second device based on the longitudinal federated learning modeling, and transmitting the second share of the second-party model update parameters to the second device, so that the second device can determine the second target regression model parameters based on the first share of the second-party model update parameters and the second share of the second-party model update parameters determined by the longitudinal federated learning modeling;

[0167] In this embodiment, it should be noted that the first share of the first-party model update parameters includes the first update parameter and the second update parameter, the second share of the second-party model update parameters includes the third update parameter and the fourth update parameter, the first share of the second-party model update parameters includes the fifth update parameter and the sixth update parameter, and the second share of the first-party model update parameters includes the seventh update parameter and the eighth update parameter.

[0168] Receive the second share of the first-party model update parameters determined by the second device based on the vertical federated learning modeling, and send the second share of the second-party model update parameters to the second device, so that the second device can determine the second target regression model parameters based on the first share of the second-party model update parameters and the second share of the second-party model update parameters determined by the second device based on the vertical federated learning modeling. Specifically, receive the seventh update parameter and the eighth update parameter sent by the second device, and send the third update parameter and the fourth update parameter to the second device, so that the second device can calculate the sum of the third update parameter and the fifth update parameter to obtain the second-party first-type model update parameters, calculate the sum of the fourth update parameter and the fifth update parameter to obtain the second-party second-type model update parameters, and use the second-party first-type model update parameters and the second-party second-type model update parameters together as the second target regression model parameters, wherein the calculation expression of the second-party first-type model update parameters is as follows:

[0169]

[0170] in, is the second party first type model update parameter. In addition, the calculation expression of the second party second type model update parameter is as follows:

[0171]

[0172] in, Update parameters for the second party second type model.

[0173] Step S322: Aggregate the first share of the first-party model update parameters and the second share of the first-party model update parameters to obtain the first target regression model parameters.

[0174] In this embodiment, the first share of the first-party model update parameter and the second share of the first-party model update parameter are aggregated to obtain the first target regression model parameter. Specifically, the sum of the first update parameter and the seventh update parameter is calculated to obtain the first-party first-type model update parameter, and the sum of the second update parameter and the eighth update parameter is calculated to obtain the first-party second-type model update parameter, and the first-party first-type model update parameter and the first-party second-type model update parameter are used together as the first target regression model parameter, wherein the calculation expression of the first-party first-type model update parameter is as follows:

[0175]

[0176] in, is the first party first type model update parameter, and in addition, the calculation expression of the first party second type model update parameter is as follows:

[0177]

[0178] in, Update parameters for the first-party second-type model.

[0179] This embodiment provides a method for updating model parameters of a vertical federated factor decomposition machine regression model based on a secret sharing regression error, that is, first, the first device updates the secret sharing model parameters based on the secret sharing regression error by calculating the gradient method to obtain the updated parameters of the secret sharing regression model of this round of iteration, and the second device simultaneously updates the second party secret sharing model parameters based on the second party secret sharing regression error to obtain the updated parameters of the second party secret sharing regression model of this round of iteration, until the preset federated learning end condition is reached, then based on the secret sharing mechanism, the first device and the second device perform decryption interaction, the first device updates the parameters based on the secret sharing regression model, assists the second device to update the parameters based on the second party secret sharing regression model, and determines the second target regression model parameters, and at the same time, the second device updates the parameters based on the second party secret sharing regression model, assists the first device to update the parameters based on the secret sharing regression model, and determines the first target regression model parameters, thereby completing the construction of the vertical federated factor decomposition machine regression model, thereby laying a foundation for overcoming the technical defect in the prior art that a regression model is constructed by using an unencrypted two-party federated learning method or a homomorphically encrypted two-party vertical federated learning modeling method, which results in the inability to protect the data privacy of each participant in the vertical federated learning modeling.

[0180] Further, refer to Figure 3 Based on the first and second embodiments of the present application, in another embodiment of the present application, the personalized recommendation method is applied to a first device, and the personalized recommendation method includes:

[0181] Step A10, performing secret sharing with the second device to obtain secret sharing to-be-recommended user data and secret sharing model parameters;

[0182] In this embodiment, it should be noted that the first device and the second device are both participants in vertical federated learning, and before secret sharing, the first device and the second device have trained a preset scoring model based on secret sharing and vertical federated learning, wherein the preset scoring model is a trained factor decomposition machine regression model, which is used to predict the score of the user's corresponding item, and then the model expression of the preset scoring model is as follows:

[0183] z(x)=<w,x> +∑ i<j <Vi , V j >x i x j

[0184] Among them, x is the model input data, w and V are model parameters, and z(x) is the model output, which is the user's rating of the item.

