Data processing method and device, computer equipment and storage medium

Abstract

The invention relates to a data processing method and device, computer equipment, a storage medium and a computer program product. The method comprises the steps of obtaining a reference dimension of user data through a reference platform, querying a target dimension complementary to the reference dimension in a preset dimension list of the user data, determining a target platform of the user data according to the target dimension, and controlling the reference platform and the target platform, and performing homomorphic encryption processing on the user data of the reference platform and the user data of the target platform to obtain at least two pieces of homomorphic encryption data, decrypting the combined homomorphic encryption data to obtain combined user data, and inputting the combined user data into a pre-trained user portrait evaluation model, according to the method, the user data of the multiple platforms are integrated by querying the target dimension complementary with the reference dimension, and the safety among the user data of the platforms is guaranteed.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a data processing method, apparatus, computer equipment, storage medium and computer program product. Background Art

[0002] With the development of the economy, in order to improve the effectiveness of product marketing, various banks or institutions will analyze information such as users' consumption habits, identity information and customer characteristics to obtain user portraits.

[0003] In the existing method, user data from multiple banks or institutions is obtained and user portraits are generated based on the data of multiple users. This method has the risk of customer information leakage. Summary of the invention

[0004] Based on this, it is necessary to provide a data processing method, apparatus, computer device, computer-readable storage medium and computer program product that can improve the security of user data in response to the above technical problems.

[0005] In a first aspect, the present application provides a data processing method. The method comprises:

[0006] If the user identifier is an old user identifier, a reference dimension of the user data is obtained according to the reference platform, and a target dimension complementary to the reference dimension is searched in a preset dimension list of the user data;

[0007] Determining a target platform for user data according to the target dimension;

[0008] Controlling the reference platform and the target platform to perform homomorphic encryption processing on user data of the reference platform and user data of the target platform respectively, to obtain at least two homomorphic encrypted data;

[0009] Combining the at least two homomorphically encrypted data;

[0010] Decrypting the combined homomorphically encrypted data to obtain combined user data, and inputting the combined user data into a pre-trained user profile evaluation model to obtain a user profile of the user;

[0011] Recommend products to the user based on the user portrait.

[0012] In one embodiment, the method further comprises:

[0013] If the user's ID is a new user ID, the default dimension of the user data is obtained;

[0014] In the preset dimension list of the user data, query the associated platform associated with the default dimension of the user data;

[0015] Controlling the associated platform to perform homomorphic encryption processing on user data in the associated platform to obtain associated homomorphic encrypted data;

[0016] The associated homomorphically encrypted data is decrypted and input into a pre-trained user portrait evaluation model to obtain a user portrait of the user.

[0017] In one embodiment, the method further comprises:

[0018] In response to the selection operation of each platform for a key, obtaining a session key of the platform, wherein the session key includes a private key for homomorphic decryption and a public key for homomorphic encryption;

[0019] Configuring the public key for each of the platforms;

[0020] The private key is configured for each platform according to the type of the platform.

[0021] In one embodiment, combining the at least two homomorphically encrypted data includes:

[0022] Obtaining an arrangement order of the user data dimensions according to a preset dimension list of the user data;

[0023] Based on the arrangement order of the user data dimensions, the at least two homomorphically encrypted data are combined.

[0024] In one embodiment, the method further comprises:

[0025] If the key indicator in the result of the product recommendation does not reach the preset value, querying the missing dimension of the user data in the preset dimension list of the user data, wherein the missing dimension, the reference dimension and the target dimension do not overlap with each other;

[0026] Based on the missing dimension, obtaining user data having the same missing dimension from other platforms;

[0027] Performing homomorphic encryption processing on the user data of the missing dimension, the reference dimension, and the target dimension respectively to obtain at least three homomorphic encrypted data;

[0028] The at least three homomorphically encrypted data are combined.

[0029] In one embodiment, the dimension of the user data includes at least one of a basic dimension, a platform dimension, a user behavior dimension, a platform product dimension, and an early warning dimension, and the method further includes:

[0030] Based on the preset dimension list of the user data, the recommendation results of the product recommendation are displayed.

[0031] In a second aspect, the present application also provides a data processing device. The device comprises:

[0032] an acquisition module, configured to acquire a reference dimension of user data according to a reference platform if the user identifier is an old user identifier, and search a preset dimension list of user data for a target dimension complementary to the reference dimension;

[0033] A determination module, used to determine a target platform of user data according to the target dimension;

[0034] An encryption module, used to control the reference platform and the target platform, and perform homomorphic encryption processing on the user data of the reference platform and the user data of the target platform respectively, to obtain at least two homomorphic encrypted data;

[0035] A combining module, used for combining the at least two homomorphically encrypted data;

[0036] A decryption module, used to decrypt the combined homomorphically encrypted data to obtain combined user data, and input the combined user data into a pre-trained user portrait evaluation model to obtain a user portrait of the user;

[0037] A recommendation module is used to recommend products to the user based on the user portrait.

[0038] In a third aspect, the present application further provides a computer device. The computer device includes a memory and a processor, the memory stores a computer program, and the processor implements the following steps when executing the computer program:

[0039] If the user identifier is an old user identifier, a reference dimension of the user data is obtained according to the reference platform, and a target dimension complementary to the reference dimension is searched in a preset dimension list of the user data;

[0040] Determining a target platform for user data according to the target dimension;

[0041] Controlling the reference platform and the target platform to perform homomorphic encryption processing on user data of the reference platform and user data of the target platform respectively, to obtain at least two homomorphic encrypted data;

[0042] Combining the at least two homomorphically encrypted data;

[0043] Decrypting the combined homomorphically encrypted data to obtain combined user data, and inputting the combined user data into a pre-trained user profile evaluation model to obtain a user profile of the user;

[0044] Recommend products to the user based on the user portrait.

