Data risk prediction method, device, electronic device, medium and program product

Abstract

The present disclosure provides a data risk prediction method, which can be applied to the field of cloud computing technology. The method includes: obtaining a target data set, wherein the target data set contains at least one encrypted data; using at least one ciphertext model to perform risk prediction based on the at least one original data, and obtaining a risk feature set that retains the encrypted information, wherein each ciphertext model is a model for risk prediction of encrypted data obtained by converting a corresponding plaintext model for risk prediction of non-encrypted data; and based on the risk feature set, generating a risk prediction result that retains the encrypted information. The present disclosure also provides a data risk prediction device, an electronic device, a computer-readable storage medium, and a computer program product.

Description

Technical Field

[0001] The present disclosure relates to the technical field of cloud computing, and in particular to a data risk prediction method, device, electronic device, computer-readable storage medium, and computer program product. Background Art

[0002] In the financial field, some risk models unique to the financial industry, such as abnormal or illegal behavior screening models, transaction fraud models, anti-money laundering models, etc., are the result of a high level of professionalism, experience, and data analysis. Large financial institutions have the corresponding self-development capabilities. Small and medium-sized financial enterprises generally purchase such models from large financial institutions to enhance their own financial risk prevention. Summary of the invention

[0003] One aspect of the present disclosure provides a data risk prediction method, including: obtaining a target data set, wherein the target data set includes at least one encrypted data; using at least one ciphertext model to perform risk prediction on the at least one encrypted data to obtain a risk feature set that retains the encrypted information, wherein each ciphertext model is a model for risk prediction on encrypted data obtained by converting a corresponding plaintext model for risk prediction on non-encrypted data; and generating a risk prediction result that retains the encrypted information based on the risk feature set.

[0004] Optionally, the above-mentioned use of at least one ciphertext model to perform risk prediction based on the above-mentioned at least one original data includes: for any ciphertext model among the above-mentioned at least one ciphertext model, extracting at least one target data that needs to use the ciphertext model to perform risk characteristic prediction from the above-mentioned target data set; and inputting the above-mentioned at least one target data into the ciphertext model for risk prediction to obtain a corresponding risk characteristic value.

[0005] Optionally, any one of the at least one ciphertext model is converted by the following operations: determining a plaintext model corresponding to the ciphertext model, wherein the plaintext model is composed of at least one target basis function; converting each of the at least one target basis function into a corresponding homomorphic basis function to obtain at least one corresponding homomorphic basis function; and constructing the ciphertext model using the at least one homomorphic basis function.

[0006] Optionally, based on the above-mentioned risk feature set, a risk prediction result retaining encrypted information is generated, including: obtaining the weight of each risk feature value in the above-mentioned risk feature set, and generating the above-mentioned risk prediction result based on the obtained weight and the above-mentioned risk feature set.

[0007] Optionally, each of the at least one target basis function is converted into a corresponding homomorphic basis function using a preset conversion interface.

[0008] Optionally, it also includes: verifying the at least one encrypted data based on a preset calculation key; wherein, in response to the verification being passed, performing risk prediction on the at least one encrypted data using at least one ciphertext model.

[0009] Another aspect of the present disclosure provides a data risk prediction method, including: encrypting at least one original data contained in a first data set to obtain a second data set about the encrypted data; initiating a risk prediction request based on the second data set, so that the server responds to the risk prediction request, obtains and generates a risk prediction result corresponding to the encrypted information based on the second data set; obtains the risk prediction result; decrypts the risk prediction result to obtain the decrypted risk prediction result.

[0010] Another aspect of the present disclosure provides a data risk prediction device, including: a first acquisition module, used to acquire a target data set, wherein the target data set contains at least one original data, and the at least one original data is encrypted data; a risk prediction module, used to use at least one ciphertext model to perform risk prediction based on the at least one original data to obtain a risk feature set that retains encrypted information, wherein each ciphertext model is a model for risk prediction of encrypted data obtained by converting a corresponding plaintext model for risk prediction of non-encrypted data; and a generation module, used to generate a risk prediction result that retains encrypted information based on the risk feature set.

