User information updating method and device, electronic equipment and medium

Abstract

The invention provides a user information updating method and device, equipment and a medium, which can be applied to the fields of big data processing and financial science and technology. The method comprises the following steps: acquiring user data of a plurality of users in different channels at the t-th moment; for each user, performing vector conversion processing on multiple pieces of attribute data in the user data to obtain a tth user attribute vector corresponding to the user data; according to the tth user attribute vector and the (t-1) th user attribute vector, information similarity is generated, and the (t-1) th user attribute vector is generated according to the user data stored in the target database at the (t-1) th moment; under the condition that the information similarity meets a preset similarity threshold value, the user data is updated to a user information graph in a target database through a graph neural network, an updated user information graph is obtained, and the user information graph comprises user nodes representing different user data and connecting edges representing the relation between the different user nodes.

Description

Technical Field

[0001] The present disclosure relates to the field of big data processing and financial technology, and specifically to a user information updating method, device, electronic device, computer storage medium and program product. Background Art

[0002] Bank user information is a valuable asset in the financial field, including personal identity, transaction records, etc. This information is not only used for identity verification, but also supports banks in providing personalized financial services, optimizing customer experience, and assessing loan risks based on customer credit.

[0003] However, in traditional banking, updating customer information usually involves cumbersome processes, which may lead to problems such as inaccurate information, long update cycles, and poor user experience. Therefore, a more efficient way to manage and update customer information is needed. Summary of the invention

[0004] In view of the above problems, the present disclosure provides a user information updating method, device, electronic device, computer storage medium and program product.

[0005] According to a first aspect of the present disclosure, a user information updating method is provided, comprising:

[0006] At time t, user data of multiple users in different channels are obtained, wherein the user data includes multiple attribute data representing information of different users;

[0007] For each user, multiple attribute data in the user data are subjected to vector conversion processing to obtain the t-th user attribute vector corresponding to the user data;

[0008] Generate information similarity based on the t-th user attribute vector and the t-1-th user attribute vector, wherein the t-1-th user attribute vector is generated based on the user data at the t-1-th time stored in the target database;

[0009] When the information similarity meets the preset similarity threshold, the graph neural network is used to update the user data to the user information graph in the target database to obtain an updated user information graph, wherein the user information graph includes user nodes representing different user data and connection edges representing the relationship between different user nodes.

[0010] According to an embodiment of the present disclosure, generating information similarity according to the t-th user attribute vector and the t-1-th user attribute vector includes:

[0011] Obtain the user data of the user at time t-1 from the target database;

[0012] Perform vector conversion processing on the user data at time t-1 to obtain the t-1 user attribute vector;

[0013] The cosine similarity function is used to process the t-th user attribute vector and the t-1-th user attribute vector to generate information similarity.

[0014] According to an embodiment of the present disclosure, a plurality of attribute data in the user data is subjected to vector conversion processing to obtain a t-th user attribute vector corresponding to the user data, including:

[0015] For each attribute data, preprocessing the attribute data to obtain preprocessed attribute data;

[0016] Using a preset word vector model to process each preprocessed attribute data to obtain multiple attribute vectors;

[0017] Multiple attribute vectors are averaged to obtain the t-th user attribute vector.

[0018] According to an embodiment of the present disclosure, before performing the vector conversion process, the method further includes:

[0019] Perform data cleaning on multiple attribute data in the user data to obtain cleaned user data;

[0020] The attribute data of each category of the multiple cleaned user data of the user are standardized to obtain multiple standardized attribute data, so as to perform vector conversion processing using the standardized attribute data.

[0021] According to an embodiment of the present disclosure, multiple attribute data in user data are cleansed to obtain cleaned user data, including:

[0022] For any type of attribute data, when the attribute data is of numerical type, a visualization graph is drawn based on the multiple attribute data;

[0023] When the difference between the extreme value and the average value in the visualization diagram is greater than the first preset threshold, the attribute data corresponding to the extreme value is deleted;

[0024] When the difference between the extreme value and the average value in the visualization graph is between the first preset threshold and the second preset threshold, the extreme value is replaced with a filling value to obtain replaced attribute data, wherein the filling value is determined based on multiple attribute data.

[0025] According to an embodiment of the present disclosure, each category of attribute data of a plurality of cleaned user data of a user is standardized to obtain a plurality of standardized attribute data, including:

[0026] For a plurality of attribute data of any category, determining a maximum value and a minimum value of the plurality of attribute data;

[0027] For each attribute data, the standardized attribute data is obtained according to the attribute data, the maximum value of the data and the minimum value of the data.

[0028] According to an embodiment of the present disclosure, before performing the vector conversion process, the method further includes:

[0029] For each user, compare the user ID of the user at time t with the user IDs of multiple user data stored at time t-1 in the target database to obtain a comparison result;

[0030] If the comparison result shows that the user identifier at time t does not exist in the target database, the user data is stored in the target database;

[0031] When the comparison result indicates that the user identifier of the user at the tth moment exists in the target database, vector conversion processing is performed on the user data of the user.

[0032] According to an embodiment of the present disclosure, the user data is updated to the user information graph in the target database using a graph neural network to obtain an updated user information graph, including:

[0033] Constructing a verification information graph according to multiple user data of multiple users, wherein the verification information graph includes multiple user nodes and multiple verification edges, and the verification edge represents the relationship between two connected user nodes;

[0034] Embed the nodes and verification edges in the verification information graph into the initial graph neural network to obtain a first graph neural network, wherein the first graph neural network includes a plurality of network nodes, and the network nodes represent the relationship between different users;

[0035] Performing graph convolution operations on the first graph neural network and the second graph neural network respectively to obtain a plurality of first feature vectors corresponding to the first graph neural network and a plurality of second feature vectors corresponding to the second graph neural network, wherein the second graph neural network is constructed according to a user information graph in a target database;

[0036] For any user, the user information graph is updated according to the first feature vector and the second feature vector corresponding to the user to obtain an updated user information graph.

[0037] According to an embodiment of the present disclosure, the user information graph is updated according to the first feature vector and the second feature vector corresponding to the user to obtain an updated user information graph, including:

[0038] In the case where there is a difference between the first eigenvector and the second eigenvector, the network node corresponding to the second eigenvector in the second graph neural network is updated using the first eigenvector to obtain an updated second graph neural network;

[0039] For the first graph neural network, when the difference value between the first feature vector of the network node where the user is located and the first feature vector of the network node where the neighbor node associated with the user is located is greater than a preset difference threshold, the first feature vectors corresponding to the network node and the neighbor node are used to update the network node and the neighbor node in the second graph neural network respectively to obtain an updated second graph neural network, wherein the updated second graph neural network represents an updated user information graph.