[0185] The method comprises the following steps: performing secret sharing with the second device to obtain secret shared user data to be recommended and secret shared model parameters, specifically, obtaining first-party scoring model parameters of a preset scoring model and first-party user data to be recommended, wherein the data to be recommended is associated data of the user to be recommended, for example, the hobbies of the user to be recommended and the historical scoring data of the user to be recommended on items, and at the same time, the second device obtains second-party scoring model parameters of the preset scoring model and second-party user data to be recommended, wherein, since the preset scoring model is constructed based on longitudinal federated learning, part of the model parameters of the preset scoring model held by the first device are first-party scoring model parameters, and part of the model parameters of the preset scoring model held by the second device are second-party scoring model parameters, and the The first-party user data to be recommended is the association data of the user to be recommended collected by the first device, and the second-party user data to be recommended is the association data of the user to be recommended to the item collected by the second device, wherein both the first-party user data to be recommended and the second-party user data to be recommended can be represented by vectors. For example, assuming that the first-party user data to be recommended is a vector (1, 0, 1, 0), wherein code 1 indicates that the user clicks on the corresponding item, and code 0 indicates that the user does not click on the corresponding item, then the vector (1, 0, 1, 0) indicates that the user clicks on item A and item C, but does not click on item B and item D. Furthermore, based on the first-party scoring model parameters and the first-party user data to be recommended, secret sharing is performed with the second device, wherein , the second device provides the second-party scoring model parameters and the second-party user data to be recommended in secret sharing, and then the first device obtains the secret sharing model parameters and the secret sharing user data to be recommended, and the second device obtains the second-party secret sharing model parameters and the second-party secret sharing user data to be recommended, wherein the secret sharing model parameters include the first shared first-party model parameters and the first shared second-party model parameters, the secret sharing user data to be recommended includes the first shared first-party user data to be recommended and the first shared second-party user data to be recommended, the second-party secret sharing model parameters include the second shared first-party model parameters and the second shared second-party model parameters, and the second-party secret sharing user data to be recommended includes the second shared first-party model parameters and the second shared second-party model parameters The first shared first-party model parameter is the first share of the first-party scoring model parameter, the second shared first-party model parameter is the second share of the first-party scoring model parameter, the first shared second-party model parameter is the first share of the second scoring model parameter, the second shared second-party model parameter is the second share of the second scoring model parameter, the first shared first-party user data to be recommended is the first share of the first-party user data to be recommended, the second shared first-party user data to be recommended is the second share of the first-party user data to be recommended, and the first shared second-party user data to be recommended is the first share of the second-party user data to be recommended,The second shared second-party user data to be recommended is a second share of the second-party user data to be recommended.

[0186] Step A20, inputting the secret sharing user data to be recommended into a preset scoring model, so as to score the recommended items corresponding to the secret sharing user data to be recommended based on the secret sharing model parameters, and obtaining a first secret sharing scoring result;

[0187] In this embodiment, the secret shared user data to be recommended is input into a preset scoring model, so as to score the recommended items corresponding to the secret shared user data to be recommended based on the secret shared model parameters, and obtain a first secret shared scoring result. Specifically, the first shared first-party user data to be recommended and the first shared second-party user data to be recommended are respectively input into the preset scoring model, so as to substitute the first shared first-party user data to be recommended and the first shared first-party model parameters into the model expression of the preset scoring model, calculate the first shared first-party score through secret shared multiplication, and substitute the first shared second-party user data to be recommended and the first shared first-party model parameters into the model expression of the preset scoring model, calculate the first shared first-party score through secret shared multiplication, and obtain the first secret shared first-party score by substituting the first shared second-party user data to be recommended and the first shared second-party model parameters into the model expression of the preset scoring model. The two-party model parameters are substituted into the model expression of the preset scoring model, the first shared second-party score is calculated by secret shared multiplication, and the first shared first-party score and the first shared second-party score are used as the first secret shared scoring result, wherein the first shared first-party score and the first shared second-party score are both model output values, and similarly, the second device will calculate the second shared first-party score by secret shared multiplication based on the second shared first-party user data to be recommended and the second shared first-party model parameters, and calculate the second shared second-party score by secret shared multiplication based on the second shared second-party user data to be recommended and the second shared second party model parameters.

