Data encryption method, cross-mechanism risk assessment method, device, equipment, medium and program product

Abstract

The invention provides a data encryption method, and relates to the field of privacy computing, the field of information security, the field of financial science and technology or other fields. The method comprises the steps of obtaining transaction information of a target user under the condition of obtaining authorization; m wavelet basis functions are called to perform wavelet transformation on the transaction information, any two wavelet basis functions are configured to adaptively process transaction data of different dimensions in the transaction information, and M is an integer greater than or equal to 2; obtaining a transaction feature set according to respective wavelet transformation results of the M wavelet basis functions; and calling a homomorphic encryption algorithm to perform homomorphic encryption on the transaction feature set to obtain transaction encrypted data. The invention further provides a cross-mechanism risk assessment method.

Description

Technical Field

[0001] The present disclosure relates to the field of privacy computing, information security, financial technology or other fields, and more specifically, to data encryption methods, cross-institutional risk assessment methods, devices, equipment, media and program products. Background Art

[0002] User risk assessment is an important scenario in the financial business field. Financial institutions can conduct comprehensive assessments based on user information, historical transaction data, asset status and other information to better understand the user's credit status, debt repayment ability, and repayment ability, thereby assessing whether the user is eligible for borrowing, etc. The data source for user risk assessment can be obtained by sharing data across financial institutions, and data encryption is used to transmit data across institutions during the sharing process.

[0003] In the process of implementing the disclosed invention, the inventors found that if the amount of encrypted data is too large, it will consume too much computing resources required for encryption, resulting in slow data transmission and low overall efficiency. Sharing a small amount of encrypted data across institutions will affect the accuracy of risk assessment. Therefore, how to balance the efficiency and security of cross-institutional data transmission and the accuracy of risk assessment is an urgent problem to be solved. Summary of the invention

[0004] In view of the above problems, the present disclosure provides a data encryption method, a cross-institutional risk assessment method, an apparatus, a device, a medium and a program product.

[0005] According to a first aspect of the present disclosure, a data encryption method is provided, comprising: obtaining transaction information of a target user under authorization; calling M wavelet basis functions to perform wavelet transform on the transaction information, wherein any two wavelet basis functions are configured to adapt to process transaction data of different dimensions in the transaction information, and M is an integer greater than or equal to 2; obtaining a transaction feature set according to the wavelet transform results of each of the M wavelet basis functions; and calling a homomorphic encryption algorithm to perform homomorphic encryption on the transaction feature set to obtain transaction encrypted data.

[0006] In some embodiments, the wavelet transform results of each of the M wavelet basis functions include multiple groups of wavelet coefficient data corresponding one-to-one to multiple dimensional transaction data; obtaining the transaction feature set based on the wavelet transform results of each of the M wavelet basis functions includes: calculating the transaction feature set based on the multiple groups of wavelet coefficient data.

[0007] In some embodiments, the M wavelet basis functions include a first wavelet basis function, and calling the M wavelet basis functions to perform a wavelet transform on the transaction information includes: calling the first wavelet basis function to perform a wavelet transform on the transaction data of the consumption habit dimension to obtain a set of wavelet coefficient data corresponding to the consumption habit dimension; wherein the transaction data of the consumption habit dimension includes a variety of consumption data including transaction amount, credit points, transaction time, transaction location and transaction currency, and the set of wavelet coefficient data corresponding to the consumption habit dimension includes a plurality of first wavelet coefficient data corresponding one-to-one to the plurality of consumption data.

[0008] In some embodiments, the transaction feature set calculated based on the multiple groups of wavelet coefficient data includes: obtaining a sum result by adding the absolute values ​​of all wavelet coefficients in each of the first wavelet coefficient data; assigning a weight coefficient to each of the sum results to perform a weighted calculation on each of the sum results; and summing each of the weighted sum results to obtain the first transaction feature in the transaction feature set.

[0009] In some embodiments, the assigning of a weight coefficient to each of the summed results comprises: calculating the information entropy of each of the summed results; and assigning to each of the summed results a weight coefficient that is positively correlated with its information entropy.

[0010] In some embodiments, calculating the information entropy of each of the summed results includes: obtaining the information entropy of the summed result according to the proportion of each of the summed results in the total summed result.

[0011] In some embodiments, the M wavelet basis functions include a second wavelet basis function, and calling the M wavelet basis functions to perform a wavelet transform on the transaction information includes: calling the second wavelet basis function to perform a wavelet transform on the transaction data in the transaction frequency fluctuation dimension to obtain a set of wavelet coefficient data corresponding to the transaction frequency fluctuation dimension; wherein the transaction data in the transaction frequency fluctuation dimension includes a plurality of transaction data sets obtained from a plurality of transaction cycles, and the set of wavelet coefficient data corresponding to the transaction frequency fluctuation dimension includes a plurality of second wavelet coefficient data corresponding one-to-one to the plurality of transaction data sets.

[0012] In some embodiments, each of the second wavelet coefficient data includes a wavelet coefficient matrix and transaction frequency data, and the transaction feature set calculated based on the multiple sets of wavelet coefficient data includes: forming a matrix variable based on all the wavelet coefficient matrices and transaction frequency data; calling the frequent pattern growth algorithm to perform spatiotemporal correlation mining on the matrix variables to obtain the second transaction feature in the transaction feature set.

[0013] Another aspect of an embodiment of the present disclosure provides a cross-institutional risk assessment method, comprising: obtaining S encrypted transaction data of a target user in S institutions, wherein the encrypted transaction data of each institution is obtained by homomorphic encryption according to the data encryption method as described in any of the above items, and S is an integer greater than or equal to 2; inputting the S encrypted transaction data into a risk assessment model to obtain a risk assessment result.

[0014] In some embodiments, the S institutions include a lead institution, and obtaining the S transaction encryption data of the S institutions includes: transmitting the transaction encryption data of each of the S-1 institutions to the lead institution; wherein, during the transmission process, key exchange and signature verification are performed according to the elliptic curve digital signature algorithm.

[0015] Another aspect of an embodiment of the present disclosure provides a data encryption device, including: a first acquisition module, used to obtain transaction information of a target user under authorization; a first calling module, used to call M wavelet basis functions to perform wavelet transform on the transaction information, wherein any two wavelet basis functions are configured to adapt to process transaction data of different dimensions in the transaction information, and M is an integer greater than or equal to 2; a feature calculation module, used to obtain a transaction feature set based on the wavelet transform results of each of the M wavelet basis functions; and a homomorphic encryption module, used to call a homomorphic encryption algorithm to perform homomorphic encryption on the transaction feature set to obtain transaction encrypted data.