[0045] In a fourth aspect, the present application further provides a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the following steps are implemented:

[0046] If the user identifier is an old user identifier, a reference dimension of the user data is obtained according to the reference platform, and a target dimension complementary to the reference dimension is searched in a preset dimension list of the user data;

[0047] Determining a target platform for user data according to the target dimension;

[0048] Controlling the reference platform and the target platform to perform homomorphic encryption processing on user data of the reference platform and user data of the target platform respectively, to obtain at least two homomorphic encrypted data;

[0049] Combining the at least two homomorphically encrypted data;

[0050] Decrypting the combined homomorphically encrypted data to obtain combined user data, and inputting the combined user data into a pre-trained user profile evaluation model to obtain a user profile of the user;

[0051] Recommend products to the user based on the user portrait.

[0052] In a fifth aspect, the present application further provides a computer program product. The computer program product includes a computer program, and when the computer program is executed by a processor, the following steps are implemented:

[0053] If the user identifier is an old user identifier, a reference dimension of the user data is obtained according to the reference platform, and a target dimension complementary to the reference dimension is searched in a preset dimension list of the user data;

[0054] Determining a target platform for user data according to the target dimension;

[0055] Controlling the reference platform and the target platform to perform homomorphic encryption processing on user data of the reference platform and user data of the target platform respectively, to obtain at least two homomorphic encrypted data;

[0056] Combining the at least two homomorphically encrypted data;

[0057] Decrypting the combined homomorphically encrypted data to obtain combined user data, and inputting the combined user data into a pre-trained user profile evaluation model to obtain a user profile of the user;

[0058] Recommend products to the user based on the user portrait.

[0059] The above-mentioned data processing method, device, computer equipment, storage medium and computer program product, if the user's identification is an old user identification, obtains the baseline dimension of the user data according to the baseline platform, queries the target dimension complementary to the baseline dimension in the preset dimension list of the user data, and queries the target dimension complementary to the baseline dimension from the preset dimension list, so that the dimension of the user data is more comprehensive, which helps to improve the subsequent prediction accuracy of the user portrait. Determine the platform of the user data according to the target dimension, control the baseline platform and the target platform to perform homomorphic encryption processing on the user data respectively, and obtain at least two homomorphic encrypted data. The user data of different platforms are homomorphically encrypted respectively, so that the plaintext of the user data of different platforms is invisible to each other, which improves the security of the user data. Combine at least two homomorphic encrypted data, decrypt the combined homomorphic encrypted data, obtain the combined user data, and input the combined user data into the pre-trained user portrait evaluation model to obtain the user portrait of the user. Based on homomorphic encryption, when the user data is available but not visible, the user data of multiple platforms are integrated to construct a more three-dimensional user portrait. Product recommendations are made to users based on their user portraits. This method queries the target dimension that is complementary to the baseline dimension, and then integrates user data from multiple platforms. This ensures the security of user data on each platform while building a more three-dimensional user portrait and improving the accuracy of product recommendations. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] FIG1 is an application environment diagram of a data processing method in one embodiment;

[0061] FIG2 is a schematic flow chart of a data processing method in one embodiment;

[0062] FIG3 is a schematic flow chart of a data processing method in another embodiment;

[0063] FIG4 is a block diagram of a data processing device in one embodiment;

[0064] FIG. 5 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0065] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0066] The data processing method provided in the embodiment of the present application can be applied in the application environment shown in FIG1 . In which, the terminal 102 communicates with the server 104 through a network. The data storage system can store the data that the server 104 needs to process. The data storage system can be integrated on the server 104, or it can be placed on the cloud or other network servers.

[0067] If the user's identifier is an old user identifier, the server 104 obtains the baseline dimension of the user data through the baseline platform 106, and searches for the target dimension complementary to the baseline dimension in the preset dimension list of the user data. The server 104 determines the target platform 108 of the user data according to the target dimension. The server 104 controls the baseline platform 106 and the target platform 108, and performs homomorphic encryption processing on the user data of the baseline platform 106 and the user data of the target platform 108 respectively to obtain at least two homomorphic encrypted data. The server 104 combines the at least two homomorphic encrypted data. The server 104 inputs the combined homomorphic encrypted data into a pre-trained user portrait evaluation model to obtain an encrypted predicted portrait of the user, and decrypts the encrypted predicted portrait of the user to obtain a user portrait of the user. The server 104 recommends products to the user based on the user portrait and pushes them to the terminal 102.

[0068] The terminal 102 may be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, IoT devices and portable wearable devices. The IoT devices may be smart speakers, smart TVs, smart air conditioners, smart car-mounted devices, etc. Portable wearable devices may be smart watches, smart bracelets, head-mounted devices, etc. The server 104 may be implemented by an independent server or a server cluster consisting of multiple servers. The reference platform 106 and the target platform 106 may be system platforms of banks or other institutions, and may be implemented by an independent server or a server cluster consisting of multiple servers.

[0069] In one embodiment, as shown in FIG. 2 , a data processing method is provided, which is described by taking the method applied to the server in FIG. 1 as an example, including:

[0070] S202: If the user identifier is an old user identifier, a reference dimension of the user data is obtained through a reference platform, and a target dimension complementary to the reference dimension is searched in a preset dimension list of the user data.