[0188] Step A30: Based on the first secret sharing scoring result, perform a federation interaction with the second device to calculate a target score in conjunction with the second secret sharing scoring result determined by the second device.

[0189] In this embodiment, based on the first secret sharing scoring result, a federal interaction is performed with the second device to calculate a target score in combination with the second secret sharing scoring result determined by the second device. Specifically, based on the first secret sharing scoring result, a federal interaction is performed with the second device to aggregate the first secret sharing scoring result and the second secret sharing scoring result to obtain a target score, wherein it should be noted that the target score is the user's score for the recommended item calculated by a preset scoring model.

[0190] The first secret-sharing personalized recommendation result includes a first shared first-party score and a first shared second-party score, and the second secret-sharing personalized recommendation result includes a second shared first-party score and a second shared second-party score.

[0191] The step of performing federated interaction with the second device based on the first secret sharing scoring result to calculate the target score in conjunction with the second secret sharing scoring result determined by the second device comprises:

[0192] Step A31, receiving the second shared first-party score and the second shared second-party score sent by the second device;

[0193] Step A32, calculating a first-party score based on the first shared first-party score and the second shared first-party score;

[0194] In this embodiment, the first party score is calculated based on the first shared first party score and the second shared first party score. Specifically, the sum of the first shared first party score and the second shared first party score is calculated to obtain the first party score.

[0195] Step A33, calculating a second-party score based on the first shared second-party score and the second shared second-party score;

[0196] In this embodiment, the second party score is calculated based on the first shared second party score and the second shared second party score. Specifically, the sum of the first shared second party score and the second shared second party score is calculated to obtain the second party score.

[0197] Step A34, aggregating the first-party score and the second-party score to obtain the target score.

[0198] In this embodiment, the first-party score and the second-party score are aggregated to obtain the target score. Specifically, based on preset aggregation rules, the first-party score and the second-party score are aggregated to obtain the target score, wherein the preset aggregation rules include summation and weighted averaging, etc.

[0199] Step A40: generating a target recommendation list corresponding to the items to be recommended based on the target scores.

[0200] In this embodiment, based on the target score, a target recommendation list corresponding to the to-be-recommended item is generated. Specifically, the target score is repeatedly obtained to obtain target scores of different target users for the same to-be-recommended item, and then the target users are sorted based on the size of each target score to generate a recommended user list of the to-be-recommended item, and the recommended user list is used as the target recommendation list.

[0201] In another feasible scheme, based on the target score, a target recommendation list corresponding to the user to be recommended corresponding to the secret shared user data to be recommended is generated. Specifically, the target score is repeatedly obtained to obtain the target score of the same target user for different items to be recommended, and then the items to be recommended are sorted based on the size of each target score to generate a recommended item list for the user to be recommended, and the recommended item list is used as the target recommendation list.

[0202] This embodiment provides a method for personalized recommendation based on secret sharing and vertical federated learning, wherein secret sharing is performed with a second device to obtain secret shared user data to be recommended and secret sharing model parameters, and then the secret shared user data to be recommended is input into a preset scoring model to score the recommended items corresponding to the secret shared user data to be recommended based on the secret sharing model parameters to obtain a first secret sharing scoring result, and then based on the first secret sharing scoring result, a federated interaction is performed with the second device to calculate a target score in conjunction with a second secret sharing scoring result determined by the second device, and then based on the target score, a target recommendation list corresponding to the recommended items is generated. In the personalized recommendation process, when the first device and the second device interact, the data sent or received are all secret shared data, and there is no need to encrypt the data with a public-private key generated by a third party. All data transmission processes are performed between the two parties participating in the vertical federated learning, which protects the privacy of the data while reducing the complex encryption and decryption calculation process of the data. Moreover, since only simple mathematical operations are required during the secret sharing process and the decryption process corresponding to the secret sharing, the calculation complexity is reduced, thereby improving the calculation efficiency of the factor decomposition machine classification model for personalized recommendation.