[0016] Another aspect of an embodiment of the present disclosure provides a cross-institutional risk assessment device, including: a second acquisition module, used to obtain S encrypted transaction data of a target user in S institutions, wherein the encrypted transaction data of each institution is obtained by homomorphic encryption according to the data encryption method as described in any of the above items, and S is an integer greater than or equal to 2; a risk assessment module, used to input the S encrypted transaction data into a risk assessment model to obtain a risk assessment result.

[0017] Another aspect of an embodiment of the present disclosure provides an electronic device, including: 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 execute the method described above.

[0018] Another aspect of an embodiment of the present disclosure provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the processor executes the method as described above.

[0019] Another aspect of an embodiment of the present disclosure provides a computer program product, including a computer program, which implements the above method when executed by a processor.

[0020] One or more of the above embodiments have the following beneficial effects:

[0021] 1) Using wavelet transform to process the target user's transaction information can capture the complex feature changes in massive transaction data, reduce data dimensions, and improve the efficiency of subsequent homomorphic encryption. Among them, using different wavelet basis functions to adapt and process transaction data of different dimensions in transaction information can more accurately extract relevant features, so that the encrypted transaction data can reflect the characteristics of the target user. Therefore, without leaking user privacy, it can achieve the effects of reducing data transmission and encryption calculation, improving transmission efficiency, and obtaining more accurate transaction features.

[0022] 2) The use of wavelet transform and homomorphic encryption technology can effectively reduce the encrypted data dimension, reduce the amount of data transmission, and improve the efficiency of data transmission and homomorphic encryption. Under the premise of not leaking user privacy, cross-institutional multi-party secure computing can be achieved. On the basis of ensuring the security of data transmission, the participating institutions can complete the cross-institutional user risk assessment without sharing the original data, which improves the security and convenience of cross-institutional risk assessment. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The above contents and other objects, features and advantages of the present disclosure will become more apparent through the following description of the embodiments of the present disclosure with reference to the accompanying drawings, in which:

[0024] Figure 1 A diagram schematically shows an application scenario of a data encryption method or a cross-institutional risk assessment method according to an embodiment of the present disclosure;

[0025] Figure 2 The flowchart of the data encryption method according to the embodiment of the present disclosure is schematically shown;

[0026] Figure 3 A flowchart of obtaining a first transaction feature according to an embodiment of the present disclosure is schematically shown;

[0027] Figure 4 A flowchart of obtaining a second transaction feature according to an embodiment of the present disclosure is schematically shown;

[0028] Figure 5 A flowchart of a cross-institutional risk assessment method according to another embodiment of the present disclosure is schematically shown;

[0029] Figure 6 A flowchart of establishing a secure transmission channel according to an embodiment of the present disclosure is schematically shown;

[0030] Figure 7 A flowchart of multi-party secure computing according to an embodiment of the present disclosure is schematically shown;

[0031] Figure 8 The structure block diagram of the data encryption device according to the embodiment of the present disclosure is schematically shown;

[0032] Fig. 9 A structural block diagram of a data encryption device according to an embodiment of the present disclosure is schematically shown; and

[0033] Fig.10 A block diagram of an electronic device suitable for implementing a data encryption method or a cross-institutional risk assessment method according to an embodiment of the present disclosure is schematically shown. DETAILED DESCRIPTION

[0034] 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.

[0035] In the technical solution of the present disclosure, the user information (including but not limited to user personal information, user transaction information, user image information, user device information, such as location information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved are all information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of the relevant data comply with relevant laws, regulations and standards, take necessary confidentiality measures, do not violate public order and good morals, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0036] In the scenario of using personal information for automated decision-making, the methods, devices, and systems provided by the embodiments of the present disclosure provide users with corresponding operation portals for users to choose to agree or reject the automated decision-making results; if the user chooses to reject, the expert decision-making process will be entered. The expression "automated decision-making" here refers to the activity of automatically analyzing and evaluating a person's behavioral habits, interests and hobbies, or economic, health, credit status, etc. through computer programs, and making decisions. The expression "expert decision-making" here refers to the activity of making decisions by people who specialize in a certain field, have specialized experience, knowledge and skills, and have reached a certain level of professionalism.

[0037] Figure 1 The following schematically shows an application scenario diagram of the data encryption method or the cross-institutional risk assessment method according to an embodiment of the present disclosure. It should be noted that: Figure 1What is shown are merely examples to which the embodiments of the present disclosure can be applied, so as 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.

[0038] like Figure 1 As shown, the application scenario 100 according to this embodiment may include a first server 101, a second server 102, a third server 103, a fourth server 104 and a network 105. The network 105 is used to provide a medium for communication links between the servers. The network 105 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0039] According to an embodiment of the present disclosure, the first server 101, the second server 102, the third server 103, and the fourth server 104 can form a centralized network or a point-to-point network (point to point, p2p) for communication. For example, communication in a decentralized form can make each institution relatively independent and improve the overall high availability. The first server 101, the second server 102, the third server 103, and the fourth server 104 can be servers of four institutions in the multi-party secure computing technology, such as financial institution 1, financial institution 2, financial institution 3, and financial institution 4. Each institution can act as a query party to obtain data from other institutions. Each participant participates in multi-party secure computing as a data source that provides local data. Local data can cover user behavior data, credit data, transaction data, etc. to comprehensively portray user portraits, and can use obfuscation encapsulation, homomorphic encryption technology, public-private key pairs and other methods to flexibly use to protect user privacy.

[0040] The staff can use the terminal device to interact with the first server 101, the second server 102, the third server 103, and the fourth server 104 to receive or send messages, etc. Various communication client applications can be installed on the terminal device, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social platform software, etc. (only as examples).

[0041] The terminal device may be any electronic device having a display screen and supporting web browsing, including but not limited to a smart phone, a tablet computer, a laptop computer, a desktop computer, and the like.

[0042] The first server 101, the second server 102, the third server 103, and the fourth server 104 may be servers that provide various services, such as a backend management server that provides support for websites browsed by staff using terminal devices (only as an example). The backend management server may analyze and process the received data such as staff requests, and feed back the processing results (such as web pages, information, or data obtained or generated according to staff requests) to the terminal device.

[0043] It should be understood that Figure 1 The number of networks and servers in the embodiment is only for illustration. Any number of networks and servers may be provided as required.

[0044] The following will be based on Figure 1 The described scenario provides a detailed description of the data encryption method and cross-institutional risk assessment method of the embodiments of the present disclosure.