[0011] Another aspect of the present disclosure provides a data risk prediction device, including: an encryption module, used to encrypt at least one original data contained in a first data set to obtain a second data set about the encrypted data; a request module, used to initiate a risk prediction request based on the above-mentioned second data set, so that the server responds to the above-mentioned risk prediction request, obtains and generates a risk prediction result corresponding to the encrypted information based on the above-mentioned second data set; a second acquisition module, used to obtain the above-mentioned risk prediction result; and a decryption module, used to decrypt the above-mentioned risk prediction result to obtain the decrypted risk prediction result.

[0012] Another aspect of the present disclosure provides an electronic device, comprising: one or more processors; and a memory for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the method of an embodiment of the present disclosure.

[0013] Another aspect of the present disclosure provides a computer-readable storage medium storing computer-executable instructions, which are used to implement the method of the embodiment of the present disclosure when executed.

[0014] Another aspect of the present disclosure provides a computer program product, including a computer program, and the above-mentioned computer program implements the above-mentioned method of the embodiment of the present disclosure when executed by a processor. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] For a more complete understanding of the present disclosure and its advantages, reference will now be made to the following description taken in conjunction with the accompanying drawings, in which:

[0016] Figure 1 The system architecture of the data risk prediction method and device according to the embodiment of the present disclosure is schematically shown;

[0017] Figure 2 A flowchart of a data risk prediction method according to an embodiment of the present disclosure is schematically shown;

[0018] Figure 3 A flowchart of a data risk prediction method according to another embodiment of the present disclosure is schematically shown;

[0019] Figure 4 A flowchart of a data risk prediction method according to another embodiment of the present disclosure is schematically shown;

[0020] Figure 5 The schematic diagram shows a principle diagram of a data risk prediction method according to an embodiment of the present disclosure;

[0021] Figure 6 A block diagram of a data risk prediction device according to an embodiment of the present disclosure is schematically shown;

[0022] Figure 7 A block diagram schematically shows a data risk prediction device according to another embodiment of the present disclosure; and

[0023] Figure 8 The block diagrams schematically illustrate an electronic device, a computer-readable storage medium, and a computer program product according to embodiments of the present disclosure. DETAILED DESCRIPTION

[0024] Hereinafter, embodiments of the present disclosure will be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the present disclosure. In the following detailed description, for ease of explanation, many specific details are set forth to provide a comprehensive understanding of the embodiments of the present disclosure. However, it is apparent that one or more embodiments may also be implemented without these specific details. In addition, in the following description, descriptions of known structures and technologies are omitted to avoid unnecessary confusion of the concepts of the present disclosure.

[0025] The terms used herein are only for describing specific embodiments and are not intended to limit the present disclosure. The terms "include", "comprising", etc. used herein indicate the existence of the above-mentioned features, steps, operations and / or components, but do not exclude the existence or addition of one or more other features, steps, operations or components.

[0026] All terms (including technical and scientific terms) used herein have the meanings commonly understood by those skilled in the art, unless otherwise defined. It should be noted that the terms used herein should be interpreted as having a meaning consistent with the context of this specification, and should not be interpreted in an idealized or overly rigid manner.

[0027] When using expressions such as "at least one of A, B, and C, etc.", they should generally be interpreted according to the meaning of the expression commonly understood by those skilled in the art (for example, "a system having at least one of A, B, and C" should include but is not limited to a system having A alone, B alone, C alone, A and B, A and C, B and C, and / or A, B, C, etc.).

[0028] Some block diagrams and / or flow charts are shown in the accompanying drawings. It should be understood that some boxes or combinations thereof in the block diagrams and / or flow charts can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that these instructions can create a device for implementing the functions / operations described in these block diagrams and / or flow charts when executed by the processor. The technology of the present disclosure can be implemented in the form of hardware and / or software (including firmware, microcode, etc.). In addition, the technology of the present disclosure can take the form of a computer program product on a computer-readable storage medium storing instructions, which can be used by an instruction execution system or used in conjunction with an instruction execution system.

[0029] In the financial field, small and medium-sized financial enterprises generally purchase risk models from large financial institutions to improve their ability to prevent financial risks. However, such models need to be customized and developed according to the database type, data table structure, and calculation method of indicators on the client side. The delivery cycle is often long and difficult to deliver in a standardized form. In addition, the calculation of such models requires the transaction data of the customers of small and medium-sized financial enterprises, which is private data. Therefore, it is basically impossible for small and medium-sized financial enterprises to transmit large amounts of private data to large financial institutions.