[0203] Reference Figure 4 , Figure 4 It is a schematic diagram of the device structure of the hardware operating environment involved in the embodiment of the present application.

[0204] like Figure 4 As shown, the factor decomposition machine regression model building device may include: a processor 1001, such as a CPU, a memory 1005, and a communication bus 1002. The communication bus 1002 is used to realize the connection and communication between the processor 1001 and the memory 1005. The memory 1005 may be a high-speed RAM memory or a stable memory (non-volatile memory), such as a disk memory. The memory 1005 may optionally be a storage device independent of the aforementioned processor 1001.

[0205] Optionally, the factor decomposition machine regression model building device may also include a rectangular user interface, a network interface, a camera, an RF (Radio Frequency) circuit, a sensor, an audio circuit, a WiFi module, etc. The rectangular user interface may include a display screen (Display), an input submodule such as a keyboard (Keyboard), and the optional rectangular user interface may also include a standard wired interface and a wireless interface. The network interface may optionally include a standard wired interface and a wireless interface (such as a WI-FI interface).

[0206] Those skilled in the art will understand that Figure 4 The factor decomposition machine regression model building device structure shown in does not constitute a limitation of the factor decomposition machine regression model building device, and may include more or fewer components than shown in the figure, or a combination of certain components, or a different arrangement of components.

[0207] like Figure 4 As shown, the memory 1005 as a computer storage medium may include an operating system, a network communication module, and a factor decomposition machine regression model construction program. The operating system is a program that manages and controls the hardware and software resources of the factor decomposition machine regression model construction device, and supports the operation of the factor decomposition machine regression model construction program and other software and / or programs. The network communication module is used to realize the communication between the components inside the memory 1005, and to communicate with other hardware and software in the factor decomposition machine regression model construction system.

[0208] exist Figure 4 In the factor decomposition machine regression model building device shown, the processor 1001 is used to execute the factor decomposition machine regression model building program stored in the memory 1005 to implement the steps of any of the above-mentioned factor decomposition machine regression model building methods.

[0209] The specific implementation of the factor decomposition machine regression model construction device of the present application is basically the same as the various embodiments of the factor decomposition machine regression model construction method described above, and will not be repeated here.

[0210] Reference Figure 5 , Figure 5 It is a schematic diagram of the device structure of the hardware operating environment involved in the embodiment of the present application.

[0211] like Figure 5As shown, the personalized recommendation device may include: a processor 1001, such as a CPU, a memory 1005, and a communication bus 1002. The communication bus 1002 is used to realize the connection and communication between the processor 1001 and the memory 1005. The memory 1005 may be a high-speed RAM memory or a stable memory (non-volatile memory), such as a disk memory. The memory 1005 may also be a storage device independent of the aforementioned processor 1001.

[0212] Optionally, the personalized recommendation device may also include a rectangular user interface, a network interface, a camera, an RF (Radio Frequency) circuit, a sensor, an audio circuit, a WiFi module, etc. The rectangular user interface may include a display screen (Display), an input submodule such as a keyboard (Keyboard), and the optional rectangular user interface may also include a standard wired interface and a wireless interface. The network interface may optionally include a standard wired interface and a wireless interface (such as a WI-FI interface).

[0213] Those skilled in the art will understand that Figure 5 The personalized recommendation device structure shown in the figure does not constitute a limitation on the personalized recommendation device, and may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0214] like Figure 5 As shown, the memory 1005 as a computer storage medium may include an operating system, a network communication module, and a personalized recommendation program. The operating system is a program that manages and controls the hardware and software resources of the personalized recommendation device, and supports the operation of the personalized recommendation program and other software and / or programs. The network communication module is used to realize the communication between the components inside the memory 1005, and the communication with other hardware and software in the personalized recommendation system.

[0215] exist Figure 5 In the personalized recommendation device shown, the processor 1001 is used to execute the personalized recommendation program stored in the memory 1005 to implement the steps of any of the personalized recommendation methods described above.

[0216] The specific implementation of the personalized recommendation device of the present application is basically the same as the embodiments of the personalized recommendation method described above, and will not be repeated here.