[0040] According to an embodiment of the present disclosure, the user information updating method further includes:

[0041] The updated user information map is encrypted to obtain an encrypted user information map, so that an external device can access the encrypted user information map.

[0042] According to an embodiment of the present disclosure, the updated user information graph includes a plurality of updated user data corresponding to each user;

[0043] The updated user information graph is encrypted to obtain an encrypted user information graph, including:

[0044] For any user, generate N round keys used for multi-round encryption based on the randomly generated initial key;

[0045] In the i-th round of encryption, an XOR operation is performed with the i-th round key and the user data updated by the multiple input data to obtain multiple initial encrypted data, wherein the input data includes the updated user data or the i-1th encrypted data;

[0046] Based on a preset substitution cipher rule, generating a plurality of substitution cipher information according to a plurality of initial encrypted data;

[0047] Performing a password shift operation on a plurality of substitute password information to obtain intermediate password information, or performing a matrix change process on a password matrix composed of a plurality of substitute password information to obtain intermediate password information;

[0048] Generate the i-th encrypted data according to the intermediate password information and the i-th round key;

[0049] In the case of i=N, the i-th encrypted data is determined as the encrypted user data of the user, wherein the encrypted user information graph includes the encrypted user data of a plurality of users.

[0050] A second aspect of the present disclosure provides a user information updating device, comprising:

[0051] An acquisition module, used to acquire user data of multiple users in different channels at time t, wherein the user data includes multiple attribute data representing different user information;

[0052] A processing module, used for performing vector conversion processing on multiple attribute data in the user data for each user, to obtain a t-th user attribute vector corresponding to the user data;

[0053] A generating module, used for generating information similarity according to the t-th user attribute vector and the t-1-th user attribute vector, wherein the t-1-th user attribute vector is generated according to the user data at the t-1-th time stored in the target database;

[0054] The updating module is used to update the user data to the user information graph in the target database by using the graph neural network when the information similarity meets the preset similarity threshold, so as to obtain the updated user information graph, wherein the user information graph includes user nodes representing different user data and connection edges representing the relationship between different user nodes.

[0055] A third aspect of the present disclosure provides an electronic device, including:

[0056] one or more processors;

[0057] a storage device for storing one or more programs,

[0058] When one or more programs are executed by one or more processors, the one or more processors execute the above-mentioned user information updating method.

[0059] A fourth aspect of the present disclosure provides a computer-readable storage medium having executable instructions stored thereon, which, when executed by a processor, causes the processor to execute the above-mentioned user information updating method.

[0060] A fifth aspect of the present disclosure provides a computer program product, including a computer program, which implements the above-mentioned user information updating method when executed by a processor.

[0061] According to the user information update method, device, equipment, medium and program product provided by the present disclosure, multiple attribute data in each user data are vectorized to obtain a user attribute vector, and then the information similarity between the user attribute vector and the user attribute vector in the target database is calculated to determine whether an update is required, thereby using a graph neural network to update the user data to the user information graph in the target database to obtain an updated user information graph. The graph neural network is used to quickly and accurately update and maintain the personal information of bank customers, store the updated information, and achieve the purpose of real-time dynamic updating of in-bank customer information while protecting customer information security, solving the cumbersome problem of updating customer information. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] 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:

[0063] Figure 1 The following schematically shows an application scenario diagram of the user information updating method according to an embodiment of the present disclosure;

[0064] Figure 2 A flowchart of a method for updating user information according to an embodiment of the present disclosure is schematically shown;

[0065] Figure 3 A flowchart of a method for updating a user information map according to an embodiment of the present disclosure is schematically shown;

[0066] Figure 4 A schematic diagram showing a structural block diagram of a user information updating device according to an embodiment of the present disclosure; and

[0067] Figure 5 A block diagram of an electronic device suitable for implementing a user information updating method according to an embodiment of the present disclosure is schematically shown. DETAILED DESCRIPTION

[0068] 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.

[0069] The terms used herein are only for describing specific embodiments and are not intended to limit the present disclosure. The terms "comprise", "include", etc. used herein indicate the existence of features, steps, operations and / or components, but do not exclude the existence or addition of one or more other features, steps, operations or components.

[0070] All terms (including technical and scientific terms) used herein have the meanings commonly understood by those skilled in the art, unless otherwise defined. It should be noted that the terms used herein should be interpreted as having a meaning consistent with the context of this specification, and should not be interpreted in an idealized or overly rigid manner.

[0071] When using expressions such as "at least one of A, B, and C, etc.", they should generally be interpreted according to the meaning of the expression commonly understood by those skilled in the art (for example, "a system having at least one of A, B, and C" should include but is not limited to a system having A alone, B alone, C alone, A and B, A and C, B and C, and / or A, B, C, etc.).

[0072] In the process of implementing the present disclosure, it is found that bank user information is not only used for identity verification, but also supports banks in providing personalized financial services, etc. However, in traditional banking services, the update of customer information usually involves cumbersome processes, which may lead to inaccurate information, long update cycles, poor user experience, etc., so a more effective user information update method is urgently needed.

[0073] In view of this, the embodiments of the present disclosure provide a user information update method, device, electronic device and medium. The method includes: obtaining user data of multiple users under different channels at the tth time; for each user, performing vector conversion processing on multiple attribute data in the user data to obtain the tth user attribute vector corresponding to the user data; generating information similarity based on the tth user attribute vector and the t-1th user attribute vector, wherein the t-1th user attribute vector is generated based on the user data at the t-1th time stored in the target database; when the information similarity meets the preset similarity threshold, using the graph neural network to update the user data to the user information graph in the target database, and obtaining an updated user information graph, wherein the user information graph includes user nodes representing different user data and connection edges representing the relationship between different user nodes.

[0074] It should be noted that the user information update method and user information update device provided by the present disclosure can be used in the financial field, such as banks and other financial institutions, and can also be used in any field other than the financial field. Therefore, the application field of the user information update method and user information update device provided by the present disclosure is not limited.

[0075] In the technical solution of the present invention, the user information (including but not limited to user personal 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 relevant data comply with the relevant laws, regulations and standards of relevant countries and regions, 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.