[0071] Specifically, before obtaining the user identifier as an old user identifier, the user identifier can be obtained by obtaining a request message initiated by the user on the reference platform and parsing the request message. The user identifier is an identifier of whether the user has user data stored on the reference platform. The user identifier includes an old user identifier and a new user identifier. The old user identifier represents that the user data of the user has been stored in the reference platform, and the new user identifier represents that the user data is not stored in the reference platform. Taking the platform as a user as an example, the old user identifier represents that the user is an old user of the bank, and the new user identifier represents that the user is a new customer.

[0072] Among them, the request message is a product request message initiated by the request message user at the benchmark platform. Specifically, after the user sends a request message to the benchmark platform, different judgments are made based on the user identifier in the request message. If the user identifier is an old user identifier, the user portrait can be constructed based on the user data of the benchmark platform. The user portrait can also be constructed based on a joint analysis of user data in the benchmark platform and other related platforms. If the user identifier is a new user identifier, it is necessary to obtain user data from other platforms to construct the user portrait. When the user portrait is constructed, product recommendations can be made to the user based on the user portrait, and the product recommendation results are written into the response message of the request message and returned to the user.

[0073] Among them, the benchmark platform can be a platform where users initiate request messages. The benchmark platform can be a system platform of a bank or other financial institution. Users can interact directly with the benchmark platform. For example, a user initiates a request message for a financial product in the benchmark platform, and the benchmark platform obtains the processing result of the request message (recommended product A) from the server and returns the processing result to the user as a response message.

[0074] Specifically, if the user identifier is an old user identifier, it means that user data is stored in the benchmark platform, and the user portrait can be predicted based on the user data of the benchmark platform, or by combining user data from other platforms.

[0075] Among them, the user data stored for a specific user in different platforms is different. For example, the basic attribute data of the user (identity, age, gender and other basic attributes) is stored in the benchmark platform A, and the user behavior attribute data (user login, transaction, logout and other behavior attributes in a certain APP) is stored in other platforms. Therefore, the user data of different dimensions of multiple platforms can be integrated to establish a comprehensive user data sample to improve the prediction accuracy of user portraits.

[0076] The preset dimension list is a list of multiple dimensions included in the user data. The preset dimension list can be set to different sizes according to the product distribution of the benchmark platform. For example, when the benchmark platform has more product types, the user data in the preset dimension list will have a wider range of dimensions. The preset dimension list can be a pre-established table and can be enriched according to the business expansion of the benchmark platform. When the number of products in the benchmark platform increases, the user data dimensions in the preset dimension list will also be enriched.

[0077] The preset dimension list includes user data dimensions. Specifically, the dimensions of user data in the benchmark platform are benchmark dimensions. Generally, the dimensions of user data in the benchmark platform are not sufficient to support the prediction of user profiles. Therefore, it is necessary to obtain user data of other dimensions except the benchmark dimensions. Specifically, the user data under the target dimension can be obtained by querying the preset dimension list of user data for a target dimension that is complementary to the benchmark dimension.

[0078] Among them, the preset dimension list includes all dimensions of user data, and the dimensions of user data include: basic dimension, platform dimension, user behavior dimension, platform product dimension, early warning dimension, etc. User data in basic dimensions include: occupation, age, gender, name, etc. User data in platform dimensions include: platform location, platform logo, real-time distance between platform and user, etc. User behavior dimension includes: user login, transaction, logout and other behaviors on the platform User data in platform product dimension includes: user preference for various bank products, such as platform search preference, financial product preference, etc. User data in early warning dimension includes: user risk situation, risk situation between user and platform, etc.

[0079] Specifically, the target dimension and the benchmark dimension present a complementary relationship in the preset dimension list. For example, if the benchmark dimension includes the basic dimension, the target dimension may be at least one of the platform dimension, the user behavior dimension, the platform product dimension and the warning dimension.

[0080] S204, determining the target platform of the user data according to the target dimension.

[0081] After the target dimension is determined, the target platform of the user data is queried according to the user identifier and the target dimension. The target platform includes the user data of the target dimension, and the user data of the target dimension is obtained by querying the target dimension of the user data in each platform.

[0082] Specifically, when querying the target dimension of user data, if multiple platforms all include user data of the target dimension, the multiple platforms can be sorted in descending order according to the amount of user data contained in the platform, and the platform with the highest ranking can be determined as the target platform of the user data. It is also possible to sort multiple platforms in descending order according to the number of dimensions of user data contained in the platform, and determine the platform with the highest ranking as the target platform of the user data. It is also possible to first sort multiple platforms in descending order according to the number of dimensions of user data contained in the platform, and then weight each platform in the sorting result according to the amount of user data contained in the platform to obtain the final sorting result, and determine the platform with the highest ranking in the final sorting result as the target platform.

[0083] S206, controlling the reference platform and the target platform to perform homomorphic encryption processing on the user data of the reference platform and the user data of the target platform respectively, to obtain at least two homomorphic encrypted data.

[0084] Among them, homomorphic encryption is a cryptographic technology based on the computational complexity theory of mathematical problems. The data that has been homomorphically encrypted is processed to obtain an output, and this output is decrypted. The result is the same as the output obtained by processing the unencrypted original data in the same way. In the case of being unable to obtain multi-dimensional user data across industries, the platform does not have a comprehensive understanding of users, and it is difficult to make targeted service and product recommendations. Based on homomorphic encryption, data from inside and outside the industry can be integrated while making the data "available but invisible", to achieve matching of specific customer groups from different institutions, build a more three-dimensional user portrait, and then achieve precision marketing for specific customer groups.

[0085] Specifically, the control base platform performs homomorphic encryption on the user data of the base platform, and the control target platform performs homomorphic encryption on the user data of the target platform. Before homomorphic encryption, it is necessary to obtain the platform's session key, which includes a public key and a private key. The public key is used for each platform to perform homomorphic encryption, and the private key is used for subsequent platforms to decrypt the encrypted data.