[0217] The present application also provides a factor decomposition machine regression model construction device, which is applied to a factor decomposition machine regression model construction device, and the factor decomposition machine regression model construction device includes:

[0218] A secret sharing module, used to perform secret sharing with the second device and obtain secret sharing model parameters and secret sharing training data;

[0219] A vertical federation module, configured to perform vertical federated learning modeling with the second device based on the secret sharing training data and the secret sharing model parameters, and calculate a secret sharing regression error;

[0220] A determination module is used to determine the first target regression model parameters based on the secret shared regression error, and assist the second device in determining the second target regression model parameters to construct a longitudinal federated factor decomposition machine regression model.

[0221] Optionally, the vertical federation module includes:

[0222] A first calculation submodule, configured to calculate, based on a preset secret sharing mechanism, a secret sharing cross-feature inner product corresponding to the second type of sharing parameter and the secret sharing training data by performing federated interaction with the second device;

[0223] A second calculation submodule is used to calculate the secret sharing intermediate parameter corresponding to the secret sharing cross-feature term inner product, the secret sharing training data, the first type of sharing parameter and the second type of sharing parameter based on the preset secret sharing mechanism by performing federated interaction with the second device;

[0224] The third calculation submodule is used to substitute the secret sharing intermediate parameter and the secret sharing label data into a preset regression error calculation formula to calculate the secret sharing regression error.

[0225] Optionally, the first calculation submodule includes:

[0226] A first calculation unit, configured to calculate, based on the secret sharing multiplication, a cross inner product between each element in the first shared second type model parameter and each element in the first shared training data by performing federated interaction with the second device, to obtain a cross inner product of each first element;

[0227] A second calculation unit is configured to calculate, based on the secret sharing multiplication, a cross inner product between each element in the second type of shared parameter of the second party and each element in the second shared training data by performing federated interaction with the second device to obtain a cross inner product of each second element;

[0228] The third calculation unit is used to accumulate the first element cross inner products and the second element cross inner products respectively to obtain the first cross feature item inner products corresponding to the first element cross inner products and the second cross feature item inner products corresponding to the second element cross inner products.

[0229] Optionally, the second calculation submodule includes:

[0230] a fourth calculation unit, configured to calculate, based on the secret sharing multiplication, a first intermediate parameter item corresponding to the first type of shared parameter and the secret sharing training data by performing federated interaction with the second device;

[0231] a fifth calculation unit, configured to calculate, based on the secret sharing addition and the secret sharing multiplication, a second intermediate parameter item corresponding to the inner product of the secret sharing cross-feature item, the secret sharing training data, and the second type of shared parameter;

[0232] A sixth calculation unit is used to calculate the secret sharing intermediate parameter based on the first intermediate parameter item and the second intermediate parameter item.

[0233] Optionally, the secret sharing module includes:

[0234] an acquisition submodule, configured to acquire a first-party model parameter and first-party training label data, and use a first share of the first-party model parameter as the first shared parameter;

[0235] A first sending submodule, configured to send a second share of the first-party model parameter to the second device, so that the second device can determine a third shared parameter;

[0236] A first receiving submodule, configured to receive a second shared parameter sent by the second device, wherein the second shared parameter is a second share of a second-party model parameter acquired by the second device, and the first share of the second-party model parameter is a fourth shared parameter of the second device;

[0237] A second sending submodule, configured to use the first share of the first-party training label data as the first shared training data, and send the second share of the first-party training label data to the second device, so that the second device can determine third shared training data;

[0238] The second receiving submodule is used to receive second shared training data sent by the second device, wherein the second shared training data is the second share of the second-party training data acquired by the second device, and the first share of the second-party training data is the fourth shared training data of the second device.

[0239] Optionally, the determining module includes:

[0240] An updating submodule, used to update the secret sharing model parameters based on the secret sharing regression error to obtain the secret sharing regression model update parameters;

[0241] A decryption submodule is used to update parameters based on the secret sharing regression model, and determine the first target regression model parameters by performing decryption interaction with the second device, so that the second device can determine the second target regression model parameters.