[0045] It should be noted that the transaction encrypted data obtained using the data encryption method provided in some embodiments of the present disclosure is not limited to use in cross-institutional risk assessment scenarios, for example, it can also be used in product recommendations, data storage, user level classification and other scenarios.

[0046] Figure 2 The flowchart of the data encryption method according to the embodiment of the present disclosure is schematically shown.

[0047] like Figure 2 As shown, this embodiment includes:

[0048] In operation S210, the transaction information of the target user is obtained after obtaining authorization;

[0049] For example, transaction information includes user basic information, basic risk assessment information, personal asset status, credit score, historical transaction records, etc. For example, user basic information may include user name, user ID, certificate number, age, occupation, currency, monthly income, total assets, total liabilities, total loan amount, repayment date, repayment period (days), number of overdue repayments in the past year, cumulative number of overdue days in the past year, credit rating, purpose of borrowing, debt-to-asset ratio, number of credit record inquiries, related loan status, guarantor information, collateral value, etc. Historical transaction records may be transaction records of the target user in the past year, and may include fields such as transaction serial number, transaction date, transaction amount, currency, repayment date, status, credit score, transaction type, and payment channel, as shown in Table 1.

[0050] Table 1

[0051]

[0052] After the user authorizes to obtain his transaction information, the obtained transaction information can be cleaned and anonymized to prevent it from being directly associated with a specific user, thereby protecting the user's privacy. Specifically, the obtained transaction information can be converted to a unified format, processed with abnormal data, processed with privacy protection and unique identification, and processed with data generalization. The unified format conversion operation can convert the data into a unified format and unit, including date, time, currency, etc., to ensure data consistency and comparability. The abnormal data processing operation can remove invalid data such as garbled characters or incomplete key information records, and transaction cancellations to ensure data quality. The data generalization processing operation can generalize the data according to the common partition types of user risk assessment agreed by financial institutions, so as to better perform classification analysis for user general feature extraction and risk prediction. For example, if a user's real age is 28 years old, the age is generalized to a specific assessment age group (25~30); if a user has overdue credit card repayments in the past year at the credit institution A, the institution A can make a judgment based on the cumulative number of overdue payments, the length of overdue time, the amount, and the overall situation of the borrower, and provide the user's credit rating and whether to be included in the blacklist and other information references. Privacy protection and unique identification processing operations can encrypt / hash sensitive information such as name, user ID, certificate number, guarantor information, etc. involved in the data set; for example, the SHA-256 hash value is calculated for the certificate number, and the generated hash value result is used as a pseudo-identifier instead of a direct identifier to achieve user data privacy protection and ensure that the data set does not contain information that can directly or indirectly identify the user's identity.

[0053] In operation S220, M wavelet basis functions are called to perform wavelet transform on the transaction information, wherein any two wavelet basis functions are configured to adapt to process transaction data of different dimensions in the transaction information, and M is an integer greater than or equal to 2;

[0054] Among them, the wavelet basis function is the basic function used for wavelet transform, which can decompose the signal into components of different frequencies.

[0055] In operation S220, the transaction information after cleaning and anonymization can continue to be subjected to standardization, time feature extraction, classification feature encoding and other operations, and then suitable M wavelet basis functions are selected for wavelet transformation. Specifically, as follows:

[0056] First, data standardization is performed. Converting the data to a distribution with a mean of 0 and a standard deviation of 1 helps eliminate the dimensional effects between different features and makes the data more comparable. For transaction amount and credit score.

[0057] Then, time feature extraction. Convert the transaction date and transaction time into timestamps or numerical features that can be used for analysis, so that users can capture patterns and trends in time series data. Convert the date and time strings into datetime objects, and then extract the required time features including month / week, etc.

[0058] Then, the categorical features are encoded. The transaction currency, transaction type, transaction channel, and transaction location are one-hot encoded, and for each categorical feature, a new binary column is created to represent each category of the feature.

[0059] Then, select the feature extraction algorithm. Wavelet transform has significant advantages in processing financial transaction data, especially in time-frequency localization, multi-scale analysis, sparse representation, non-stationary signal processing, noise suppression, etc., while Fourier transform can only analyze signals in the frequency domain and cannot provide information in the time domain.

[0060] As shown in Table 2, the advantages, disadvantages and applicable scope of different wavelet basis functions are analyzed in combination with specific financial transaction scenarios.

[0061] Table 2

[0062]

[0063] In operation S230, a transaction feature set is obtained according to the wavelet transform results of each of the M wavelet basis functions;

[0064] For example, the wavelet transform results of each wavelet basis function are integrated, such as combining frequency features of different dimensions to form a transaction feature set. Its function is to extract key features that can represent transaction information and provide a data basis for subsequent encryption. Alternatively, the wavelet transform results of each of the M wavelet basis functions can be input into a pre-trained neural network model to extract key features representing transaction information. Alternatively, further calculations can be performed based on the wavelet transform results of each of the M wavelet basis functions to obtain one or more key features representing transaction information.

[0065] In operation S240, a homomorphic encryption algorithm is called to perform homomorphic encryption on the transaction feature set to obtain transaction encrypted data.

[0066] For example, the Paillier homomorphic encryption algorithm is used to encrypt the transaction feature set to protect the security of transaction information and prevent data leakage.

[0067] For example, taking the financial transaction system of a certain institution as an example, first, after the user confirms the authorization, the transaction information of the user is obtained, including the transaction amount, transaction object, transaction time, etc. Then, multiple wavelet basis functions are called to perform wavelet transform on these transaction information. For example, one wavelet basis function is used to process the time series data containing the transaction amount, and another wavelet basis function is used to process the category data of the transaction object. Based on these wavelet transform results, a set of transaction features is extracted, such as the frequency characteristics and trend characteristics of the transaction. Finally, the homomorphic encryption algorithm is called to encrypt the transaction feature set to ensure the security of the transaction data during transmission and storage.

[0068] According to the embodiments of the present disclosure, the use of wavelet transform to process the transaction information of the target user can capture the complex feature changes in the massive transaction data, reduce the data dimension, and improve the efficiency of subsequent homomorphic encryption. Among them, the use of different wavelet basis functions to adapt and process the transaction data of different dimensions in the transaction information can more accurately extract the relevant features, so that the transaction encrypted data can reflect the characteristics of the target user. Therefore, without leaking the privacy of the user, it is possible to achieve the effects of reducing the amount of data transmission and encryption calculation, improving transmission efficiency, and obtaining more accurate transaction features.

[0069] In some embodiments, the wavelet transform results of each of the M wavelet basis functions include multiple groups of wavelet coefficient data corresponding one-to-one to multiple dimensional transaction data; obtaining a transaction feature set based on the wavelet transform results of each of the M wavelet basis functions includes: calculating the transaction feature set based on the multiple groups of wavelet coefficient data.