[0030] At the same time, when deployed on the client side, such financial risk models contain transaction-related indicators in various dimensions. These indicators and their calculation methods are based on years of expert judgment, relevant field knowledge, experience, and a large amount of data accumulation by large financial institutions. They are very core content in the model and need special protection. These indicators and their calculation methods are often deployed on the client side in the form of SQL statements, which may lead to the leakage of intellectual property rights of financial risk models. Furthermore, when large financial institutions export such models to the outside world, they may also suffer losses due to model leakage.

[0031] The embodiment of the present disclosure provides a data risk prediction method and a data risk prediction device capable of applying the method. The method comprises obtaining a target data set, wherein the target data set contains at least one encrypted data; using at least one ciphertext model to perform risk prediction on the at least one original data to obtain a risk feature set that retains the encrypted information, wherein each ciphertext model is a model for risk prediction of encrypted data obtained by converting a corresponding plaintext model for risk prediction of non-encrypted data; and generating a risk prediction result that retains the encrypted information based on the risk feature set.

[0032] Figure 1 The system architecture of the data risk prediction method according to the embodiment of the present disclosure is schematically shown. It should be noted that: Figure 1 What is shown is merely an example of a system architecture to which the embodiments of the present disclosure can be applied, in order to help those skilled in the art understand the technical content of the present disclosure, but it does not mean that the embodiments of the present disclosure cannot be used in other devices, systems, environments or scenarios.

[0033] like Figure 1 As shown, the system architecture 100 according to this example may include a client 101, a network 102, and a server 103. The network 102 is used to provide a medium for a communication link between the client 101 and the server 103. The network 102 may include various connection types, such as wired and / or wireless communication links.

[0034] The client 101 can interact with the server 103 through the network 102 to transmit or receive data. The client can transmit data to the server; the server performs risk prediction based on the received data and sends the risk prediction result to the client; the client receives the risk prediction result.

[0035] A large financial institution can provide risk prediction services for multiple small and medium-sized financial institutions, and accordingly, the server 103 can establish connections with multiple clients 101. Alternatively, a small and medium-sized financial institution can select different large financial institutions to provide financial risk prediction services for it, and accordingly, a client 101 can establish connections with the servers 103 of different large financial institutions.

[0036] It should be understood that in the embodiment of the present disclosure, the server 103 may be a single server or a server cluster consisting of multiple servers.

[0037] Figure 2 The flowchart of the data risk prediction method according to the embodiment of the present disclosure is schematically shown. The method can be applied to the server side.

[0038] like Figure 2 As shown, the method 200 may include operations S210 to S230.

[0039] In operation S210, a target data set is acquired, wherein the target data set includes at least encrypted data.

[0040] For example, the original data may be private data encrypted by a homomorphic encryption algorithm, for example, the private data may be the account balance of a certain account.

[0041] In operation S220, risk prediction is performed based on the at least one encrypted data using at least one ciphertext model to obtain a risk feature set that retains the encrypted information, wherein each ciphertext model is a model for risk prediction of encrypted data obtained by converting a corresponding plaintext model for risk prediction of non-encrypted data.

[0042] For example, the ciphertext model can predict the risk of encrypted data and obtain the risk characteristic value that retains the encrypted information. The plaintext model can predict the risk of unencrypted private data. For example, it can predict the risk based on private data such as account balance and account transaction frequency. The ciphertext model is converted from the plaintext model. The ciphertext model can predict the risk of encrypted data. For example, the ciphertext model can predict the risk based on encrypted data such as encrypted account balance and encrypted account transaction frequency.

[0043] In operation S230, based on the risk feature set, a risk prediction result retaining the encrypted information is generated.

[0044] For example, the risk feature set includes at least one risk feature value, and a risk prediction result can be obtained by performing calculations based on the at least one risk feature value.

[0045] Through the embodiments of the present disclosure, the data acquired and processed by the server are all encrypted data, and the ciphertext models are all deployed on the server, which effectively protects the intellectual property rights of the models.

[0046] Figure 3 The flowchart of the data risk prediction method according to an embodiment of the present disclosure is schematically shown. The method can be applied to the server side.

[0047] like Figure 3 As shown, the data risk prediction method 300 includes obtaining a target data set. The implementation of this operation is similar to the implementation of the above-mentioned operation S210, and the technical effects achieved are also similar, which will not be repeated here.