[0076] Figure 1 The application scenario diagram of the user information updating method according to the embodiment of the present disclosure is schematically shown.

[0077] like Figure 1 As shown, the application scenario 100 according to this embodiment may include a first terminal device 101, a second terminal device 102, a third terminal device 103, a network 104 and a server 105. The network 104 is used to provide a medium for a communication link between the terminal devices 101, 102, 103 and the server 105. The network 104 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0078] Users can use terminal devices 101, 102, 103 to interact with server 105 through network 104 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 101, 102, 103, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social platform software, etc. (only examples).

[0079] The terminal devices 101 , 102 , and 103 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, and desktop computers.

[0080] The server 105 may be a server that provides various services, such as a background management server (only an example) that provides support for websites browsed by users using the terminal devices 101, 102, and 103. The background management server may analyze and process the received data such as user requests, and feed back the processing results (such as web pages, information, or data obtained or generated according to user requests) to the terminal device.

[0081] It should be noted that the user information updating method provided in the embodiment of the present disclosure can generally be executed by the server 105. Accordingly, the user information updating device provided in the embodiment of the present disclosure can generally be set in the server 105. The user information updating method provided in the embodiment of the present disclosure can also be executed by a server or server cluster that is different from the server 105 and can communicate with the terminal devices 101, 102, 103 and / or the server 105. Accordingly, the user information updating device provided in the embodiment of the present disclosure can also be set in a server or server cluster that is different from the server 105 and can communicate with the terminal devices 101, 102, 103 and / or the server 105.

[0082] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0083] The following will be based on Figure 1 The scene described by Figure 2-3 The user information updating method of the disclosed embodiment is described in detail.

[0084] Figure 2 The flowchart of the user information updating method according to the embodiment of the present disclosure is schematically shown.

[0085] like Figure 2 As shown, the method 200 includes operations S210 to S240.

[0086] In operation S210, user data of a plurality of users in different channels is obtained at time t, wherein the user data includes a plurality of attribute data representing information of different users.

[0087] In operation S220, for each user, a plurality of attribute data in the user data is subjected to vector conversion processing to obtain a t-th user attribute vector corresponding to the user data.

[0088] In operation S230, information similarity is generated based on the t-th user attribute vector and the t-1-th user attribute vector, wherein the t-1-th user attribute vector is generated based on user data at the t-1-th time stored in the target database.

[0089] In operation S240, when the information similarity satisfies a preset similarity threshold, the user data is updated to a user information graph in a target database using a graph neural network to obtain an updated user information graph, wherein the user information graph includes user nodes representing different user data and connection edges representing the relationship between different user nodes.

[0090] According to an embodiment of the present disclosure, user data of multiple users are obtained from different channels, that is, user information is obtained from all channels, and all channels include account opening application, transaction records, online banking or mobile banking, customer surveys, marketing activities, third-party data suppliers, partners, social media and other channels.

[0091] According to an embodiment of the present disclosure, the attribute data may be personal information such as name, ID number, contact information, or user transaction information such as deposits, withdrawals, transfers, loans, etc.

[0092] According to an embodiment of the present disclosure, a plurality of attribute data corresponding to each user is subjected to vector conversion processing to obtain a user attribute vector corresponding to each user.

[0093] According to an embodiment of the present disclosure, the t-1th user attribute vector represents a user attribute vector generated based on the user data of the user last updated or stored in the target database.

[0094] According to an embodiment of the present disclosure, the tth user attribute vector represents a user attribute vector generated based on the user data of the current user.

[0095] According to an embodiment of the present disclosure, information similarity is calculated based on the current user attribute vector and the user attribute vector in the target database. The information similarity is used to characterize the textual similarity between the current user information and the user data in the target database.

[0096] According to an embodiment of the present disclosure, the user information graph may include user nodes of different user data and connection edges representing the relationship between different user nodes. The relationship between different user nodes may include user relationships, transaction patterns between users, etc.

[0097] According to an embodiment of the present disclosure, vector conversion processing is performed on multiple attribute data in each user data to obtain a user attribute vector, and then the information similarity between this user attribute vector and the user attribute vector in the target database is calculated to determine whether it needs to be updated, so as to use the graph neural network to update the user data to the user information graph in the target database to obtain an updated user information graph.

[0098] According to the embodiments of the present disclosure, through the combined application of graph neural network algorithms, the personal information of bank customers can be quickly and accurately updated and maintained, and the updated information can be stored, so as to achieve the purpose of real-time dynamic updating of in-bank customer information while protecting the security of customer information, thus solving the cumbersome problem of updating customer information.

[0099] According to an embodiment of the present disclosure, generating information similarity according to the t-th user attribute vector and the t-1-th user attribute vector includes:

[0100] Obtain the user data of the user at time t-1 from the target database;

[0101] Perform vector conversion processing on the user data at time t-1 to obtain the t-1 user attribute vector;

[0102] The cosine similarity function is used to process the t-th user attribute vector and the t-1-th user attribute vector to generate information similarity.

[0103] According to an embodiment of the present disclosure, the user data of the user is obtained from the target database and vector conversion is performed to obtain a user attribute vector generated by the user data of the user last updated or stored in the target database, that is, the t-1th user attribute vector.

[0104] According to an embodiment of the present disclosure, a cosine similarity function is used to calculate the information similarity between a current user attribute vector and a user attribute vector generated based on user data in a target database.

[0105] According to an embodiment of the present disclosure, information similarity can be used to determine whether an old customer needs to be updated. When the old customer's information is consistent with that retained in the target database last time, no update is required; otherwise, an update is required.

[0106] According to an embodiment of the present disclosure, when the information similarity is greater than the similarity threshold, it is considered that the information of the old customer is consistent with the last retained in the target database and does not need to be updated; when the information similarity is less than or equal to the similarity threshold, it needs to be updated.

[0107] According to an embodiment of the present disclosure, it is possible to further determine whether the user data needs to be updated based on the information similarity result between the current user attribute vector and the user attribute vector generated based on the user data in the target database.

[0108] According to an embodiment of the present disclosure, a plurality of attribute data in the user data is subjected to vector conversion processing to obtain a t-th user attribute vector corresponding to the user data, including:

[0109] For each attribute data, preprocessing the attribute data to obtain preprocessed attribute data;

[0110] Using a preset word vector model to process each preprocessed attribute data to obtain multiple attribute vectors;

[0111] Multiple attribute vectors are averaged to obtain the t-th user attribute vector.