[0086] S208, combine at least two homomorphically encrypted data.

[0087] In order to distinguish the dimensions corresponding to the homomorphically encrypted data, the user data dimensions of the homomorphically encrypted data may be marked. Specifically, at least two homomorphically encrypted data may be marked according to the order of the dimensions of the user data in the preset dimension list, and then combined according to the marking results of the at least two homomorphically encrypted data to obtain the combined homomorphically encrypted data.

[0088] Customer information can be encrypted through homomorphic encryption. Since the same homomorphic encryption scheme is used, the encrypted user data of each platform can be calculated without decryption. By decrypting the calculation results, the results of the calculation at the plaintext level can be obtained. In this way, data sharing can be achieved without leaking the user data of each platform.

[0089] Among them, the combination method can be an additive combination or a multiplicative combination. Taking the additive combination as an example, the encrypted homomorphically encrypted data are added to obtain the combined homomorphically encrypted data, and the combined homomorphically encrypted data are decrypted to obtain the combined user data.

[0090] S210, decrypt the combined homomorphically encrypted data to obtain combined user data, and input the combined user data into a pre-trained user portrait evaluation model to obtain a user portrait of the user.

[0091] Among them, the user portrait evaluation model can be a classification model for the user's product preferences. For example, the user portrait evaluation model converts the combined user data into the user's product preferences. The product preferences may include: financial product preferences, investment product preferences, and savings product preferences. The product preferences can represent the user's historical product selection and are long-term indicators. The classification model can also generate attention to different product preferences of users. The attention can represent the degree of importance that users attach to different types of products and are short-term indicators.

[0092] Specifically, a user profile of the user is obtained based on product preferences and the degree of attention paid to the product preferences.

[0093] S212, recommend products to users based on their user portraits.

[0094] Among them, product recommendations can be made to users based on the product preferences and attention in the user portrait.

[0095] Specifically, the product tendencies in the user portrait are compared with the preset product types in the benchmark platform. If the preset product types include products corresponding to the product tendencies of the user portrait, the user's attention is compared with the preset attention threshold. If the user's attention is greater than or equal to the preset attention threshold, the product type in the benchmark platform is recommended to the user.

[0096] If the user's attention is less than the preset attention threshold, the products are sorted in descending order according to the user's attention, and the product types with the highest ranking are pushed to the user.

[0097] In the above data processing method, if the user's identifier is an old user identifier, the baseline dimension of the user data is obtained according to the baseline platform, and the target dimension complementary to the baseline dimension is queried in the preset dimension list of the user data. The target dimension complementary to the baseline dimension is queried from the preset dimension list, so that the dimension of the user data is more comprehensive, which helps to improve the subsequent prediction accuracy of the user portrait. Determine the platform of the user data according to the target dimension, control the baseline platform and the target platform to perform homomorphic encryption processing on the user data respectively, and obtain at least two homomorphic encrypted data. The user data of different platforms are homomorphically encrypted respectively, so that the plaintext of the user data of different platforms is invisible to each other, thereby improving the security of the user data. Combine at least two homomorphic encrypted data, decrypt the combined homomorphic encrypted data, obtain the combined user data, and input the combined user data into the pre-trained user portrait evaluation model to obtain the user's user portrait. Based on homomorphic encryption, when the user data is available but not visible, the user data of multiple platforms are integrated to construct a more three-dimensional user portrait. Product recommendations are made to users based on their user portraits. This method queries the target dimension that is complementary to the baseline dimension, and then integrates user data from multiple platforms. This ensures the security of user data on each platform while building a more three-dimensional user portrait and improving the accuracy of product recommendations.

[0098] In one embodiment, the data processing method also includes: if the user identifier is a new user identifier, obtaining the default dimension of the user data, querying the associated platform associated with the default dimension of the user data in the preset dimension list of the user data, controlling the associated platform, performing homomorphic encryption processing on the user data in the associated platform to obtain associated homomorphic encrypted data, decrypting the associated homomorphic encrypted data and inputting it into a pre-trained user portrait evaluation model to obtain a user portrait of the user.

[0099] Among them, the default dimension can be the basic dimension of the user data in the benchmark platform. It can be understood that the default dimension corresponds to the basic information of the user, such as occupation, age, gender, name, personal identification number, etc.

[0100] After obtaining the default dimension of the user data, in the preset dimension list of the user data, query the associated platform associated with the default dimension of the user data. Specifically, you can query in the list according to the name, personal identification number, etc. in the default dimension to obtain the associated platform containing the name and personal identification number.

[0101] When the amount of user data on the current benchmark platform is too small, it is necessary to refer to the associated platform and process the user data through the associated platform to achieve prediction of the user portrait.

[0102] Control the associated platform, perform homomorphic encryption on the user data in the associated platform to obtain associated homomorphic encrypted data, decrypt the associated homomorphic encrypted data and input it into a pre-trained user portrait evaluation model to obtain a user portrait of the user.

[0103] In this embodiment, if the user's identifier is a new user identifier, the default dimension of the user data is obtained, and in the preset dimension list of the user data, the associated platform associated with the default dimension of the user data is queried, and then the user data is processed according to the associated platform to obtain a user portrait of the user. When the amount of user data on the current benchmark platform is too small, the prediction accuracy of the user portrait is improved.

[0104] In one embodiment, the data processing method further includes: in response to the selection operation of each platform for the key, obtaining the session key of the platform, wherein the session key includes a private key for homomorphic decryption and a public key for homomorphic encryption, configuring a public key for each platform, and configuring a private key for each platform according to the type of the platform.