[0242] Optionally, the updating submodule includes:

[0243] a seventh calculation unit, configured to calculate first gradient information of the secret shared regression error with respect to the first secret shared model parameter, and calculate second gradient information of the secret shared regression error with respect to the first shared second type model parameter;

[0244] A first updating unit, configured to update the first secret sharing model parameters based on the first gradient information and a preset first learning parameter until a preset federated learning end condition is met, thereby obtaining the first shared regression model parameters;

[0245] The second updating unit is used to update the second secret sharing model parameters based on the second gradient information and the preset second learning parameters until the preset federated learning end condition is met to obtain the second shared regression model parameters.

[0246] Optionally, the decryption submodule includes:

[0247] an assisting decryption unit, configured to receive a second share of first-party model update parameters determined by the second device based on longitudinal federated learning modeling, and transmit the second share of the second-party model update parameters to the second device, so that the second device can determine the second target regression model parameters based on the first share of second-party model update parameters and the second share of second-party model update parameters determined by longitudinal federated learning modeling;

[0248] An aggregation unit is used to aggregate the first share of the first-party model update parameters and the second share of the first-party model update parameters to obtain the first target regression model parameters.

[0249] The specific implementation of the factor decomposition machine regression model construction device of the present application is basically the same as the various embodiments of the factor decomposition machine regression model construction method mentioned above, and will not be repeated here.

[0250] The embodiment of the present application further provides a personalized recommendation device, which is applied to a personalized recommendation device, and includes:

[0251] A secret sharing module, used to perform secret sharing with the second device, and obtain secret sharing recommended user data and secret sharing model parameters;

[0252] A scoring module, configured to input the secret sharing user data to be recommended into a preset scoring model, so as to score the recommended items corresponding to the secret sharing user data to be recommended based on the secret sharing model parameters, and obtain a first secret sharing scoring result;

[0253] a calculation module, configured to perform a federation interaction with the second device based on the first secret sharing score result, so as to calculate a target score in conjunction with a second secret sharing score result determined by the second device;

[0254] A generation module is used to generate a target recommendation list corresponding to the to-be-recommended items based on the target scores.

[0255] Optionally, the calculation module includes:

[0256] a receiving unit, configured to receive the second shared first-party score and the second shared second-party score sent by the second device;

[0257] a first calculation unit, configured to calculate a first-party score based on the first shared first-party score and the second shared second-party score;

[0258] a second calculation unit, configured to calculate a second-party score based on the first shared second-party score and the second shared second-party score;

[0259] An aggregation unit is used to aggregate the first-party score and the second-party score to obtain the target score.

[0260] The specific implementation of the personalized recommendation device of the present application is basically the same as the embodiments of the personalized recommendation method described above, and will not be repeated here.

[0261] An embodiment of the present application provides a readable storage medium, and the readable storage medium stores one or more programs, and the one or more programs can also be executed by one or more processors to implement the steps of the factor decomposition machine regression model construction method described in any one of the above items.

[0262] The specific implementation of the readable storage medium of the present application is basically the same as the above-mentioned embodiments of the factor decomposition machine regression model construction method, and will not be repeated here.

[0263] An embodiment of the present application provides a readable storage medium, and the readable storage medium stores one or more programs, and the one or more programs can also be executed by one or more processors to implement the steps of any of the personalized recommendation methods described above.

[0264] The specific implementation of the readable storage medium of the present application is basically the same as the embodiments of the above-mentioned personalized recommendation method, and will not be repeated here.

[0265] The above are only preferred embodiments of the present application, and are not intended to limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent processing scope of the present application.

Claims

1. A factor decomposition machine regression model construction method, characterized in that: The factor decomposition machine regression model construction method is applied to a first device, and the factor decomposition machine regression model construction method includes: Perform secret sharing with the second device to obtain secret sharing model parameters and secret sharing training data; Based on the secret sharing training data and the secret sharing model parameters, perform longitudinal federated learning modeling with the second device to calculate the secret sharing regression error; The secret sharing model parameters include first type sharing parameters and second type sharing parameters, the secret sharing training data include secret sharing label data, wherein the secret sharing label data is a sample label of a portion of the share owned by the first device after secret sharing, The step of performing longitudinal federated learning modeling with the second device based on the secret sharing training data and the secret sharing model parameters and calculating the secret sharing regression error comprises: Based on a preset secret sharing mechanism, by performing federated interaction with the second device, calculating the inner product of the secret sharing cross-feature items corresponding to the second type of sharing parameter and the secret sharing training data; Based on the preset secret sharing mechanism, by performing federated interaction with the second device, calculating the secret sharing intermediate parameter corresponding to the inner product of the secret sharing cross-feature term, the secret sharing training data, the first type of sharing parameter, and the second type of sharing parameter; Substituting the secret sharing intermediate parameter and the secret sharing label data into a preset regression error calculation formula to calculate the secret sharing regression error; Based on the secret shared regression error, first target regression model parameters are determined, and the second device is assisted in determining second target regression model parameters to construct a longitudinal federated factorization machine regression model.