[0048] In the data risk prediction method 300, at least one ciphertext model is used to perform risk prediction based on the at least one original data, including operations S321 to S323; based on the risk feature set, a risk prediction result retaining encrypted information is generated, including operations S331 to S332.

[0049] In operation S321, the at least one original data is verified based on a preset calculation key.

[0050] After the server receives the data, it can use the calculation key to verify the original data in the target data set before making risk predictions. For example, target data set A needs to be verified with a calculation key A, and target data set B needs to be verified with another calculation key B.

[0051] For example, the encrypted data in the target dataset A can be verified using the calculation key A. If the server has the authority to process the target dataset A, it has the calculation key A. The calculation key A can be included in the ciphertext model. When using the ciphertext model to perform risk prediction on at least one encrypted data, first verify whether the encrypted data originates from the target dataset A that needs to be risk predicted based on the calculation key A. Through the pre-set calculation key, it can be ensured that only the server with the corresponding calculation key can perform risk prediction on the target dataset, which further improves the security of the data and strengthens the privacy protection of the data.

[0052] In operation S322, in response to the verification being passed, for any one of the at least one ciphertext model, at least one target data that needs to be predicted for risk characteristics using the ciphertext model is extracted from the target data set.

[0053] For example, the original data set may include data such as bank card number and balance; Card_Num in the target data set represents the bank card number, N_AMT represents the "balance", and the encrypted data corresponding to Card_Num and N_AMT may be a string. For example, the original data set includes the original data corresponding to the "balance" of bank card A, and the original data corresponding to the "balance" of other bank cards. The corresponding target data set includes the encrypted data corresponding to the "balance" of bank card A, and the encrypted original data corresponding to the "balance" of other bank cards. When calculating the risk feature value, a ciphertext model Modell of the server needs to use the "balance" data, so the encrypted data corresponding to N_AMT can be extracted from the target data set.

[0054] In operation S323, the at least one target original data is input into the ciphertext model for risk prediction to obtain a corresponding risk feature value.

[0055] For example, when calculating the risk feature value, a ciphertext model Model1 of the server needs to use the "balance" data. After extracting the encrypted data corresponding to N_AMT from the target data set, the encrypted data can be input into the ciphertext model Model1 to calculate the risk feature value Model1_result.

[0056] In operation S331, the weight of each risk feature value in the risk feature set is obtained.

[0057] For example, the weights of the risk feature values ​​may be preset, for example, a weight is set for each feature value based on experience in risk prediction.

[0058] For example, the weight of the risk feature value Model1_result calculated by the ciphertext model Model1 is 0.7, and the weight of the risk feature Model2_result calculated by the ciphertext model Model2 is 0.3. Among them, the ciphertext model Model1 can be calculated based on "balance" and "transaction frequency", and the ciphertext model Model2 can be calculated based on data such as "account opening time".

[0059] In operation S332, the risk prediction result is generated based on the acquired weight and the risk feature set.

[0060] For example, the risk prediction result may be a weighted average or weighted sum of the risk feature values ​​in the risk feature set.

[0061] For example, the risk prediction result He_result = He_add(He_mul(0.7, Model1_result), He_mul(0.3, Model2_result)), He_add represents homomorphic addition operation, He_mul represents homomorphic multiplication operation, He_result, Model1_result, Model2_result are all ciphertext, and He is the abbreviation of Homomorphic Encryptfon. The risk prediction result He_result should be consistent with the risk prediction result Result calculated based on the plaintext model. The risk prediction result Result calculated based on the plaintext model = 0.7×Model1′_result+0.3×Model2′_result, Result, Model1′_result, Model2`_result are all plaintext. The risk prediction result obtained by weighted operation can effectively characterize financial risks.

[0062] As an optional embodiment of the present disclosure, any one of the above-mentioned at least one ciphertext model is converted by the following operations: determining a plaintext model corresponding to the ciphertext model, wherein the plaintext model is composed of at least one target basis function; converting each of the above-mentioned at least one target basis function into a corresponding homomorphic basis function to obtain the corresponding at least one homomorphic basis function; and constructing the ciphertext model using the above-mentioned at least one homomorphic basis function.