[0112] According to an embodiment of the present disclosure, preprocessing is performed on a plurality of attribute data in the t-th user data and a plurality of attribute data in the t-1-th user data in the target database to obtain preprocessed attribute data.

[0113] According to an embodiment of the present disclosure, preprocessing includes operation steps such as removing stop words and extracting stems.

[0114] According to an embodiment of the present disclosure, each preprocessed attribute data is converted into a word vector representation using a preset word vector model (Word2Vec) to obtain multiple attribute vectors.

[0115] According to an embodiment of the present disclosure, an average process is performed on multiple attribute vectors of the user to obtain the current user attribute vector of the user and the user attribute vector in the target database.

[0116] According to an embodiment of the present disclosure, before performing the vector conversion process, the method further includes:

[0117] Perform data cleaning on multiple attribute data in the user data to obtain cleaned user data;

[0118] The attribute data of each category of the multiple cleaned user data of the user are standardized to obtain multiple standardized attribute data, so as to perform vector conversion processing using the standardized attribute data.

[0119] According to an embodiment of the present disclosure, before vector conversion processing, for each user, multiple attribute data in the user data are cleaned to obtain cleaned user data, and then each category of attribute data is standardized to obtain multiple standardized attribute data.

[0120] According to the embodiments of the present disclosure, the formats and standards of information collected by various channels are different, and it is difficult to achieve unified updates. The acquired user data is first cleaned and then standardized to ensure that the used data maintains consistent standards between different features, so that a unified format facilitates unified standard updates of data from different channels.

[0121] According to an embodiment of the present disclosure, multiple attribute data in user data are cleansed to obtain cleaned user data, including:

[0122] For any type of attribute data, when the attribute data is of numerical type, a visualization graph is drawn based on the multiple attribute data;

[0123] When the difference between the extreme value and the average value in the visualization diagram is greater than the first preset threshold, the attribute data corresponding to the extreme value is deleted;

[0124] When the difference between the extreme value and the average value in the visualization graph is between the first preset threshold and the second preset threshold, the extreme value is replaced with a filling value to obtain replaced attribute data, wherein the filling value is determined based on multiple attribute data.

[0125] According to an embodiment of the present disclosure, for different categories of attribute data, it is necessary to identify outliers. When the attribute data is of numerical type, a visual graph is drawn to identify outliers, such as a box plot, a scatter plot, etc.

[0126] According to an embodiment of the present disclosure, a statistical method is used to calculate the difference between the extreme value and the average value in the visualization diagram. When the difference is greater than a first preset threshold, it is regarded as an extreme outlier, and the attribute data corresponding to the extreme value is deleted, that is, the outlier is deleted.

[0127] According to an embodiment of the present disclosure, when the attribute data is of text type, abnormal value deletion processing is also performed, such as null values, garbled characters, worthless data, etc.

[0128] According to an embodiment of the present disclosure, when the difference between the extreme value and the average value in the visualization graph is within the first preset threshold and the second preset threshold, it is considered as a slight abnormal value, and the extreme value is replaced with a suitable filling value to obtain replaced attribute data. The filling methods include mean, median, mode, etc. For example, the age attribute can be filled with mode.

[0129] According to an embodiment of the present disclosure, each category of attribute data of a plurality of cleaned user data of a user is standardized to obtain a plurality of standardized attribute data, including:

[0130] For a plurality of attribute data of any category, determining a maximum value and a minimum value of the plurality of attribute data;

[0131] For each attribute data, the standardized attribute data is obtained according to the attribute data, the maximum value of the data and the minimum value of the data.

[0132] In one embodiment, the normalization formula is as shown in (1):

[0133]

[0134] Among them, X represents an attribute data of this category, max represents the maximum value of the attribute data of this category, min represents the minimum value of the attribute data of this category, and x scaled Represents the standardized value of the attribute data.

[0135] According to an embodiment of the present disclosure, the user attribute data after data cleaning is standardized to ensure the consistency of scales between different attributes so as to be better applied to data analysis.

[0136] According to an embodiment of the present disclosure, before performing the vector conversion process, the method further includes:

[0137] For each user, compare the user ID of the user at time t with the user IDs of multiple user data stored at time t-1 in the target database to obtain a comparison result;

[0138] If the comparison result shows that the user identifier at time t does not exist in the target database, the user data is stored in the target database;

[0139] When the comparison result indicates that the user identifier of the user at the tth moment exists in the target database, vector conversion processing is performed on the user data of the user.

[0140] According to an embodiment of the present disclosure, user attribute data after data standardization processing is obtained, and the user identification is used as a key matching field, such as a customer ID.

[0141] According to an embodiment of the present disclosure, a user identification at time t is obtained and compared with user identifications of multiple user data stored at time t-1 in the target database, for example, using a customer ID for comparison to obtain a comparison result.

[0142] According to an embodiment of the present disclosure, the comparison result may be 0 or 1, where 0 represents a new customer and 1 represents an old customer.

[0143] According to an embodiment of the present disclosure, when the compared user identifier does not exist in the target database, that is, the comparison result is 0, it is determined to be a new customer, and the user data of this user is stored in the target database.

[0144] According to an embodiment of the present disclosure, when a compared user identifier exists in the target database, that is, the comparison result is 1, it is determined to be an old customer, and the user data of this user is subjected to vector conversion processing.

[0145] According to an embodiment of the present disclosure, the acquired user data is compared with the user data stored in the target database to determine whether the customer is a new customer. If the customer is a new customer, the user data is stored. Otherwise, vector conversion processing is performed to further determine whether the user data in the target database needs to be updated.

[0146] Figure 3 A flow chart of a method for updating a user information graph according to an embodiment of the present disclosure is schematically shown.

[0147] like Figure 3 As shown, the user data is updated to the user information graph in the target database using a graph neural network to obtain an updated user information graph. The method 300 includes operations S310 to S340.

[0148] In operation S310, a verification information graph is constructed according to a plurality of user data of a plurality of users, wherein the verification information graph includes a plurality of user nodes and a plurality of verification edges, and the verification edge represents a relationship between two connected user nodes.

[0149] In operation S320, the nodes and verification edges in the verification information graph are embedded into the initial graph neural network to obtain a first graph neural network, wherein the first graph neural network includes a plurality of network nodes, and the network nodes represent the relationship between different users.