2. The method for constructing a factor decomposition machine regression model according to claim 1, characterized in that: The second type of shared parameters includes a first shared second type model parameter and a second shared second type model parameter, the secret shared training data includes a first shared training data and a second shared training data, the secret shared cross-feature item inner product includes a first cross-feature item inner product and a second cross-feature item inner product, the preset secret sharing mechanism includes a secret shared multiplication, The step of calculating the inner product of the secret sharing cross-feature term corresponding to the second type of sharing parameter and the secret sharing training data based on the preset secret sharing mechanism by performing federated interaction with the second device comprises: Based on the secret sharing multiplication, by performing federated interaction with the second device, a cross inner product between each element in the first shared second type model parameter and each element in the first shared training data is calculated to obtain a cross inner product of each first element; Based on the secret sharing multiplication, by performing federated interaction with the second device, a cross inner product between each element in the second shared second type model parameter and each element in the second shared training data is calculated to obtain a cross inner product of each second element; The first element cross inner products and the second element cross inner products are accumulated respectively to obtain the first cross characteristic term inner products corresponding to the first element cross inner products and the second cross characteristic term inner products corresponding to the second element cross inner products.

3. The method for constructing a factor decomposition machine regression model according to claim 1, wherein: The preset secret sharing mechanism includes secret sharing multiplication and secret sharing addition, The step of calculating the secret sharing intermediate parameter corresponding to the secret sharing cross-feature term inner product, the secret sharing training data, the first type of sharing parameter and the second type of sharing parameter by performing federated interaction with the second device based on the preset secret sharing mechanism comprises: Based on the secret sharing multiplication, calculating a first intermediate parameter item corresponding to the first type of shared parameter and the secret sharing training data by performing federated interaction with the second device; Based on the secret sharing addition and the secret sharing multiplication, calculating a second intermediate parameter item corresponding to the inner product of the secret sharing cross-feature item, the secret sharing training data and the second type of shared parameter; The secret sharing intermediate parameter is calculated based on the first intermediate parameter item and the second intermediate parameter item.

4. The factor decomposition machine regression model construction method according to claim 1, characterized in that: The secret sharing model parameters include a first sharing parameter and a second sharing parameter, and the secret sharing training data include a first sharing training data and a second sharing training data. The step of performing secret sharing with the second device and obtaining secret sharing model parameters and secret sharing training data comprises: Acquire a first-party model parameter and first-party training label data, and use a first share of the first-party model parameter as the first shared parameter; sending a second share of the first party model parameter to the second device for the second device to determine a third shared parameter; receiving a second shared parameter sent by the second device, wherein the second shared parameter is a second share of a second-party model parameter acquired by the second device, and the first share of the second-party model parameter is a fourth shared parameter of the second device; Using the first share of the first-party training label data as the first shared training data, and sending the second share of the first-party training label data to the second device so that the second device can determine third shared training data; Second shared training data sent by the second device is received, wherein the second shared training data is a second share of second-party training data acquired by the second device, and the first share of the second-party training data is fourth shared training data of the second device.

5. The method for constructing a factor decomposition machine regression model according to claim 1, characterized in that: The step of determining the first target regression model parameters based on the secret shared regression error and assisting the second device in determining the second target regression model parameters comprises: Based on the secret sharing regression error, updating the secret sharing model parameters to obtain the secret sharing regression model update parameters; Based on the secret shared regression model update parameters, the first target regression model parameters are determined by performing decryption interaction with the second device, so that the second device can determine the second target regression model parameters.