[0063] For example, the target basic function can be an addition function, a subtraction function, a multiplication function, a division function, a comparison function, a remainder function, or a combination of the above functions. For example, the addition function supports plain text addition operations, such as X`+Y, where X` and Y' are plain text numbers. Corresponding to the addition function, He_add represents a homomorphic addition operation and supports ciphertext operations. For example, He_add(X, Y) represents a homomorphic addition operation on X and Y, where X is a string encrypted by X` and Y is a string encrypted by Y`.

[0064] For example, the plain text model Model1` is:

[0065] Model1`_result=(N_AMT1`+N_AMT2`)×N_Frequency`;

[0066] Among them, Modell`_result is the risk characteristic value calculated according to the plaintext model Model1`, N_AMTl` and N_AMT2` are both plaintext of account balance, and N_Frequency` is the transaction frequency. Accordingly, the ciphertext model model1 converted from the plaintext model Model1` is:

[0067] Model1_result=He_mul(He_add(N_AMT1, N_AMT2), N_Frequency);

[0068] Among them, Modell_result is the risk characteristic value calculated according to the ciphertext model Model1, N_AMT1 and N_AMT2 respectively represent the ciphertext of the account balance, and N_Frequency is the transaction frequency.

[0069] It should be understood that the formulas of the plaintext model and the ciphertext model in the embodiments of the present disclosure may not be used for actual financial risk prediction, but are only for illustrating the conversion method between the ciphertext model and the plaintext model.

[0070] Through the embodiments of the present disclosure, a plaintext model can be converted into a ciphertext model, and then risk prediction can be performed based on the encrypted data in the target data set. Risk prediction can be achieved without decrypting the ciphertext, thereby reducing the risk of privacy data leakage.

[0071] As an optional embodiment of the present disclosure, when risk prediction is performed on the at least one encrypted data using at least one ciphertext model, the number of homomorphic encryption operations using a single ciphertext model does not exceed 40.

[0072] Further, as an optional embodiment of the present disclosure, each of the at least one target basis function is converted into a corresponding homomorphic basis function using a preset conversion interface.

[0073] Further, as an optional embodiment of the present disclosure, the ciphertext model is integrated into the mySQL database.

[0074] Through the embodiments of the present disclosure, the target basis function can be quickly converted into a homomorphic basis function using a preset conversion interface.

[0075] Figure 4 The flowchart of the data risk prediction method according to an embodiment of the present disclosure is schematically shown. The method can be applied to the client side.

[0076] like Figure 4 As shown, the data risk prediction method 400 includes operations S410 to S440.

[0077] In operation S410, at least one original data included in the first data set is encrypted to obtain a second data set related to the encrypted data.

[0078] For example, a homomorphic encryption algorithm may be used to encrypt at least one original data contained in a first data set to obtain a second data set related to the encrypted data.

[0079] In operation S420, a risk prediction request is initiated based on the second data set, so that the server responds to the risk prediction request, obtains and generates a risk prediction result corresponding to the encrypted information based on the second data set.

[0080] For example, the client sends a second data set about encrypted data to the server, and the server obtains the encrypted data and cannot directly obtain the private data.

[0081] In operation S430, the risk prediction result is obtained.

[0082] For example, the client can obtain the risk prediction result immediately after it is generated, or obtain the result from the server when needed.

[0083] In operation S440, the risk prediction result is decrypted to obtain a decrypted risk prediction result.

[0084] For example, the client uses a homomorphic encryption algorithm to decrypt the risk prediction result to obtain a decrypted risk prediction result. Based on the homomorphic encryption algorithm, the decrypted risk prediction result is basically consistent with the risk prediction result generated according to the first data set, for example, the first 15 significant digits of the two are the same.

[0085] Through the embodiments of the present disclosure, the client can achieve financial risk prediction without sending private data to the server, which effectively reduces the risk of client private data leakage.

[0086] Figure 5 The schematic diagram shows a principle diagram of a data risk prediction method according to an embodiment of the present disclosure.

[0087] like Figure 5 As shown, an encryption module 511 and a decryption module 512 are arranged on the client 510, and the encryption module 511 and the decryption module 512 can perform encryption or operation based on the homomorphic encryption algorithm. The encryption module 510 encrypts at least one original data in the first data set 501 to form a second data set 502 about the encrypted data. The client 510 initiates a risk prediction request based on the second data set 502, so that the server 502 responds to the risk prediction request, obtains and uses the second data set 502 as the target data set.