[0150] In operation S330, graph convolution operations are performed on the first graph neural network and the second graph neural network, respectively, to obtain multiple first feature vectors corresponding to the first graph neural network and multiple second feature vectors corresponding to the second graph neural network, wherein the second graph neural network is constructed based on the user information graph in the target database.

[0151] In operation S340, for any user, the user information graph is updated according to the first feature vector and the second feature vector corresponding to the user to obtain an updated user information graph.

[0152] According to an embodiment of the present disclosure, user data and associated user data are constructed into a verification information graph, which includes multiple user nodes and multiple verification edges. The user nodes may include user attribute data, such as customer characteristics, transaction amounts, etc.; the verification edges may represent the relationships between users, such as transactions, cooperation, etc.

[0153] According to an embodiment of the present disclosure, the nodes and verification edges in the verification information graph are embedded into the initial graph neural network to obtain a first graph neural network. The embedded user node representation obtains the relationship between users, such as transaction patterns, cooperative relationships, etc.

[0154] According to an embodiment of the present disclosure, the second graph neural network represents the user attribute data and relationships of the user and related users in the user information graph of the target database.

[0155] According to an embodiment of the present disclosure, the first graph neural network and the second graph neural network perform graph convolution operations on each user node, and generate a first feature vector and a second feature vector representation corresponding to each user node by aggregating user nodes and associated user node information, and the user nodes are represented as low-dimensional vectors.

[0156] According to an embodiment of the present disclosure, for any user, the user information graph is updated according to the first feature vector and the second feature vector corresponding to the user to obtain an updated user information graph.

[0157] According to an embodiment of the present disclosure, by comparing different user nodes through the embedded verification information graph, changes or anomalies in transaction patterns can be identified, thereby discovering user data that may need to be updated.

[0158] According to an embodiment of the present disclosure, the user information graph is updated according to the first feature vector and the second feature vector corresponding to the user to obtain an updated user information graph, including:

[0159] In the case where there is a difference between the first eigenvector and the second eigenvector, the network node corresponding to the second eigenvector in the second graph neural network is updated using the first eigenvector to obtain an updated second graph neural network;

[0160] For the first graph neural network, when the difference value between the first feature vector of the network node where the user is located and the first feature vector of the network node where the neighbor node associated with the user is located is greater than a preset difference threshold, the first feature vectors corresponding to the network node and the neighbor node are used to update the network node and the neighbor node in the second graph neural network respectively to obtain an updated second graph neural network, wherein the updated second graph neural network represents an updated user information graph.

[0161] According to an embodiment of the present disclosure, when there is a difference between the first eigenvector and the second eigenvector, the first eigenvector is used to update the network node corresponding to the second eigenvector in the second graph neural network to obtain an updated second graph neural network.

[0162] According to an embodiment of the present disclosure, when the difference value between the first feature vector of the network node where the user is located and the first feature vector of the network node where the neighbor node associated with the user is located is greater than a preset difference threshold, the network nodes and neighbor nodes in the second graph neural network are updated using the first feature vectors corresponding to the network nodes and neighbor nodes, respectively, to obtain an updated second graph neural network.

[0163] According to an embodiment of the present disclosure, updates are passed from one node to other network nodes. When the information of one node is updated, the neighboring nodes associated with the user in the second graph neural network can be gradually updated.

[0164] According to an embodiment of the present disclosure, using the first graph neural network and the second graph neural network, anomaly detection can be performed by monitoring user nodes and associated neighbor nodes. If there is a significant difference, it may indicate that its transaction behavior is abnormal and its information needs to be updated.

[0165] According to the embodiments of the present disclosure, by updating user information through a graph neural network, the association relationship between user nodes can be effectively utilized to transmit information in the network. Each user node can aggregate information from associated neighbor nodes to update its own information. In terms of user data updates, the accuracy and timeliness of data updates can be ensured.

[0166] According to an embodiment of the present disclosure, updating user information further includes:

[0167] The updated user information map is encrypted to obtain an encrypted user information map, so that an external device can access the encrypted user information map.

[0168] According to an embodiment of the present disclosure, the updated user information graph and the newly added user information graph are stored in the enterprise customer information platform (ECIS) and AES encrypted to obtain the encrypted user information graph.

[0169] According to an embodiment of the present disclosure, when an external device accesses an enterprise-level customer information platform, the access state is an encrypted user information graph; when an internal device accesses an enterprise-level customer information platform, the access state is an open source user information graph, that is, a user information graph in an unencrypted state.

[0170] According to the embodiments of the present disclosure, updated user data is stored and encrypted to protect customer information data, thereby improving user information security.

[0171] According to an embodiment of the present disclosure, the updated user information graph includes a plurality of updated user data corresponding to each user;

[0172] The updated user information graph is encrypted to obtain an encrypted user information graph, including:

[0173] For any user, generate N round keys used for multi-round encryption based on the randomly generated initial key;

[0174] In the i-th round of encryption, an XOR operation is performed with the i-th round key and the user data updated by the multiple input data to obtain multiple initial encrypted data, wherein the input data includes the updated user data or the i-1th encrypted data;

[0175] Based on a preset substitution cipher rule, generating a plurality of substitution cipher information according to a plurality of initial encrypted data;

[0176] Performing a password shift operation on a plurality of substitute password information to obtain intermediate password information, or performing a matrix change process on a password matrix composed of a plurality of substitute password information to obtain intermediate password information;

[0177] Generate the i-th encrypted data according to the intermediate password information and the i-th round key;

[0178] In the case of i=N, the i-th encrypted data is determined as the encrypted user data of the user, wherein the encrypted user information graph includes the encrypted user data of a plurality of users.

[0179] According to an embodiment of the present disclosure, for any user, N rounds of keys are generated according to the input key, which are used for each round of encryption operation.

[0180] According to an embodiment of the present disclosure, in the first round of encryption, the first round key is used to perform an XOR operation with multiple updated user data to obtain initial encrypted data; in the i-th round of encryption, i≥2, the i-th round key is used to perform an XOR operation with the previous round of initial encrypted data to obtain multiple initial encrypted data.

[0181] According to an embodiment of the present disclosure, a plurality of substitution cipher information is generated by searching a preset substitution cipher rule and a plurality of initial encrypted data in an S-box.

[0182] According to an embodiment of the present disclosure, a cryptographic move operation is performed on each row of a plurality of alternative cryptographic information to obtain intermediate cryptographic information, or a matrix change operation is performed on each column of a cryptographic matrix composed of a plurality of alternative cryptographic information to obtain intermediate cryptographic information.