[0105] Among them, a public key is configured for each platform, and the public keys of each platform are the same. The user data of each platform is encrypted according to the public key. Since the same homomorphic encryption method is used, the encrypted user data of each platform can be calculated without decryption.

[0106] According to the type of platform, a private key is configured for each platform. The type of platform includes a reference platform, a target platform, etc. The number of target platforms can be multiple, and different private keys are configured for each platform.

[0107] Specifically, before obtaining the platform's session key, you need to generate a session key. The key generation process usually involves a series of mathematical operations and random number generation to ensure that the generated key pair is secure and reliable. The key generation usually needs to meet mathematical properties, such as homomorphism and composability.

[0108] In this embodiment, by configuring session keys for each platform and using the same homomorphic encryption method, the encrypted user data of each platform can be calculated without decryption, making the user data available but invisible, thereby improving the security of user data sharing.

[0109] In one embodiment, at least two homomorphically encrypted data are combined, including: obtaining an arrangement order of user data dimensions according to a preset dimension list of user data, and combining at least two homomorphically encrypted data based on the arrangement order of user data dimensions.

[0110] Among them, in order to distinguish the dimensions of user data belonging to at least two homomorphically encrypted data, when at least two homomorphically encrypted data are combined, the order of each homomorphically encrypted data is obtained, so that the order of each homomorphically encrypted data is always maintained during the continuous homomorphic encryption process to avoid confusion between ciphertexts.

[0111] Specifically, the arrangement order of user data dimensions can be obtained according to a preset dimension list of user data, and the homomorphically encrypted data after the user data is homomorphically encrypted can be marked, so that at least two homomorphically encrypted data can be combined based on the arrangement order of user data dimensions and the marking order of homomorphically encrypted data.

[0112] In this embodiment, the order of each homomorphically encrypted data is obtained so that the order of each homomorphically encrypted data is always maintained during the continuous homomorphic encryption process, thereby avoiding confusion between ciphertexts.

[0113] In one embodiment, if the key indicators in the product recommendation results do not reach the preset values, the missing dimensions of the user data are queried in the preset dimension list of the user data, wherein the missing dimensions, the baseline dimensions, and the target dimensions do not overlap with each other. Based on the missing dimensions, user data identical to the missing dimensions are obtained from other platforms, and the user data of the missing dimensions, the baseline dimensions, and the target dimensions are homomorphically encrypted respectively to obtain at least three homomorphically encrypted data, and the at least three homomorphically encrypted data are combined.

[0114] Among them, the result of product recommendation can refer to the user's feedback on the recommended product, for example, the user's evaluation result of satisfaction with the recommended product. If the satisfaction evaluation result of the recommended product does not reach the preset value, it means that the effect of the recommended product has not met the platform's expectations and the product needs to be re-predicted and pushed.

[0115] Among them, the dimensions of user data can be divided into dimensions of the same level and dimensions of different levels. The dimension levels of user data can include a basic level and a mining level. The user data corresponding to the dimensions of the basic level can be directly obtained, and the user data corresponding to the dimensions of the mining level need to be obtained by algorithm mining of user behavior. Specifically, the baseline dimension and the target dimension belong to the same level, for example, they belong to the basic level and the mining level. It should be noted that for a certain dimension of user data, the level of the dimension to which the user data belongs under this dimension can be the basic level and / or the mining level. For example, the user data of the basic dimension (name, gender, age, etc.) belongs to the basic level dimension. For another example, the user data of the platform product dimension (platform search preference, financial product preference, etc.) belongs to the mining level dimension. For another example, the user data of the warning dimension (number of times the user defaults, user performance) belongs to both the basic level dimension and the mining level dimension.

[0116] When the user portrait predicted based on the user data of the same level of dimension is inaccurate, obtain the missing dimensions of the user data in different levels, that is, query the missing dimensions of the user data in the preset dimension list of the user data, and based on the missing dimensions, obtain the user data with the same missing dimensions from other platforms, and perform homomorphic encryption on the user data of the missing dimension, baseline dimension and target dimension respectively to obtain at least three homomorphic encrypted data, and combine the at least three homomorphic encrypted data.

[0117] In this embodiment, the accuracy of the user portrait can be improved by further completing the original user data from different levels by querying the missing dimensions.

[0118] In one embodiment, the dimensions of the user data include: at least one of: a basic dimension, a platform dimension, a user behavior dimension, a platform product dimension and an early warning dimension, and the method further includes: displaying the recommendation results of the product recommendation based on a preset dimension list of the user data.

[0119] Among them, user data in the basic dimension include: occupation, age, gender, name, etc.; user data in the platform dimension include: platform location, platform logo, real-time distance between the platform and the user, etc.; user behavior dimension include: user login, transaction, logout and other behaviors on the platform; user data in the platform product dimension include: user preference for various bank products, for example, platform search preference, financial product preference, etc.; user data in the early warning dimension include: user risk situation, risk situation between user and platform, etc.

[0120] Specifically, based on a preset dimension list of user data, the recommendation results of the product recommendation are displayed on a display device of a terminal or a reference platform.

[0121] In this embodiment, the product recommendation results are displayed to facilitate users to select the product recommendation results, thereby improving data interaction efficiency.

[0122] In one embodiment, as shown in FIG3 , a data processing method is provided, which is described by applying the method to a scenario where cross-bank user data is shared. The method includes:

[0123] S302, in response to the selection operation of each platform with respect to the key, obtaining the session key of the platform.

[0124] Among them, the session key includes a private key for homomorphic decryption and a public key for homomorphic encryption.

[0125] S304, configuring a public key for each platform.