[0088] The server 520 is provided with a risk prediction module 521, which performs risk prediction based on the second data set 502. The risk prediction module 521 has at least one ciphertext model built in it, which can perform risk prediction on the at least one encrypted data to obtain a risk feature set that retains the encrypted information, wherein each ciphertext model is a model for risk prediction on encrypted data obtained by converting a corresponding plaintext model for risk prediction on non-encrypted data. The risk prediction model 521 generates a risk prediction result 503 that retains the encrypted information based on the risk feature set.

[0089] The client 510 may obtain the risk prediction result 503. The decryption module 512 of the client 510 may decrypt the risk prediction result 503 to obtain a decrypted result 504, that is, the decrypted risk prediction result.

[0090] Figure 6 A block diagram of a data risk prediction device according to an embodiment of the present disclosure is schematically shown.

[0091] like Figure 6 As shown, the data risk prediction device 600 includes a first acquisition module 610 , a risk prediction module 620 , and a generation module 630 .

[0092] A first acquisition module 610 is used to acquire a target data set, wherein the target data set includes at least one encrypted data;

[0093] The risk prediction module 620 is used to use at least one ciphertext model to perform risk prediction on the at least one encrypted data to obtain a risk feature set that retains the encrypted information, wherein each ciphertext model is a model for risk prediction of encrypted data obtained by converting the corresponding plaintext model for risk prediction of non-encrypted data.

[0094] The generation module 630 is used to generate a risk prediction result retaining encrypted information based on the above risk feature set.

[0095] As an optional embodiment of the present disclosure, the risk prediction module includes: an extraction unit, for any of the at least one ciphertext model, extracting at least one target data that needs to be predicted for risk characteristics using the ciphertext model from the target data set; and a calculation unit, for inputting the at least one target data into the ciphertext model for risk prediction to obtain a corresponding risk characteristic value.

[0096] As an optional embodiment of the present disclosure, a conversion module is further included, wherein the conversion module includes: a determination unit, which determines a plaintext model corresponding to the ciphertext model, wherein the plaintext model is composed of at least one target basis function. A conversion unit, which is used to convert each of the at least one target basis function into a corresponding homomorphic basis function to obtain the corresponding at least one homomorphic basis function. A construction unit, which is used to construct the ciphertext model using the at least one homomorphic basis function.

[0097] As an optional embodiment of the present disclosure, the generation module includes: a weight acquisition unit, used to acquire the weight of each risk feature value in the risk feature set, and a generation unit, used to generate the risk prediction result based on the acquired weight and the risk feature set.

[0098] As an optional embodiment of the present disclosure, the above-mentioned conversion unit is used to convert each of the above-mentioned at least one target basis function into a corresponding homomorphic basis function using a preset conversion interface.

[0099] As an optional embodiment of the present disclosure, a verification module is further included, wherein the verification module includes: a verification unit, configured to verify the at least one original data based on a preset calculation key; and a response unit, configured to perform risk prediction on the at least one original data using at least one ciphertext model in response to passing the verification.

[0100] Figure 7 A block diagram of a data risk prediction device according to another embodiment of the present disclosure is schematically shown.

[0101] like Figure 7 As shown, the data risk prediction device 700 includes an encryption module 710 , a request module 720 , a second acquisition module 730 , and a decryption module 740 .

[0102] The encryption module 710 is used to encrypt at least one original data included in the first data set to obtain a second data set related to the encrypted data.

[0103] The request module 720 is used to initiate a risk prediction request based on the above-mentioned second data set, so that the server responds to the above-mentioned risk prediction request, obtains and generates a corresponding risk prediction result with encrypted information retained based on the above-mentioned second data set.

[0104] The second acquisition module 730 is used to obtain the above risk prediction results.

[0105] The decryption module 740 is used to decrypt the above risk prediction results to obtain the decrypted risk prediction results.

[0106] It should be noted that the implementation method of the device part is similar to the implementation method of the method part, and the technical effects achieved are also similar, which will not be repeated here.