[0070] The wavelet coefficient data includes a number of wavelet coefficient values ​​or wavelet coefficient matrices extracted by the wavelet basis function from the transaction data of the adaptation dimension. For example, a transaction feature set can be obtained by performing statistical analysis on multiple groups of wavelet coefficient data, such as calculating the mean, variance, etc. Alternatively, the wavelet coefficients in each group of wavelet coefficient data are screened, and the wavelet coefficients above the average value are weighted and summed.

[0071] According to the embodiments of the present disclosure, by calculating a transaction feature set from multiple groups of wavelet coefficient data, it is possible to more accurately extract the features of transaction information based on the wavelet coefficients, thereby improving the pertinence and security of subsequent encryption.

[0072] There are many transaction records and amount data in the target user's transaction information. If homomorphic encryption is used, the number of ciphertext bits will increase, which will reduce the efficiency of privacy computing in terms of data transmission and storage. Therefore, it is considered to use wavelet transform to reduce the data dimension before homomorphic encryption processing and transmission, so as to improve the processing efficiency of subsequent homomorphic encryption and transmission process, while taking into account the accuracy of features after wavelet transform processing.

[0073] The following example uses the Daubechies wavelet basis function as the first wavelet basis function to perform wavelet transform on the transaction data in the dimension of consumption habits, and uses the Morlet wavelet basis function as the second wavelet basis function to perform wavelet transform on the transaction data in the dimension of transaction frequency fluctuation.

[0074] Figure 3 The flowchart of obtaining the first transaction feature according to an embodiment of the present disclosure is schematically shown.

[0075] like Figure 3 As shown, this embodiment includes operations S310 to S360.

[0076] In operation S310, Daubechies wavelet transform decomposition is performed.

[0077] For example, assuming that the decomposition scale is n levels, the wavelet coefficients of each field include cA (approximate coefficients), and cD1, cD2, .....cDn (n detail coefficients).

[0078] For example, the first wavelet basis function is called to perform wavelet transform on the transaction data of the consumption habit dimension to obtain a set of wavelet coefficient data corresponding to the consumption habit dimension; wherein the transaction data of the consumption habit dimension includes a variety of consumption data such as transaction amount, credit points, transaction time, transaction location and transaction currency, and the set of wavelet coefficient data corresponding to the consumption habit dimension includes a plurality of first wavelet coefficient data corresponding one-to-one to the plurality of consumption data.

[0079] The consumption habit dimension can reflect the consumption behavior patterns of users at different times and places. One or more data such as transaction amount, credit points, transaction location, transaction time and currency can be processed to more accurately extract the characteristic value of the consumption habit dimension.

[0080] In some embodiments, calculating a transaction feature set based on multiple groups of wavelet coefficient data includes: obtaining a sum result by adding the absolute values ​​of all wavelet coefficients in each first wavelet coefficient data; assigning a weight coefficient to each sum result to perform a weighted calculation on each sum result; and summing each sum result after the weighted calculation to obtain the first transaction feature in the transaction feature set.

[0081] The weight coefficient of each summation result may be preset, for example, weight coefficients of different sizes are allocated in different ranges according to the summation result. Alternatively, the weight coefficients may be calculated and allocated based on information entropy.

[0082] According to the embodiments of the present disclosure, by summing up the absolute values ​​of all wavelet coefficients in each first wavelet coefficient data, the complex wavelet coefficient data is simplified into a single summation result, and the overall characteristics of the dimension data are extracted, reflecting the comprehensive characteristics of the transaction data of the dimension after the wavelet transform. The weight coefficient of each summation result is assigned, and the weighted calculation is performed, which can be adjusted according to the importance of transaction data of different dimensions. By summing up each summation result after the weighted calculation, the transaction information of multiple dimensions can be integrated.

[0083] In some embodiments, assigning a weight coefficient to each sum result includes: calculating the information entropy of each sum result; and assigning a weight coefficient positively correlated with its information entropy to each sum result. In some embodiments, calculating the information entropy of each sum result includes: obtaining the information entropy of the sum result according to the proportion of each sum result in the total sum result. This is further described below in conjunction with operations S320 to S360.

[0084] In operation S320, each first wavelet coefficient data is aggregated.

[0085] For example, to sum the wavelet coefficients of the transaction amount, the processing formula is as follows.

[0086] =∣ ∣+

[0087] According to the same method, the sum of the absolute values ​​of the wavelet coefficients of the credit score, transaction location, transaction time and currency can be calculated, which are represented as sum_credit, sum_location, sum_time and sum_encoded respectively.

[0088] In operation S330, the probability distribution is normalized.

[0089] For example, the above sum is normalized to obtain a probability distribution result, such as the probability distribution of the transaction amount Pamount, and the calculation formula is as follows.

[0090]

[0091] According to the same method, the probability distribution of the wavelet coefficients of credit points, transaction location, transaction time and currency can be calculated, which are Pcredit, Plocation, Ptime and Pencoded respectively.

[0092] In operation S340, each information entropy is calculated.

[0093] For the information entropy Hamount of the transaction amount, the calculation formula is as follows.

[0094]

[0095] According to the same method, the information entropy of the wavelet coefficients of credit score, transaction location, transaction time and currency can be calculated, which are Hcredit, Hlocation, Htime and Hencoded respectively.

[0096] In operation S350, each weight coefficient is calculated.

[0097] The weight coefficients are positively correlated according to the information entropy of the transaction amount, credit score, transaction location, transaction time and currency. For example, the weight coefficients can be determined by a positive correlation function (such as a linear function, exponential function, etc.) based on the calculated information entropy. The weights can be reasonably allocated according to the degree of uncertainty reflected by the information entropy, so that the greater the uncertainty, the greater the impact.

[0098] In operation S360, the consumption characteristic Cr of the user in the past year, ie, the first transaction characteristic, is obtained.

[0099] The sum of the absolute values ​​of the wavelet coefficients of each variable is multiplied by the corresponding weight coefficient, and then the sum is calculated to obtain the first transaction feature. The calculation formula is as follows:

[0100]

[0101] In some embodiments, calling M wavelet basis functions to perform wavelet transform on transaction information includes: calling a second wavelet basis function to perform wavelet transform on transaction data in a transaction frequency fluctuation dimension to obtain a set of wavelet coefficient data corresponding to the transaction frequency fluctuation dimension; wherein the transaction data in the transaction frequency fluctuation dimension includes multiple transaction data sets obtained from multiple transaction cycles, and the set of wavelet coefficient data corresponding to the transaction frequency fluctuation dimension includes multiple second wavelet coefficient data corresponding one-to-one to the multiple transaction data sets.