[0126] S306: Configure a private key for each platform according to the type of the platform.

[0127] S308: If the user identifier is an old user identifier, a reference dimension of the user data is obtained through a reference platform, and a target dimension complementary to the reference dimension is searched in a preset dimension list of the user data.

[0128] Among them, if the user's identifier is a new user identifier, the default dimension of the user data is obtained, and in the preset dimension list of the user data, the associated platform associated with the default dimension of the user data is queried, the associated platform is controlled, and the user data in the associated platform is homomorphically encrypted to obtain associated homomorphically encrypted data, and the associated homomorphically encrypted data is decrypted and input into a pre-trained user portrait evaluation model to obtain a user portrait of the user.

[0129] S310, determining a target platform for user data according to a target dimension.

[0130] S312, controlling the reference platform and the target platform, and performing homomorphic encryption processing on the user data of the reference platform and the user data of the target platform respectively, to obtain at least two homomorphic encrypted data.

[0131] S314, obtaining an arrangement order of user data dimensions according to a preset dimension list of user data.

[0132] S316, combining at least two homomorphically encrypted data based on the arrangement order of the user data dimension.

[0133] S318, decrypt the combined homomorphically encrypted data to obtain combined user data, and input the combined user data into a pre-trained user portrait evaluation model to obtain a user portrait of the user.

[0134] S320, recommend products to users based on their user portraits.

[0135] S322: If the key indicator in the result of the product recommendation does not reach the preset value, query the missing dimension of the user data in the preset dimension list of the user data.

[0136] Among them, the missing dimension, benchmark dimension and target dimension do not overlap with each other.

[0137] S324, based on the vacant dimension, obtaining user data having the same vacant dimension from other platforms.

[0138] S326, perform homomorphic encryption processing on the user data of the missing dimension, the reference dimension and the target dimension respectively to obtain at least three homomorphic encrypted data.

[0139] S328, combine at least three homomorphically encrypted data and return to S318.

[0140] In this embodiment, if the user's identifier is an old user identifier, the base dimension of the user data is obtained according to the base platform, and the target dimension complementary to the base dimension is queried in the preset dimension list of the user data. The target dimension complementary to the base dimension is queried from the preset dimension list, so that the dimension of the user data is more comprehensive, which helps to improve the accuracy of subsequent prediction of the user portrait. The platform of the user data is determined according to the target dimension, and the base platform and the target platform are controlled to perform homomorphic encryption processing on the user data respectively to obtain at least two homomorphic encrypted data. The user data of different platforms are homomorphically encrypted respectively, so that the plaintext of the user data of different platforms is mutually invisible, thereby improving the security of the user data. At least two homomorphic encrypted data are combined, and the combined homomorphic encrypted data are decrypted to obtain the combined user data, and the combined user data is input into the pre-trained user portrait evaluation model to obtain the user portrait of the user. Based on homomorphic encryption, when the user data is available but not visible, the user data of multiple platforms are integrated to construct a more three-dimensional user portrait. According to the user portrait, the user is recommended a product to the user. This method, by querying the target dimension complementary to the base dimension, further integrates the user data of multiple platforms, ensures the security between the user data of each platform, and constructs a more three-dimensional user portrait, thereby improving the accuracy of product recommendation.

[0141] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0142] Based on the same inventive concept, the embodiment of the present application also provides a data processing device for implementing the data processing method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in the one or more data processing device embodiments provided below can refer to the limitations on the data processing method above, and will not be repeated here.

[0143] In one embodiment, as shown in FIG. 4 , a data processing device is provided, including: an acquisition module 402, a determination module 404, an encryption module 406, a combination module 408, a decryption module 410, and a recommendation module 412, wherein:

[0144] The acquisition module 402 is used to acquire a reference dimension of the user data through a reference platform if the user identifier is an old user identifier, and search for a target dimension complementary to the reference dimension in a preset dimension list of the user data;

[0145] A determination module 404, configured to determine a target platform for user data according to a target dimension;

[0146] The encryption module 406 is used to control the reference platform and the target platform to perform homomorphic encryption processing on the user data of the reference platform and the user data of the target platform respectively to obtain at least two homomorphic encrypted data;

[0147] A combining module 408, configured to combine at least two homomorphically encrypted data;

[0148] A decryption module 410 is used to decrypt the combined homomorphically encrypted data to obtain combined user data, and input the combined user data into a pre-trained user profile evaluation model to obtain a user profile of the user;

[0149] The recommendation module 412 is used to recommend products to users based on their user portraits.

[0150] In one of the embodiments, the data processing device also includes: an association module, which is used to obtain the default dimension of the user data if the user identifier is a new user identifier, query the association platform associated with the default dimension of the user data in the preset dimension list of the user data, control the association platform, perform homomorphic encryption on the user data in the association platform to obtain associated homomorphic encrypted data, decrypt the associated homomorphic encrypted data and input it into a pre-trained user portrait evaluation model to obtain a user portrait of the user.

[0151] In one of the embodiments, the data processing device also includes: a configuration module, used to obtain the session key of the platform in response to the selection operation of each platform for the key, wherein the session key includes a private key for homomorphic decryption and a public key for homomorphic encryption, configure the public key for each platform, and configure the private key for each platform according to the type of the platform.

[0152] In one of the embodiments, the combining module 408 is further used to obtain an arrangement order of user data dimensions according to a preset dimension list of user data, and combine at least two homomorphically encrypted data based on the arrangement order of user data dimensions.