[0107] According to the modules and units of the embodiments of the present disclosure, any multiple or at least part of the functions of any multiple thereof can be implemented in one module. According to the modules and units of the embodiments of the present disclosure, any one or more can be split into multiple modules for implementation. According to the modules and units of the embodiments of the present disclosure, any one or more can be at least partially implemented as a hardware circuit, such as a field programmable gate array (FPGA), a programmable logic array (PLA), a system on a chip, a system on a substrate, a system on a package, an application-specific integrated circuit (ASIC), or can be implemented by hardware or firmware of any other reasonable way of integrating or packaging the circuit, or implemented in any one of the three implementation methods of software, hardware and firmware or in any appropriate combination of any of them. Alternatively, according to the modules and units of the embodiments of the present disclosure, one or more can be at least partially implemented as a computer program module, and when the computer program module is run, the corresponding function can be executed.

[0108] For example, any of the first acquisition module 610, the risk prediction module 620, the generation module 630, the encryption module 710, the request module 720, the second acquisition module 730, and the decryption module 740 can be combined in one module for implementation, or any of the modules can be split into multiple modules. Alternatively, at least part of the functions of one or more of these modules can be combined with at least part of the functions of other modules and implemented in one module. According to an embodiment of the present disclosure, at least one of the first acquisition module 610, the risk prediction module 620, the generation module 630, the encryption module 710, the request module 720, the second acquisition module 730, and the decryption module 740 can be at least partially implemented as a hardware circuit, such as a field programmable gate array (FPGA), a programmable logic array (PLA), a system on a chip, a system on a substrate, a system on a package, an application specific integrated circuit (ASIC), or can be implemented by hardware or firmware such as any other reasonable way of integrating or packaging the circuit, or by any one of the three implementation methods of software, hardware, and firmware, or by a suitable combination of any of them. Alternatively, at least one of the first acquisition module 610, the risk prediction module 620, and the generation module 630 may be at least partially implemented as a computer program module, and when the computer program module is executed, the corresponding function may be executed.

[0109] Figure 8 A block diagram of an electronic device according to an embodiment of the present disclosure is schematically shown. Figure 8The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0110] like Figure 8 As shown, the electronic device 800 includes a processor 810 and a computer-readable storage medium 820. The electronic device 800 can execute the method according to the embodiment of the present disclosure.

[0111] Specifically, the processor 810 may include, for example, a general-purpose microprocessor, an instruction set processor and / or a related chipset and / or a special-purpose microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 810 may also include an onboard memory for cache purposes. The processor 810 may be a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of the present disclosure.

[0112] The computer-readable storage medium 820 may be, for example, a non-volatile computer-readable storage medium, specific examples of which include but are not limited to: magnetic storage devices, such as magnetic tapes or hard disks (HDDs); optical storage devices, such as compact disks (CD-ROMs); memories, such as random access memories (RAMs) or flash memories; and the like.

[0113] The computer-readable storage medium 820 may include a computer program 821 , which may include code / computer-executable instructions that, when executed by the processor 810 , cause the processor 810 to perform a method according to an embodiment of the present disclosure or any variation thereof.

[0114] The computer program 821 may be configured to have, for example, a computer program code including a computer program module. For example, in an example embodiment, the code in the computer program 821 may include one or more program modules, for example, including 821A, module 821B, ... It should be noted that the division method and number of modules are not fixed, and those skilled in the art may use appropriate program modules or program module combinations according to actual conditions, and when these program module combinations are executed by the processor 810, the processor 810 may execute the method according to the embodiment of the present disclosure or any variation thereof.

[0115] According to an embodiment of the present disclosure, at least one of the first acquisition module 610, the risk prediction module 620, and the generation module 630 can be implemented as a reference. Figure 8 The computer program modules described herein, when executed by the processor 810, can implement the corresponding operations described above.

[0116] According to an embodiment of the present disclosure, at least one of the encryption module 710, the request module 720, the second acquisition module 730, and the decryption module 740 may be implemented as a reference. Figure 8The computer program modules described herein, when executed by the processor 810, can implement the corresponding operations described above.

[0117] The present disclosure also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments; or may exist independently without being assembled into the device / apparatus / system. The above computer-readable storage medium carries one or more programs, and when the above one or more programs are executed, the method according to the embodiment of the present disclosure is implemented.

[0118] The flow charts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each box in the flow chart or block diagram can represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flow chart, and the combination of the boxes in the block diagram or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0119] It will be appreciated by those skilled in the art that, although the present disclosure has been shown and described with reference to specific exemplary embodiments of the present disclosure, it will be appreciated by those skilled in the art that, without departing from the spirit and scope of the present disclosure as defined by the appended claims and their equivalents, various changes in form and detail may be made to the present disclosure. Therefore, the scope of the present disclosure should not be limited to the above-described embodiments, but should be determined not only by the appended claims, but also by the equivalents of the appended claims.