[0102] The fluctuation of transaction frequency reflects the change of transaction frequency in different time windows. Analyzing the correlation characteristics of the periodicity reflected by the fluctuation of transaction frequency and the specific transaction time points is important for predicting user transaction behavior and identifying abnormal credit risks. The transaction cycle can be a time window such as day, week, month, etc. The transaction data set is collected from the transaction information of the target user in the corresponding transaction cycle.

[0103] It can be understood that the present disclosure is not limited to the consumption habit dimension and the transaction frequency fluctuation dimension, for example, it can also include the transaction peak dimension, the transaction channel dimension, the transaction object dimension, etc.

[0104] In some embodiments, each second wavelet coefficient data includes a wavelet coefficient matrix and transaction frequency data, and the transaction feature set is calculated based on multiple sets of wavelet coefficient data, including: forming a matrix variable based on all wavelet coefficient matrices and transaction frequency data; calling a frequent pattern growth algorithm to perform spatiotemporal correlation mining on the matrix variable to obtain a second transaction feature in the transaction feature set. Figure 4 Provide explanation.

[0105] Figure 4 The flowchart of obtaining the second transaction feature according to an embodiment of the present disclosure is schematically shown.

[0106] like Figure 4 As shown, this embodiment includes operations S401 to S411.

[0107] In operation S401, a pre-processed data set is input, for example, transaction information processed by operations such as cleaning and anonymization;

[0108] In operation S402, a period for aggregating transaction data is selected, such as monthly aggregation.

[0109] In operation S403, a data set in the current aggregated transaction cycle is obtained and recorded as Xi;

[0110] like:

[0111]

[0112] Among them, Xi includes the transaction amount greater than a certain threshold and the corresponding transaction time. In addition, intermediate variables such as the total transaction amount per month and the number of transactions can also be calculated.

[0113] In operation S404, it is determined whether there are any unprocessed transaction cycle data sets; if so, operations S405 to S407 are executed. If not, operations S405 to S411 are executed.

[0114] In operation S405, extract the current transaction cycle data and perform Morlet wavelet transform;

[0115] In operation S406, the wavelet coefficient matrix coeffs_Mi and the transaction frequency Fi of the corresponding data set are obtained, which are represented by [.....[coeffs_Mi,Fi]....], where coeffs_Mi,Fi are the results automatically output by the Morlet wavelet transform.

[0116] In operation S407, the corresponding feature combination is added to the matrix variable T and the counting variable i+1;

[0117] In operation S408, outputting the summarized matrix variable T;

[0118] For example, input Xi, sampling period sampling_period value and other data, after the i-th execution, output the two-dimensional wavelet coefficient matrix coeffs_Mi (where each row corresponds to a specific scale and each column corresponds to a specific time point), and output the transaction frequency Fi. The generated wavelet coefficient matrix coeffs_Mi and transaction frequency Fi are combined and stored in the matrix variable T.

[0119] In operation S409, Boolean format conversion: obtain the summary result of the matrix variable T and convert it into Boolean format.

[0120] In operation S410, FP-Growth (Frequent Pattern Growth Algorithm) is performed to mine frequent item sets; the Frequent Pattern Growth (FP-Growth) algorithm is an effective algorithm for mining frequent item sets. It is based on the fact that if an item set is frequent, then all its non-empty subsets are also frequent. The algorithm adopts a divide-and-conquer strategy to decompose the mining task into smaller subtasks and store the frequent item information in the transaction database by constructing a data structure called FP-tree (Frequent- Pattern Tree).

[0121] In operation S411, the association rule feature value Tr is output.

[0122] Call FP-Growth to mine spatiotemporal association rules, input a feature vector set including periodic change features and time features, and output the mined association rule features, namely the second transaction features, as shown in Table 3.

[0123] Table 3

[0124]

[0125] Among them, Feature_N represents the mined feature.

[0126] According to an embodiment of the present disclosure, by mining the spatiotemporal correlation of multiple transaction data sets of multiple transaction cycles, it is possible to evaluate possible transaction risks from the perspective of transaction frequency fluctuations.

[0127] The following further describes the cross-institutional risk assessment method in combination with the above-mentioned data encryption method.

[0128] In some embodiments, S encrypted transaction data of a target user in S institutions are obtained, wherein the encrypted transaction data of each institution is obtained by homomorphic encryption according to the data encryption method of any of the above embodiments, and S is an integer greater than or equal to 2; the decrypted S encrypted transaction data are input into a risk assessment model to obtain a risk assessment result.

[0129] Figure 5The flowchart of a cross-institutional risk assessment method according to another embodiment of the present disclosure is schematically shown.

[0130] like Figure 5 As shown, this embodiment includes operations S501 to S508, wherein each organization can perform operations S501 to S506 and S508 respectively, and the leading organization performs operation S507.

[0131] In operation S501, user information, historical transaction data, asset status, etc. of the financial institutions participating in the assessment are collected;

[0132] In operation S502, format conversion, abnormal data processing, generalization and anonymization processing are unified;

[0133] In operation S503, multiple wavelet basis functions are used to perform feature calculation to obtain a transaction feature set;

[0134] In operation S504, Paillier homomorphic encryption is used to select a prime number, a generator, and a random number for multiplication, and the transaction feature set is encrypted to obtain encrypted transaction data;

[0135] In operation S505, a certificate-based two-way identity authentication mechanism secure channel transmission is established, and ECDSA is used to improve the encryption transmission speed;

[0136] In some embodiments, the S institutions include a lead institution, and obtaining the S transaction encrypted data of the S institutions includes: transmitting the transaction encrypted data of each of the S-1 institutions to the lead institution; wherein, during the transmission process, key exchange and signature verification are performed according to the Elliptic Curve Digital Signature Algorithm (hereinafter referred to as ECDSA). Figure 6 Provide explanation.

[0137] Figure 6 The flowchart of establishing a secure transmission channel according to an embodiment of the present disclosure is schematically shown.

[0138] In operation S610, generate a certificate authority key and a self-signed certificate; use a specific tool to generate the certificate authority's key and self-signed certificate, generate an ECDSA key pair and a certificate signing request (CSR) for each authority, sign the CSR with a CA certificate, and generate a formal ECDSA certificate.

[0139] In operation S620, the server and client contexts are configured; the respective ECDSA certificates and private keys are loaded on the server and client of each organization, and are set to a mode that requires the other party's certificate.