[0153] In one of the embodiments, the data processing device also includes: a missing module, which is used to query the missing dimensions of the user data in the preset dimension list of the user data if the key indicators in the product recommendation results do not reach the preset values, wherein the missing dimensions, the baseline dimensions and the target dimensions do not overlap with each other, and based on the missing dimensions, obtain user data with the same missing dimensions from other platforms, and perform homomorphic encryption processing on the user data of the missing dimensions, the baseline dimensions and the target dimensions respectively to obtain at least three homomorphic encrypted data, and combine the at least three homomorphic encrypted data.

[0154] In one of the embodiments, the data processing device also includes: the dimensions of user data include: basic dimension, platform dimension, user behavior dimension, platform product dimension and at least one of warning dimension, and the method also includes: a display module, which is used to display the recommendation results of product recommendations based on a preset dimension list of user data.

[0155] Each module in the above data processing device can be implemented in whole or in part by software, hardware or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to each module.

[0156] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be shown in FIG5. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store user portrait data. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a data processing method is implemented.

[0157] Those skilled in the art will understand that the structure shown in FIG. 5 is merely a block diagram of a partial structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different arrangement of components.

[0158] In one embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the following steps are implemented:

[0159] If the user identifier is an old user identifier, a benchmark dimension of the user data is obtained through the benchmark platform, and a target dimension complementary to the benchmark dimension is searched in a preset dimension list of the user data;

[0160] Determine the target platform for user data based on the target dimension;

[0161] Controlling the reference platform and the target platform, and performing homomorphic encryption processing on the user data of the reference platform and the user data of the target platform respectively, to obtain at least two homomorphic encrypted data;

[0162] Combine at least two homomorphically encrypted data;

[0163] Decrypt the combined homomorphic encrypted data to obtain the combined user data, and input the combined user data into the pre-trained user portrait evaluation model to obtain the user portrait of the user;

[0164] Recommend products to users based on their user portraits.

[0165] In one embodiment, the processor further implements the following steps when executing the computer program: if the user identifier is a new user identifier, obtain the default dimension of the user data, query the associated platform associated with the default dimension of the user data in the preset dimension list of the user data, control the associated platform, perform homomorphic encryption processing on the user data in the associated platform to obtain associated homomorphic encrypted data, decrypt the associated homomorphic encrypted data and input it into a pre-trained user portrait evaluation model to obtain a user portrait of the user.

[0166] In one embodiment, when the processor executes the computer program, the following steps are also implemented: in response to the selection operation of each platform for the key, the session key of the platform is obtained, wherein the session key includes a private key for homomorphic decryption and a public key for homomorphic encryption, a public key is configured for each platform, and a private key is configured for each platform according to the type of the platform.

[0167] In one embodiment, when the processor executes the computer program, it also implements the following steps: obtaining an arrangement order of user data dimensions according to a preset dimension list of user data, and combining at least two homomorphically encrypted data based on the arrangement order of user data dimensions.

[0168] In one embodiment, the processor further implements the following steps when executing the computer program: if the key indicator in the product recommendation result does not reach the preset value, then query the missing dimension of the user data in the preset dimension list of the user data, wherein the missing dimension, the baseline dimension and the target dimension do not overlap with each other, and based on the missing dimension, obtain the user data with the same missing dimension from other platforms, perform homomorphic encryption processing on the user data of the missing dimension, the baseline dimension and the target dimension respectively, obtain at least three homomorphic encrypted data, and combine the at least three homomorphic encrypted data.

[0169] In one embodiment, when the processor executes the computer program, the following steps are also implemented: based on a preset dimension list of user data, the recommendation results of the product recommendation are displayed.

[0170] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

[0171] If the user identifier is an old user identifier, a benchmark dimension of the user data is obtained through the benchmark platform, and a target dimension complementary to the benchmark dimension is searched in a preset dimension list of the user data;

[0172] Determine the target platform for user data based on the target dimension;

[0173] Controlling the reference platform and the target platform, and performing homomorphic encryption processing on the user data of the reference platform and the user data of the target platform respectively, to obtain at least two homomorphic encrypted data;

[0174] Combine at least two homomorphically encrypted data;

[0175] Decrypt the combined homomorphic encrypted data to obtain the combined user data, and input the combined user data into the pre-trained user portrait evaluation model to obtain the user portrait of the user;

[0176] Recommend products to users based on their user portraits.

[0177] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: if the user identifier is a new user identifier, the default dimension of the user data is obtained, in the preset dimension list of the user data, the associated platform associated with the default dimension of the user data is queried, the associated platform is controlled, the user data in the associated platform is homomorphically encrypted to obtain associated homomorphically encrypted data, the associated homomorphically encrypted data is decrypted and input into a pre-trained user portrait evaluation model to obtain a user portrait of the user.

[0178] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: in response to the selection operation of each platform for the key, the session key of the platform is obtained, wherein the session key includes a private key for homomorphic decryption and a public key for homomorphic encryption, a public key is configured for each platform, and a private key is configured for each platform according to the type of the platform.

[0179] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: according to a preset dimension list of user data, an arrangement order of user data dimensions is obtained, and based on the arrangement order of user data dimensions, at least two homomorphically encrypted data are combined.

[0180] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: if the key indicator in the product recommendation result does not reach the preset value, then query the missing dimension of the user data in the preset dimension list of the user data, wherein the missing dimension, the baseline dimension and the target dimension do not overlap with each other, based on the missing dimension, obtain the user data with the same missing dimension from other platforms, perform homomorphic encryption processing on the user data of the missing dimension, the baseline dimension and the target dimension respectively, obtain at least three homomorphic encrypted data, and combine the at least three homomorphic encrypted data.

[0181] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: based on a preset dimension list of user data, the recommendation results of the product recommendation are displayed.