[0183] According to an embodiment of the present disclosure, an XOR operation is performed based on the intermediate password information and the i-th round key to generate the i-th encrypted data;

[0184] According to an embodiment of the present disclosure, in the last round, the last round of encrypted data is determined as the encrypted user data of the user, and an encrypted user information graph is generated based on the encrypted user data of multiple users.

[0185] Based on the above user information updating method, the present disclosure also provides a user information updating device. Figure 4 The device is described in detail.

[0186] Figure 4 The structure block diagram of the user information updating device according to the embodiment of the present disclosure is schematically shown.

[0187] like Figure 4 As shown, the user information updating device 400 of this embodiment includes an acquisition module 410 , a processing module 420 , a generation module 430 and an updating module 440 .

[0188] The acquisition module 410 is used to acquire user data of multiple users in different channels at the tth time, wherein the user data includes multiple attribute data representing different user information. In one embodiment, the acquisition module 410 can be used to perform the operation S210 described above, which will not be repeated here.

[0189] The processing module 420 is used to perform vector conversion processing on multiple attribute data in the user data for each user to obtain the t-th user attribute vector corresponding to the user data. In one embodiment, the processing module 420 can be used to perform the operation S220 described above, which will not be repeated here.

[0190] The generating module 430 is used to generate information similarity according to the t-th user attribute vector and the t-1-th user attribute vector, wherein the t-1-th user attribute vector is generated according to the user data at the t-1-th time stored in the target database. In one embodiment, the generating module 430 can be used to perform the operation S230 described above, which will not be repeated here.

[0191] The updating module 440 is used to update the user data to the user information graph in the target database using the graph neural network when the information similarity meets the preset similarity threshold, and obtain an updated user information graph, wherein the user information graph includes user nodes representing different user data and connection edges representing the relationship between different user nodes. In one embodiment, the updating module 440 can be used to perform the operation S240 described above, which will not be repeated here.

[0192] According to the embodiments of the present disclosure, multiple attribute data in each user data are vectorized to obtain a user attribute vector, and then the information similarity between the user attribute vector and the user attribute vector in the target database is calculated to determine whether it needs to be updated, so as to update the user data to the user information graph in the target database using a graph neural network to obtain an updated user information graph. The graph neural network is used to quickly and accurately update and maintain the personal information of bank customers, store the updated information, and achieve the purpose of real-time dynamic updating of in-bank customer information while protecting customer information security, solving the cumbersome problem of updating customer information.

[0193] According to an embodiment of the present disclosure, the generation module 430 includes an acquisition submodule, a first generation submodule and a second generation submodule.

[0194] The acquisition submodule is used to obtain the user data of the user at the t-1th time from the target database.

[0195] The first generating submodule is used to perform vector conversion processing on the user data at the t-1th moment to obtain the t-1th user attribute vector.

[0196] The second generating submodule is used to process the t-th user attribute vector and the t-1-th user attribute vector by using the cosine similarity function to generate information similarity.

[0197] According to an embodiment of the present disclosure, the processing module 420 includes a first processing submodule, a second processing submodule, and a second processing submodule.

[0198] The first processing submodule is used to preprocess the attribute data for each attribute data to obtain the preprocessed attribute data.

[0199] The second processing submodule is used to process each preprocessed attribute data using a preset word vector model to obtain multiple attribute vectors.

[0200] The third processing submodule is used to perform average processing on multiple attribute vectors to obtain the t-th user attribute vector.

[0201] According to an embodiment of the present disclosure, the processing module 420 further includes a fourth processing submodule and a fifth processing submodule.

[0202] The fourth processing submodule is used to clean the multiple attribute data in the user data to obtain cleaned user data.

[0203] The fifth processing submodule is used to perform standardization processing on each category of the attribute data of the multiple cleaned user data of the user to obtain multiple standardized attribute data, so as to perform vector conversion processing using the standardized attribute data.

[0204] According to an embodiment of the present disclosure, the fourth processing submodule includes a visualization graph generating unit, a deleting unit and a replacing unit.

[0205] The visualization graph generating unit is used to draw a visualization graph according to a plurality of attribute data for any type of attribute data when the attribute data is of a numerical type.

[0206] The deleting unit is used to delete the attribute data corresponding to the extreme value when the difference between the extreme value and the average value in the visualization diagram is greater than a first preset threshold.

[0207] The replacement unit is used to replace the extreme value with a filling value when the difference between the extreme value and the average value in the visualization diagram is within a first preset threshold and a second preset threshold to obtain replaced attribute data, wherein the filling value is determined based on multiple attribute data.

[0208] According to an embodiment of the present disclosure, the fifth processing submodule includes a determining unit and a normalizing unit.

[0209] The determination unit is used to determine the maximum value and the minimum value of the multiple attribute data in any category.

[0210] The standardization unit is used to obtain standardized attribute data for each attribute data according to the attribute data, the maximum value of the data and the minimum value of the data.

[0211] According to an embodiment of the present disclosure, the processing module 420 further includes a sixth processing sub-module, a seventh processing sub-module and an eighth processing sub-module.

[0212] The sixth processing submodule is used to compare the user identification of each user at the tth time with the user identifications of the plurality of user data stored at the t-1th time in the target database to obtain a comparison result.

[0213] The seventh processing submodule is configured to store the user data in the target database if the comparison result indicates that the user identifier at the tth moment does not exist in the target database.

[0214] An eighth processing submodule is configured to perform vector conversion processing on the user data of the user when the comparison result indicates that the user identifier of the user at the tth moment exists in the target database.

[0215] According to an embodiment of the present disclosure, the update module 440 includes a first update submodule, a second update submodule, a third update submodule, and a fourth update submodule.

[0216] The first updating submodule is used to construct a verification information graph according to multiple user data of multiple users, wherein the verification information graph includes multiple user nodes and multiple verification edges, and the verification edge represents the relationship between two connected user nodes.

[0217] The second updating submodule is used to embed the nodes and verification edges in the verification information graph into the initial graph neural network to obtain a first graph neural network, wherein the first graph neural network includes multiple network nodes, and the network nodes represent the relationship between different users.

[0218] The third updating submodule is used to perform graph convolution operations on the first graph neural network and the second graph neural network, respectively, to obtain multiple first feature vectors corresponding to the first graph neural network and multiple second feature vectors corresponding to the second graph neural network, wherein the second graph neural network is constructed based on the user information graph in the target database.