[0140] In operation S630, a TLS connection is established to carry out secure data transmission; the client and the server use the TLS protocol to establish a secure connection and perform two-way identity authentication.

[0141] In operation S640, ECDSA is used for key exchange and signature verification. After the TLS connection is established, the client and the server can securely transmit encrypted user risk assessment data of different financial institutions. During the TLS handshake process, ECDSA is used for key exchange and signature verification to reduce the computational burden.

[0142] For example, financial institutions A, B, and C provide ECDSA certificates to the client, which uses the certificates to establish a TLS connection with the server and perform two-way identity authentication. After the connection is established, the client and the server can securely transmit encrypted data and establish a secure transmission channel with other institutions.

[0143] According to the embodiments of the present disclosure, ECDSA can reduce the handshake calculation burden and improve the processing speed, while maintaining the security of data transmission and ensuring the confidentiality and integrity of data transmission.

[0144] In operation S506, ciphertext data reception, access control and database security management; after any financial institution participating in the risk assessment receives the encrypted data, it stores it in its own database and keeps the data in an encrypted state. Access control is set for the database to allow authorized users to access the encrypted data in the database. In addition, all operations on the database can be recorded and monitored to timely discover and handle security incidents. In addition, a regular cleanup mechanism can be set up. When the cross-institutional user risk assessment is completed and the comprehensive evaluation results are obtained, the corresponding financial institution can clean up and delete the historical encrypted data in a timely manner.

[0145] In operation S507, cross-institutional risk assessment multi-party secure computation is performed; after receiving data from different institutions, the lead institution can perform a risk assessment on the user by executing a multi-party secure computation protocol.

[0146] Figure 7 The flowchart of multi-party secure computing according to an embodiment of the present disclosure is schematically shown.

[0147] like Figure 7 As shown, this embodiment includes operations S710 to S770, wherein operations S710 to S750 may be performed before operation S507.

[0148] In operation S710, the number of financial institutions participating in the assessment is determined. Assume that there are three participating institutions, namely institution A, institution B, and institution C.

[0149] In operation S720, each organization independently generates a pair of public and private keys, and the private key is stored independently. 2. Each organization independently generates a pair of public and private keys, the public key is used for encryption, and the private key is first stored independently. Assume that the private key of organization A is sk_A, the private key of organization B is sk_B, and the private key of organization N is sk_N.

[0150] In operation S730, each organization uses its own public key to perform independent encryption processing to generate ciphertext data, such as executing the data encryption method of any of the above embodiments.

[0151] In operation S740, the leading integration and aggregation organization is confirmed to integrate all private key information. Since homomorphic encryption does not support directly using one private key to decrypt the results of multiple public key encryptions, the leading integration and aggregation organization is first confirmed. Assuming that organization A is selected, organization A uses key negotiation to integrate all private key information into a decryption key sk_combined.

[0152] In operation S750, other institutions send the ciphertext data to the lead institution using a secure transmission channel, for example, the secure transmission channel can be based on Figure 6 The process shown is established.

[0153] In operation S760, the lead organization performs homomorphic addition aggregation calculation. The lead organization A performs comprehensive calculations. After receiving the ciphertext, it first performs homomorphic addition operations on Ct_A and Ct_B to obtain Ct_AB = Ct_A *Ct_B = E(M_A + M_B, pk_A * pk_B) ("*" refers to the addition operation under homomorphic encryption), then adds the C organization data M_C, and continues to perform homomorphic addition operations on Ct_AB and Ct_C to obtain Ct_ABC = Ct_AB * Ct_C = E(M_A +M_B + M_C, pk_A * pk_B * pk_C).

[0154] Furthermore, based on the collaborative processing using homomorphic encryption and secure multi-party computing, in operation S770, the lead institution can also input Ct_ABC (i.e., S encrypted transaction data) into a pre-trained risk assessment model (e.g., a neural network model) to obtain the risk assessment results, and the calculation results are returned to each financial institution in encrypted form.

[0155] In operation S508, the decrypted comprehensive evaluation result is used for subsequent comprehensive risk assessment and decision-making of users. That is, each financial institution can use its own private key to decrypt and obtain the final risk assessment result.

[0156] According to the embodiments of the present disclosure, the use of wavelet transform and homomorphic encryption technology can effectively reduce the encrypted data dimension, reduce the amount of data transmission, and improve the efficiency of data transmission and homomorphic encryption. Under the premise of not leaking user privacy, multi-party secure computing is implemented across institutions. On the basis of ensuring the security of data transmission, the institutions participating in the assessment can complete the cross-institutional user risk assessment without sharing the original data, which improves the security and convenience of cross-institutional risk assessment.

[0157] Based on the above data encryption method, the present disclosure also provides a data encryption device. Figure 8 The device is described in detail.

[0158] Figure 8 The structural block diagram of the data encryption device according to the embodiment of the present disclosure is schematically shown.

[0159] like Figure 8 As shown, the data encryption device 800 of this embodiment includes a first acquisition module 810, a first calling module 820, a feature calculation module 830 and a homomorphic encryption module 840.

[0160] The first acquisition module 810 may perform operation S210 to acquire transaction information of a target user under authorization;

[0161] The first calling module 820 may perform operation S220 for calling M wavelet basis functions to perform wavelet transform on the transaction information, wherein any two wavelet basis functions are configured to adapt to process transaction data of different dimensions in the transaction information, and M is an integer greater than or equal to 2;

[0162] The feature calculation module 830 may perform operation S230 for obtaining a transaction feature set according to the wavelet transform results of each of the M wavelet basis functions;

[0163] The homomorphic encryption module 840 may execute operation S240 to call a homomorphic encryption algorithm to perform homomorphic encryption on a transaction feature set to obtain transaction encrypted data.

[0164] In some embodiments, the wavelet transform results of each of the M wavelet basis functions include multiple groups of wavelet coefficient data corresponding one-to-one to multiple dimensional transaction data; the first calling module 820 is used to calculate the transaction feature set based on the multiple groups of wavelet coefficient data.

[0165] In some embodiments, the first calling module 820 is used to call the first wavelet basis function to perform wavelet transform on the transaction data of the consumption habit dimension to obtain a set of wavelet coefficient data corresponding to the consumption habit dimension.

[0166] In some embodiments, the feature calculation module 830 is used to obtain a sum result by adding the absolute values ​​of all wavelet coefficients in each first wavelet coefficient data; assigning a weight coefficient to each sum result to perform weighted calculation on each sum result; and summing each sum result after weighted calculation to obtain the first transaction feature in the transaction feature set.