[0182] In one embodiment, a computer program product is provided, comprising a computer program, which, when executed by a processor, implements the following steps:

[0183] If the user identifier is an old user identifier, a benchmark dimension of the user data is obtained through the benchmark platform, and a target dimension complementary to the benchmark dimension is searched in a preset dimension list of the user data;

[0184] Determine the target platform for user data based on the target dimension;

[0185] Controlling the reference platform and the target platform, and performing homomorphic encryption processing on the user data of the reference platform and the user data of the target platform respectively, to obtain at least two homomorphic encrypted data;

[0186] Combine at least two homomorphically encrypted data;

[0187] Decrypt the combined homomorphic encrypted data to obtain the combined user data, and input the combined user data into the pre-trained user portrait evaluation model to obtain the user portrait of the user;

[0188] Recommend products to users based on their user portraits.

[0189] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: if the user identifier is a new user identifier, the default dimension of the user data is obtained, in the preset dimension list of the user data, the associated platform associated with the default dimension of the user data is queried, the associated platform is controlled, the user data in the associated platform is homomorphically encrypted to obtain associated homomorphically encrypted data, the associated homomorphically encrypted data is decrypted and input into a pre-trained user portrait evaluation model to obtain a user portrait of the user.

[0190] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: in response to the selection operation of each platform for the key, the session key of the platform is obtained, wherein the session key includes a private key for homomorphic decryption and a public key for homomorphic encryption, a public key is configured for each platform, and a private key is configured for each platform according to the type of the platform.

[0191] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: according to a preset dimension list of user data, an arrangement order of user data dimensions is obtained, and based on the arrangement order of user data dimensions, at least two homomorphically encrypted data are combined.

[0192] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: if the key indicator in the product recommendation result does not reach the preset value, then query the missing dimension of the user data in the preset dimension list of the user data, wherein the missing dimension, the baseline dimension and the target dimension do not overlap with each other, based on the missing dimension, obtain the user data with the same missing dimension from other platforms, perform homomorphic encryption processing on the user data of the missing dimension, the baseline dimension and the target dimension respectively, obtain at least three homomorphic encrypted data, and combine the at least three homomorphic encrypted data.

[0193] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: based on a preset dimension list of user data, the recommendation results of the product recommendation are displayed.

[0194] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0195] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0196] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0197] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A data processing method, characterized in that: The method comprises: If the user identifier is an old user identifier, a reference dimension of the user data is obtained according to the reference platform, and a target dimension complementary to the reference dimension is searched in a preset dimension list of the user data; Determining a target platform for user data according to the target dimension; Controlling the reference platform and the target platform to perform homomorphic encryption processing on user data of the reference platform and user data of the target platform respectively, to obtain at least two homomorphic encrypted data; Combining the at least two homomorphically encrypted data; Decrypting the combined homomorphically encrypted data to obtain combined user data, and inputting the combined user data into a pre-trained user profile evaluation model to obtain a user profile of the user; Recommend products to the user based on the user portrait.

2. The method according to claim 1, characterized in that The method further comprises: If the user's ID is a new user ID, the default dimension of the user data is obtained; In the preset dimension list of the user data, query the associated platform associated with the default dimension of the user data; Controlling the associated platform to perform homomorphic encryption processing on user data in the associated platform to obtain associated homomorphic encrypted data; The associated homomorphically encrypted data is decrypted and input into a pre-trained user portrait evaluation model to obtain a user portrait of the user.

3. The method according to claim 2, characterized in that The method further comprises: In response to the selection operation of each platform for a key, obtaining a session key of the platform, wherein the session key includes a private key for homomorphic decryption and a public key for homomorphic encryption; Configuring the public key for each of the platforms; The private key is configured for each platform according to the type of the platform.

4. The method according to claim 1, characterized in that: The combining the at least two homomorphically encrypted data comprises: Obtaining an arrangement order of the user data dimensions according to a preset dimension list of the user data; Based on the arrangement order of the user data dimensions, the at least two homomorphically encrypted data are combined.

5. The method according to claim 1, characterized in that The method further comprises: If the key indicator in the result of the product recommendation does not reach the preset value, querying the missing dimension of the user data in the preset dimension list of the user data, wherein the missing dimension, the reference dimension and the target dimension do not overlap with each other; Based on the missing dimension, obtaining user data having the same missing dimension from other platforms; Performing homomorphic encryption processing on the user data of the missing dimension, the reference dimension, and the target dimension respectively to obtain at least three homomorphic encrypted data; The at least three homomorphically encrypted data are combined.

6. The method according to claim 1, characterized in that The dimension of the user data includes: at least one of a basic dimension, a platform dimension, a user behavior dimension, a platform product dimension and an early warning dimension, and the method further includes: Based on the preset dimension list of the user data, the recommendation results of the product recommendation are displayed.

7. A data processing device, characterized in that: The device comprises: an acquisition module, configured to acquire a reference dimension of user data according to a reference platform if the user identifier is an old user identifier, and search a preset dimension list of user data for a target dimension complementary to the reference dimension; A determination module, used to determine a target platform of user data according to the target dimension; An encryption module, used to control the reference platform and the target platform, and perform homomorphic encryption processing on the user data of the reference platform and the user data of the target platform respectively, to obtain at least two homomorphic encrypted data; A combining module, used for combining the at least two homomorphically encrypted data; A decryption module, used to decrypt the combined homomorphically encrypted data to obtain combined user data, and input the combined user data into a pre-trained user portrait evaluation model to obtain a user portrait of the user; A recommendation module is used to recommend products to the user based on the user portrait.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. 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 6 are implemented.

10. 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 6 are implemented.

Images