[0219] The fourth updating submodule is used to update the user information graph for any user according to the first feature vector and the second feature vector corresponding to the user to obtain an updated user information graph.

[0220] According to an embodiment of the present disclosure, the fourth updating submodule includes a first updating unit and a second updating unit.

[0221] The first updating unit is used to update the network nodes corresponding to the second eigenvector in the second graph neural network using the first eigenvector when there is a difference between the first eigenvector and the second eigenvector, so as to obtain an updated second graph neural network.

[0222] The second updating unit is used to update the network nodes and neighbor nodes in the second graph neural network using the first feature vectors corresponding to the network nodes and neighbor nodes, respectively, for the first graph neural network, when the difference value between the first feature vector of the network node where the user is located and the first feature vector of the network node where the neighbor node associated with the user is located is greater than a preset difference threshold, so as to obtain an updated second graph neural network, wherein the updated second graph neural network represents an updated user information graph.

[0223] According to an embodiment of the present disclosure, the user information updating device 400 further includes an encryption module.

[0224] The encryption module is used to encrypt the updated user information map to obtain the encrypted user information map, so that the external device can access the encrypted user information map.

[0225] According to an embodiment of the present disclosure, the encryption module includes a first encryption submodule, a second encryption submodule, a third encryption submodule, a fourth encryption submodule, a fifth encryption submodule and a sixth encryption submodule.

[0226] The first encryption submodule is used to generate N round keys used for multi-round encryption for any user according to the randomly generated initial key.

[0227] The second encryption submodule is used to perform an XOR operation on the i-th round key and the user data updated by multiple input data to obtain multiple initial encrypted data during the i-th round of encryption, wherein the input data includes the updated user data or the i-1th encrypted data.

[0228] The third encryption submodule is used to generate a plurality of replacement cipher information according to a plurality of initial encrypted data based on a preset replacement cipher rule.

[0229] The fourth encryption submodule is used to perform a password shift operation on multiple replacement password information to obtain intermediate password information, or to perform a matrix change process on a password matrix composed of multiple replacement password information to obtain intermediate password information.

[0230] The fifth encryption submodule is used to generate the i-th encrypted data according to the intermediate password information and the i-th round key.

[0231] The sixth encryption submodule is used to determine the i-th encrypted data as the encrypted user data of the user when i=N, wherein the encrypted user information graph includes the encrypted user data of multiple users.

[0232] According to an embodiment of the present disclosure, any multiple modules in the acquisition module 410, the processing module 420, the generation module 430 and the update module 440 can be combined in 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. According to an embodiment of the present disclosure, at least one of the acquisition module 410, the processing module 420, the generation module 430 and the update module 440 can be at least partially implemented as a hardware circuit, such as a field programmable gate array (FPGA), a programmable logic array (PLA), a system on a chip, a system on a substrate, a system on a package, an application-specific integrated circuit (ASIC), or can be implemented by hardware or firmware such as any other reasonable way of integrating or packaging the circuit, or implemented in any one of the three implementation modes of software, hardware and firmware or in a proper combination of any of them. Alternatively, at least one of the acquisition module 410, the processing module 420, the generation module 430 and the update module 440 can be at least partially implemented as a computer program module, and when the computer program module is run, the corresponding function can be performed.

[0233] Figure 5 A block diagram of an electronic device suitable for implementing a user information updating method according to an embodiment of the present disclosure is schematically shown.

[0234] like Figure 5 As shown, the electronic device 500 according to an embodiment of the present disclosure includes a processor 501, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 502 or a program loaded from a storage part 508 into a random access memory (RAM) 503. The processor 501 may include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor and / or a related chipset and / or a special-purpose microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 501 may also include an onboard memory for caching purposes. The processor 501 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.

[0235] In RAM 503, various programs and data required for the operation of electronic device 500 are stored. Processor 501, ROM 502 and RAM 503 are connected to each other via bus 504. Processor 501 performs various operations of the method flow according to the embodiment of the present disclosure by executing the program in ROM 502 and / or RAM 503. It should be noted that the program can also be stored in one or more memories other than ROM 502 and RAM 503. Processor 501 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.

[0236] According to an embodiment of the present disclosure, the electronic device 500 may further include an input / output (I / O) interface 505, which is also connected to the bus 504. The electronic device 500 may further include one or more of the following components connected to the I / O interface 505: an input portion 506 including a keyboard, a mouse, etc.; an output portion 507 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage portion 508 including a hard disk, etc.; and a communication portion 509 including a network interface card such as a LAN card, a modem, etc. The communication portion 509 performs communication processing via a network such as the Internet. A drive 510 is also connected to the I / O interface 505 as needed. A removable medium 511, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 510 as needed, so that a computer program read therefrom is installed into the storage portion 508 as needed.

[0237] 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.

[0238] According to an embodiment of the present disclosure, the 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 502 and / or RAM 503 described above and / or one or more memories other than ROM 502 and RAM 503.

[0239] 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 user information update method provided by the embodiment of the present disclosure.

[0240] 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 501. According to the embodiment of the present disclosure, the system, device, module, unit, etc. described above can be implemented by a computer program module.

[0241] 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 509, and / or installed from the removable medium 511. 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.

[0242] In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 509, and / or installed from the removable medium 511. When the computer program is executed by the processor 501, 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.

[0243] 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).

[0244] 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.

[0245] It will be appreciated by those skilled in the art that the features described in the various embodiments and / or claims of the present disclosure may be combined and / or combined in a variety of ways, even if such combinations and / or combinations are not explicitly described in the present disclosure. In particular, the features described in the various embodiments and / or claims of the present disclosure may be combined and / or combined in a variety of ways without departing from the spirit and teachings of the present disclosure. All of these combinations and / or combinations fall within the scope of the present disclosure.

[0246] The embodiments of the present disclosure are described above. However, these embodiments are only for illustrative purposes 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 a variety of substitutions and modifications, which should all fall within the scope of the present disclosure.

Claims

1. A method for updating user information, comprising: At time t, user data of multiple users in different channels are obtained, wherein the user data includes multiple attribute data representing information of different users; For each of the users, performing vector conversion processing on the plurality of attribute data in the user data to obtain a t-th user attribute vector corresponding to the user data; Generate information similarity based on the t-th user attribute vector and the t-1-th user attribute vector, wherein the t-1-th user attribute vector is generated based on the user data at the t-1-th time stored in the target database; When the information similarity satisfies a preset similarity threshold, the user data is updated to the user information graph in the target database using a graph neural network to obtain an updated user information graph, wherein the user information graph includes user nodes representing different user data and connection edges representing the relationship between different user nodes.