[0167] In some embodiments, the feature calculation module 830 is used to calculate the information entropy of each summation result; and assign a weight coefficient positively correlated with its information entropy to each summation result.

[0168] In some embodiments, the feature calculation module 830 is used to obtain the information entropy of the sum result according to the proportion of each sum result in the total sum result.

[0169] In some embodiments, the first calling module 820 is used to call the second wavelet basis function to perform wavelet transform on the transaction data of the transaction frequency fluctuation dimension to obtain a set of wavelet coefficient data corresponding to the transaction frequency fluctuation dimension.

[0170] In some embodiments, the feature calculation module 830 is used to form a matrix variable based on all wavelet coefficient matrices and transaction frequency data; call the frequent pattern growth algorithm to perform spatiotemporal association mining on the matrix variable to obtain the second transaction feature in the transaction feature set.

[0171] Based on the above cross-institutional risk assessment method, the present disclosure also provides a cross-institutional risk assessment device. Fig. 9 The device is described in detail.

[0172] Fig. 9 The structural block diagram of the data encryption device according to the embodiment of the present disclosure is schematically shown.

[0173] like Fig. 9 As shown, the cross-institutional risk assessment device 900 of this embodiment includes a second acquisition module 910 and a risk assessment module 920 .

[0174] The second acquisition module 910 is used to obtain S encrypted transaction data of the target user in S institutions, wherein the encrypted transaction data of each institution is obtained by homomorphic encryption according to any one of the methods of claims 1 to 8, and S is an integer greater than or equal to 2;

[0175] The risk assessment module 920 is used to input the S encrypted transaction data into the risk assessment model to obtain a risk assessment result.

[0176] In some embodiments, the second acquisition module 910 is used to transmit the encrypted transaction data of each of the S-1 institutions to the lead institution; wherein, during the transmission process, key exchange and signature verification are performed according to the elliptic curve digital signature algorithm.

[0177] For the parts not mentioned in each device part, they can be understood by referring to the various embodiments of the method corresponding to the above-mentioned device. That is, each device part includes modules for executing the various steps of any corresponding method embodiment described above. In addition, the implementation methods, technical problems solved, functions implemented, and technical effects achieved of each module / unit / subunit, etc. in each device part embodiment are respectively the same or similar to the implementation methods, technical problems solved, functions implemented, and technical effects achieved of each corresponding step in the corresponding method part embodiment, and will not be repeated here.

[0178] According to an embodiment of the present disclosure, any multiple modules in the data encryption device 800 or the cross-institutional risk assessment device 900 can be combined into one module for implementation, or any one 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.

[0179] According to an embodiment of the present disclosure, at least one of the data encryption device 800 or the cross-institutional risk assessment device 900 may 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 may be implemented by hardware or firmware in any other reasonable manner of integrating or packaging the circuit, or in any one of the three implementation methods of software, hardware, and firmware, or in an appropriate combination of any of them. Alternatively, at least one of the data encryption device 800 or the cross-institutional risk assessment device 900 may be at least partially implemented as a computer program module, which may perform corresponding functions when the computer program module is run.

[0180] Fig.10 A block diagram of an electronic device suitable for implementing a data encryption method or a cross-institutional risk assessment method according to an embodiment of the present disclosure is schematically shown.

[0181] like Fig.10As shown, the electronic device 1000 according to an embodiment of the present disclosure includes a processor 1001, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1002 or a program loaded from a storage part 1008 to a random access memory (RAM) 1003. The processor 1001 may include, for example, a general-purpose microprocessor (such as a CPU), an instruction set processor and / or a related chipset and / or a special-purpose microprocessor (for example, an application-specific integrated circuit (ASIC)), etc. The processor 1001 may also include an onboard memory for caching purposes. The processor 1001 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of the present disclosure.

[0182] In RAM 1003, various programs and data required for the operation of electronic device 1000 are stored. Processor 1001, ROM 1002 and RAM 1003 are connected to each other via bus 1004. Processor 1001 performs various operations of the method flow according to the embodiment of the present disclosure by executing the program in ROM 1002 and / or RAM 1003. It should be noted that the program can also be stored in one or more memories other than ROM 1002 and RAM 1003. Processor 1001 can also perform various operations of the method flow according to the embodiment of the present disclosure by executing the program stored in one or more memories.

[0183] According to an embodiment of the present disclosure, the electronic device 1000 may further include an input / output (I / O) interface 1005, which is also connected to the bus 1004. The electronic device 1000 may further include one or more of the following components connected to the input / output (I / O) interface 1005: an input portion 1006 including a keyboard, a mouse, etc.; an output portion 1007 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage portion 1008 including a hard disk, etc.; and a communication portion 1009 including a network interface card such as a LAN card, a modem, etc. The communication portion 1009 performs communication processing via a network such as the Internet. A drive 1010 is also connected to the input / output (I / O) interface 1005 as needed. A removable medium 1011, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 1010 as needed, so that a computer program read therefrom is installed into the storage portion 1008 as needed.

[0184] 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.

[0185] According to an embodiment of the present disclosure, a computer-readable storage medium may be a non-volatile computer-readable storage medium, for example, may include but is not limited to: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, an apparatus or a device. For example, according to an embodiment of the present disclosure, a computer-readable storage medium may include the ROM 1002 and / or RAM 1003 described above and / or one or more memories other than ROM 1002 and RAM 1003.

[0186] The embodiment of the present disclosure also includes a computer program product, which includes a computer program, and the computer program contains program code for executing the method shown in the flowchart. When the computer program product is run in a computer system, the program code is used to enable the computer system to implement the method provided by the embodiment of the present disclosure.

[0187] The above functions defined in the system / device of the embodiment of the present disclosure are performed when the computer program is executed by the processor 1001. According to the embodiment of the present disclosure, the system, device, module, unit, etc. described above can be implemented by a computer program module.

[0188] In one embodiment, the computer program may rely on tangible storage media such as optical storage devices, magnetic storage devices, etc. In another embodiment, the computer program may also be transmitted and distributed in the form of signals on a network medium, and downloaded and installed through the communication part 1009, and / or installed from the removable medium 1011. The program code contained in the computer program may be transmitted using any appropriate network medium, including but not limited to: wireless, wired, etc., or any suitable combination of the above.

[0189] In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 1009, and / or installed from the removable medium 1011. When the computer program is executed by the processor 1001, the above functions defined in the system of the embodiment of the present disclosure are performed. According to the embodiment of the present disclosure, the system, device, means, module, unit, etc. described above can be implemented by a computer program module.