6. The method for constructing a factor decomposition machine regression model according to claim 5, characterized in that: The secret sharing model parameters include a first secret sharing model parameter and a second secret sharing model parameter, and the secret sharing regression model update parameters include a first shared regression model parameter and a second shared regression model parameter. The step of updating the secret sharing model parameters based on the secret sharing regression error to obtain the updated parameters of the secret sharing regression model comprises: Calculating first gradient information of the secret shared regression error with respect to the first secret shared model parameter, and calculating second gradient information of the secret shared regression error with respect to the first shared second type model parameter; Based on the first gradient information and the preset first learning parameters, the first secret sharing model parameters are updated until a preset federated learning end condition is met, thereby obtaining the first shared regression model parameters; Based on the second gradient information and the preset second learning parameters, the second secret sharing model parameters are updated until the preset federated learning end condition is met, and the second shared regression model parameters are obtained.

7. The method for constructing a factor decomposition machine regression model according to claim 5, characterized in that: The secret shared regression model update parameters include a first share of the first party model update parameters and a second share of the second party model update parameters. The step of updating the parameters based on the secret shared regression model and determining the first target regression model parameters by performing decryption interaction with the second device so that the second device can determine the second target regression model parameters comprises: Receive a second share of first-party model update parameters determined by the second device based on longitudinal federated learning modeling, and transmit the second share of the second-party model update parameters to the second device, so that the second device can determine the second target regression model parameters based on the first share of second-party model update parameters and the second share of second-party model update parameters determined by longitudinal federated learning modeling; Aggregate the first share of the first-party model update parameters and the second share of the first-party model update parameters to obtain the first target regression model parameters.

8. A personalized recommendation method, characterized in that: The personalized recommendation method is applied to the first device described in any one of claims 1 to 7, and the personalized recommendation method includes: Perform secret sharing with the second device to obtain secret sharing to-be-recommended user data and secret sharing model parameters; Inputting the secret sharing user data to be recommended into a preset scoring model to score the recommended items corresponding to the secret sharing user data to be recommended based on the secret sharing model parameters to obtain a first secret sharing scoring result; Based on the first secret sharing scoring result, perform a federation interaction with the second device to calculate a target score in conjunction with the second secret sharing scoring result determined by the second device; Based on the target scores, a target recommendation list corresponding to the items to be recommended is generated.

9. The personalized recommendation method according to claim 8, characterized in that: The first secret sharing score result includes a first shared first-party score and a first shared second-party score, and the second secret sharing score result includes a second shared first-party score and a second shared second-party score. The step of performing federated interaction with the second device based on the first secret sharing scoring result to calculate the target score in conjunction with the second secret sharing scoring result determined by the second device comprises: receiving the second shared first-party score and the second shared second-party score sent by the second device; calculating a first-party score based on the first shared first-party score and the second shared second-party score; calculating a second-party score based on the first shared second-party score and the second shared second-party score; The first-party score and the second-party score are aggregated to obtain the target score.

10. A factor decomposition machine regression model building device, characterized in that: The factor decomposition machine regression model construction device includes: a memory, a processor, and a program stored in the memory for implementing the factor decomposition machine regression model construction method. The memory is used to store a program for implementing a factorization machine regression model construction method; The processor is used to execute a program for implementing the factor decomposition machine regression model construction method to implement the steps of the factor decomposition machine regression model construction method as described in any one of claims 1 to 7.

11. A readable storage medium, characterized in that: The readable storage medium stores a program for implementing a factor decomposition machine regression model construction method, and the program for implementing a factor decomposition machine regression model construction method is executed by a processor to implement the steps of the factor decomposition machine regression model construction method as described in any one of claims 1 to 7.

12. A personalized recommendation device, characterized in that: The personalized recommendation device includes: a memory, a processor, and a program stored in the memory for implementing the personalized recommendation method. The memory is used to store a program for implementing a personalized recommendation method; The processor is used to execute a program for implementing the personalized recommendation method, so as to implement the steps of the personalized recommendation method as claimed in any one of claims 8 to 9.

13. A readable storage medium, characterized in that: The readable storage medium stores a program for implementing the personalized recommendation method, and the program for implementing the personalized recommendation method is executed by a processor to implement the steps of the personalized recommendation method as claimed in any one of claims 8 to 9.