2. The method according to claim 1, wherein: According to the t-th user attribute vector and the t-1-th user attribute vector, information similarity is generated, including: Acquire user data of the user at time t-1 from the target database; Performing vector conversion processing on the user data at the t-1th time to obtain the t-1th user attribute vector; The t-th user attribute vector and the t-1-th user attribute vector are processed using a cosine similarity function to generate the information similarity.

3. The method according to claim 1, wherein: Performing vector conversion processing on the plurality of attribute data in the user data to obtain a t-th user attribute vector corresponding to the user data includes: For each of the attribute data, preprocess the attribute data to obtain preprocessed attribute data; Using a preset word vector model to process each of the preprocessed attribute data to obtain a plurality of attribute vectors; A plurality of the attribute vectors are averaged to obtain the t-th user attribute vector.

4. The method according to claim 1, wherein: Before the vector conversion process, it also includes: Performing data cleaning on multiple attribute data in the user data to obtain cleaned user data; The attribute data of each category of the plurality of cleaned user data of the user are standardized to obtain a plurality of standardized attribute data, so as to perform vector conversion processing using the standardized attribute data.

5. The method according to claim 4, wherein: Performing data cleaning on multiple attribute data in the user data to obtain cleaned user data includes: For any type of attribute data, when the attribute data is of a numerical type, a visualization graph is drawn according to the plurality of attribute data; When the difference between the extreme value and the average value in the visualization diagram is greater than a first preset threshold, the attribute data corresponding to the extreme value is deleted; When the difference between the extreme value and the average value in the visualization graph is between a first preset threshold and a second preset threshold, the extreme value is replaced with a filling value to obtain replaced attribute data, wherein the filling value is determined based on a plurality of the attribute data.

6. The method according to claim 4, wherein: The attribute data of each category of the plurality of cleaned user data of the user are standardized to obtain a plurality of standardized attribute data, including: For a plurality of attribute data of any category, determining a maximum value and a minimum value of the plurality of attribute data; For each of the attribute data, the standardized attribute data is obtained according to the attribute data, the data maximum value and the data minimum value.

7. The method according to claim 1, before performing the vector conversion process, further comprises: For each of the users, comparing the user ID of the user at time t with the user IDs of the plurality of user data stored at time t-1 in the target database to obtain a comparison result; If the comparison result shows that the user identifier at the time t does not exist in the target database, storing the user data in the target database; When the comparison result indicates that the user identifier of the user at the tth moment exists in the target database, vector conversion processing is performed on the user data of the user.

8. The method according to claim 1, wherein: The user data is updated to the user information graph in the target database using a graph neural network to obtain an updated user information graph, including: Constructing a verification information graph according to multiple user data of multiple users, wherein the verification information graph includes multiple user nodes and multiple verification edges, and the verification edges represent the relationship between two connected user nodes; Embed the nodes and verification edges in the verification information graph into an initial graph neural network to obtain a first graph neural network, wherein the first graph neural network includes a plurality of network nodes, and the network nodes represent the relationship between different users; Performing graph convolution operations on the first graph neural network and the second graph neural network respectively to obtain a plurality of first feature vectors corresponding to the first graph neural network and a plurality of second feature vectors corresponding to the second graph neural network, wherein the second graph neural network is constructed according to a user information graph in a target database; For any of the users, the user information graph is updated according to the first feature vector and the second feature vector corresponding to the user to obtain the updated user information graph.

9. The method according to claim 8, wherein: The method of updating the user information graph according to the first feature vector and the second feature vector corresponding to the user to obtain the updated user information graph includes: In the case where there is a difference between the first eigenvector and the second eigenvector, using the first eigenvector to update the network node corresponding to the second eigenvector in the second graph neural network to obtain an updated second graph neural network; For the first graph neural network, when the difference value between the first feature vector of the network node where the user is located and the first feature vector of the network node where the neighbor node associated with the user is located is greater than a preset difference threshold, the network node and the neighbor node in the second graph neural network are updated using the first feature vectors corresponding to the network node and the neighbor node respectively to obtain an updated second graph neural network, wherein the updated second graph neural network represents the updated user information graph.

10. The method according to claim 1, further comprising: The updated user information map is encrypted to obtain an encrypted user information map, so that an external device can access the encrypted user information map.

11. The method according to claim 10, wherein the updated user information graph comprises a plurality of updated user data corresponding to each of the users; in, The updated user information graph is encrypted to obtain an encrypted user information graph, including: For any user, generate N round keys used for multi-round encryption based on the randomly generated initial key; In the i-th round of encryption, an XOR operation is performed using the i-th round key and the user data updated by a plurality of input data to obtain a plurality of initial encrypted data, wherein the input data includes the updated user data or the i-1th encrypted data; Based on a preset substitution cipher rule, generating a plurality of substitution cipher information according to a plurality of the initial encrypted data; Performing a password shift operation on a plurality of the replacement password information to obtain intermediate password information, or performing a matrix change process on a password matrix composed of a plurality of the replacement password information to obtain intermediate password information; Generate the i-th encrypted data according to the intermediate password information and the i-th round key; In the case of i=N, the i-th encrypted data is determined as the encrypted user data of the user, wherein the encrypted user information graph includes encrypted user data of a plurality of users.

12. A user information updating device, comprising: An acquisition module, used to acquire user data of multiple users in different channels at time t, wherein the user data includes multiple attribute data representing different user information; A processing module, configured to perform vector conversion processing on the plurality of attribute data in the user data for each user, so as to obtain a t-th user attribute vector corresponding to the user data; A generating module, used to generate information similarity according to the t-th user attribute vector and the t-1-th user attribute vector, wherein the t-1-th user attribute vector is generated according to the user data at the t-1-th time stored in the target database; An updating module is used to update the user data to the user information graph in the target database using a graph neural network when the information similarity meets a preset similarity threshold, so as to obtain an updated user information graph, wherein the user information graph includes user nodes representing different user data and connection edges representing the relationship between different user nodes.

13. An electronic device, comprising: one or more processors; a storage device for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors are enabled to execute the method according to any one of claims 1 to 11.

14. A computer-readable storage medium having executable instructions stored thereon, which, when executed by a processor, causes the processor to execute the method according to any one of claims 1 to 11.

Images