[0190] According to an embodiment of the present disclosure, the program code for executing the computer program provided by the embodiment of the present disclosure can be written in any combination of one or more programming languages. Specifically, these computing programs can be implemented using high-level process and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages ​​include, but are not limited to, Java, C++, python, "C" language or similar programming languages. The program code can be executed entirely on the user computing device, partially on the user device, partially on the remote computing device, or entirely on the remote computing device or server. In the case of a remote computing device, the remote computing device can be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computing device (for example, using an Internet service provider to connect through the Internet).

[0191] 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.

[0192] It will be appreciated by those skilled in the art that the features described in the various embodiments of the present disclosure may be combined and / or combined in a variety of ways, even if such combinations or combinations are not explicitly described in the present disclosure. In particular, without departing from the spirit and teachings of the present disclosure, the features described in the various embodiments of the present disclosure may be combined and / or combined in a variety of ways. All of these combinations and / or combinations fall within the scope of the present disclosure.

[0193] The embodiments of the present disclosure are described above. However, these embodiments are only for the purpose of illustration and are not intended to limit the scope of the present disclosure. Although the embodiments are described above separately, this does not mean that the measures in the various embodiments cannot be used in combination to advantage. The scope of the present disclosure is defined by the attached claims and their equivalents. Without departing from the scope of the present disclosure, those skilled in the art may make various substitutions and modifications, which should all fall within the scope of the present disclosure.

Claims

1. A data encryption method, comprising: Obtain the target user's transaction information with authorization; Calling M wavelet basis functions to perform wavelet transform on the transaction information, wherein any two wavelet basis functions are configured to adapt to process transaction data of different dimensions in the transaction information, and M is an integer greater than or equal to 2; Obtaining a transaction feature set according to the wavelet transform results of each of the M wavelet basis functions; A homomorphic encryption algorithm is called to perform homomorphic encryption on the transaction feature set to obtain transaction encrypted data.

2. The method according to claim 1, characterized in that: The wavelet transform results of each of the M wavelet basis functions include multiple groups of wavelet coefficient data corresponding one-to-one to the transaction data of multiple dimensions; The obtaining of the transaction feature set according to the wavelet transform results of each of the M wavelet basis functions comprises: The transaction feature set is calculated based on the multiple groups of wavelet coefficient data.

3. The method according to claim 2, characterized in that The M wavelet basis functions include a first wavelet basis function, and calling the M wavelet basis functions to perform wavelet transform on the transaction information includes: Calling the first wavelet basis function to perform wavelet transform on the transaction data of the consumption habit dimension to obtain a set of wavelet coefficient data corresponding to the consumption habit dimension; Among them, the transaction data of the consumption habit dimension includes multiple consumption data such as transaction amount, credit points, transaction time, transaction location and transaction currency, and a set of wavelet coefficient data corresponding to the consumption habit dimension includes multiple first wavelet coefficient data corresponding one by one to the multiple consumption data.

4. The method according to claim 3, characterized in that The transaction feature set calculated based on the multiple groups of wavelet coefficient data includes: Obtaining a sum result by summing up the absolute values ​​of all wavelet coefficients in each of the first wavelet coefficient data; Allocating a weight coefficient to each of the summed results to perform weighted calculation on each of the summed results; Each of the weighted calculation results is summed to obtain the first transaction feature in the transaction feature set.

5. The method according to claim 4, characterized in that The weight coefficients for allocating each of the summation results include: Calculating the information entropy of each of the summation results; A weight coefficient positively correlated with the information entropy thereof is assigned to each of the summation results.

6. The method according to claim 5, characterized in that The calculating of the information entropy of each of the summation results comprises: According to the proportion of each sum result in the total sum result, the information entropy of the sum result is obtained.

7. The method according to claim 2, characterized in that: The M wavelet basis functions include a second wavelet basis function, and calling the M wavelet basis functions to perform wavelet transform on the transaction information includes: Calling the second wavelet basis function to perform wavelet transform on the transaction data of the transaction frequency fluctuation dimension to obtain a set of wavelet coefficient data corresponding to the transaction frequency fluctuation dimension; Among them, the transaction data of the transaction frequency fluctuation dimension includes multiple transaction data sets obtained from multiple transaction cycles, and a set of wavelet coefficient data corresponding to the transaction frequency fluctuation dimension includes multiple second wavelet coefficient data corresponding one-to-one to the multiple transaction data sets.

8. The method according to claim 7, characterized in that Each of the second wavelet coefficient data includes a wavelet coefficient matrix and transaction frequency data, and the transaction feature set calculated based on the multiple sets of wavelet coefficient data includes: forming a matrix variable based on all of said wavelet coefficient matrices and transaction frequency data; A frequent pattern growth algorithm is called to perform spatiotemporal association mining on the matrix variables to obtain a second transaction feature in the transaction feature set.

9. A cross-institutional risk assessment method, comprising: Obtain S encrypted transaction data of the target user in S institutions, wherein the encrypted transaction data of each institution is obtained by homomorphic encryption according to any one of the methods of claims 1 to 8, and S is an integer greater than or equal to 2; The S encrypted transaction data are input into a risk assessment model to obtain a risk assessment result.

10. The method according to claim 9, characterized in that The S institutions include a leading institution, and obtaining the S encrypted transaction data of the S institutions includes: Transmitting the encrypted transaction data of each S-1 institution to the lead institution; Among them, key exchange and signature verification are performed according to the elliptic curve digital signature algorithm during the transmission process.

11. A data encryption device, comprising: The first acquisition module is used to acquire the transaction information of the target user under the condition of obtaining authorization; A first calling module is used to call M wavelet basis functions to perform wavelet transform on the transaction information, wherein any two wavelet basis functions are configured to adapt to process transaction data of different dimensions in the transaction information, and M is an integer greater than or equal to 2; A feature calculation module, used for obtaining a transaction feature set according to the wavelet transform results of each of the M wavelet basis functions; The homomorphic encryption module is used to call the homomorphic encryption algorithm to perform homomorphic encryption on the transaction feature set to obtain transaction encrypted data.

12. A cross-institutional risk assessment device, comprising: The second acquisition module is used to obtain S encrypted transaction data of the target user in S institutions, wherein the encrypted transaction data of each institution is obtained by homomorphic encryption according to any one of the methods of claims 1 to 8, and S is an integer greater than or equal to 2; The risk assessment module is used to input the S encrypted transaction data into the risk assessment model to obtain a risk assessment result.

13. An electronic device comprising: one or more processors; a memory for storing one or more computer programs, It is characterized in that the one or more processors execute the one or more computer programs to implement the steps of the method according to any one of claims 1 to 10.

14. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 10 are implemented.

15. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 10 are implemented.

Images