Claims

1. A data risk prediction method, comprising: Acquire a target data set, wherein the target data set includes at least one encrypted data; Using at least one ciphertext model, risk prediction is performed based on the at least one encrypted data to obtain a risk feature set retaining the encrypted information, wherein each ciphertext model is a model for risk prediction of encrypted data obtained by converting a corresponding plaintext model for risk prediction of non-encrypted data; and Based on the risk feature set, generating a risk prediction result with encrypted information retained; Wherein, any ciphertext model in the at least one ciphertext model is converted by the following operations: Determining a plaintext model corresponding to the ciphertext model, wherein the plaintext model is composed of at least one target basis function; Convert each of the at least one target basis function into a corresponding homomorphic basis function to obtain the corresponding at least one homomorphic basis function; The ciphertext model is constructed using the at least one homomorphic basis function.

2. The method according to claim 1, wherein: The using at least one ciphertext model to perform risk prediction based on the at least one encrypted data includes: For any one of the at least one ciphertext model, extracting at least one target data for which risk feature prediction needs to be performed using the ciphertext model from the target data set; and The at least one target data is input into the ciphertext model for risk prediction to obtain a corresponding risk characteristic value.

3. The method according to claim 1, wherein: Based on the risk feature set, a risk prediction result with encrypted information retained is generated, including: Obtaining the weight of each risk feature value in the risk feature set, and Based on the acquired weight and the risk feature set, the risk prediction result is generated.

4. The method according to claim 1, wherein: Each of the at least one target basis function is converted into a corresponding homomorphic basis function using a preset conversion interface.

5. The method according to any one of claims 1 to 4, further comprising: Verifying the at least one encrypted data based on a preset calculation key; In which, in response to the verification being passed, risk prediction is performed on the at least one encrypted data using at least one ciphertext model.

6. A data risk prediction method, comprising: Encrypting at least one original data contained in the first data set to obtain a second data set related to the encrypted data; Based on the second data set, a risk prediction request is initiated, so that the server responds to the risk prediction request, obtains and generates a risk prediction result corresponding to the retained encrypted information based on the second data set, wherein the risk prediction result is generated based on the method described in any one of claims 1 to 5; Obtaining the risk prediction result; The risk prediction result is decrypted to obtain a decrypted risk prediction result.

7. A data risk prediction device, comprising: A first acquisition module, used to acquire a target data set, wherein the target data set includes at least one encrypted data; A risk prediction module, configured to use at least one ciphertext model to perform risk prediction based on the at least one encrypted data to obtain a risk feature set retaining the encrypted information, wherein each ciphertext model is a model for risk prediction of encrypted data obtained by converting a corresponding plaintext model for risk prediction of non-encrypted data; A generating module, configured to generate a risk prediction result retaining encrypted information based on the risk feature set; and Conversion modules, including: A determination unit, which determines a plaintext model corresponding to the ciphertext model, wherein the plaintext model is composed of at least one target basis function; a conversion unit, configured to convert each of the at least one target basis function into a corresponding homomorphic basis function to obtain the corresponding at least one homomorphic basis function; A construction unit is used to construct the ciphertext model using the at least one homomorphic basis function.

8. A data risk prediction device, comprising: An encryption module, used for encrypting at least one original data included in the first data set to obtain a second data set related to the encrypted data; A request module, configured to initiate a risk prediction request based on the second data set, so that the server responds to the risk prediction request, obtains and generates a risk prediction result corresponding to the retained encrypted information based on the second data set, wherein the risk prediction result is generated based on the method described in any one of claims 1 to 5; A second acquisition module is used to obtain the risk prediction result; The decryption module is used to decrypt the risk prediction result to obtain the decrypted risk prediction result.

9. An electronic device, comprising: one or more processors; a memory for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 6.

10. A computer-readable storage medium storing computer-executable instructions, wherein the instructions are used to implement the method according to any one of claims 1 to 6 when executed.

11. A computer program product comprising a computer program, which, when executed by a processor, implements the method according to any one of claims 1 to 6.

Images