Table processing method and device, storage medium, program product and electronic equipment

Abstract

The application discloses a table processing method and device, a storage medium, a program product and an electronic device. It relates to the field of artificial intelligence, and the method comprises the following steps: acquiring a plurality of newly added business tables in a target storage area; constructing an adjacency matrix based on the association relationship between the plurality of business tables, wherein the elements in the adjacency matrix represent the association strength between two business tables, and the association strength represents the probability value of two business tables belonging to the same user; processing the adjacency matrix and the plurality of business tables by using a target recognition model to obtain the probability value of each business table belonging to different users, wherein the target recognition model comprises a graph neural network and a table data representation learning model; and determining the user associated with each business table according to the probability value of each business table belonging to different users. Through the application, the problem of low recognition accuracy when identifying the user associated with the business table in the related art is solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of artificial intelligence, in particular to a table processing method and device, a storage medium, a program product and an electronic device. BACKGROUND

[0002] In financial risk assessment, it is often necessary to comprehensively consider the information of multiple dimensions of a customer, such as credit records, transaction behaviors, etc. These information is often scattered in different business tables, and records such as transaction behaviors usually involve a large amount of data, which need to be recorded by using multiple tables. However, not all tables directly record the customer information of the customer (for example, only the first page table records the user information, and the other page tables omit the user information), therefore, the step of determining the association relationship between the customer and the table is particularly important. At present, the related technology relies on pre-defined rules or pattern recognition of the aforementioned association relationship, but these rules or patterns are difficult to effectively adapt to the changing financial environment and user behavior, thereby there is a problem of low recognition accuracy.

[0003] In view of the above problems, no effective solution has been proposed so far. SUMMARY

[0004] The main purpose of the present application is to provide a table processing method, device, storage medium, program product and electronic device, so as to solve the problem of low recognition accuracy in identifying the user associated with the business table in the related technology.

[0005] In order to achieve the above purpose, according to one aspect of the present application, a table processing method is provided. The method comprises: acquiring a plurality of newly added business tables in a target storage area, wherein a single business table is used to record the business information of a single user, part of the plurality of business tables includes user information of the user, and the remaining business tables do not include user information of the user; constructing an adjacency matrix based on the association relationship between the plurality of business tables, wherein the elements in the adjacency matrix represent the association strength between two business tables, and the association strength represents the probability value that two business tables belong to the same user; processing the adjacency matrix and the plurality of business tables by using a target recognition model to obtain the probability value that each business table belongs to different users, wherein the target recognition model includes a graph neural network and a table data representation learning model; and determining the user associated with each business table according to the probability value that each business table belongs to different users.

[0006] Further, the table processing method further comprises: taking the business table as a node, constructing an edge between the associated nodes based on the association relationship between the plurality of business tables to obtain a target graph structure; and constructing the adjacency matrix based on the target graph structure.

[0007] Further, the table processing method further comprises: processing the adjacency matrix and the plurality of business tables through a graph neural network to obtain a first vector representation of each business table; processing the plurality of business tables using a table data representation learning model to obtain a second vector representation of each business table; for each business table, performing fusion processing on the first vector representation and the second vector representation of the business table to obtain a target vector of the business table; processing the target vector of each business table through a fully connected layer and a normalization exponential layer in the target identification model to obtain a probability value of each business table belonging to different users.

[0008] Further, the table processing method further comprises: determining a first hyperparameter corresponding to the graph neural network, and determining a second hyperparameter corresponding to the table data representation learning model; according to the first hyperparameter and the second hyperparameter, performing weighted fusion on the first vector representation and the second vector representation of the business table to obtain the target vector.

[0009] Further, the table processing method further comprises: for each business table, determining a maximum probability value from the probability value corresponding to the business table; determining the user corresponding to the maximum probability value as the user associated with the business table.

[0010] Further, the table processing method further comprises: obtaining a training sample set, wherein the training sample in the training sample set comprises a plurality of sample business tables, a sample adjacency matrix corresponding to the plurality of sample business tables, and a real label of the training sample representing a user to which each sample business table belongs; training an initial identification model based on the training sample set to obtain the target identification model.

[0011] Further, the table processing method further comprises: determining a plurality of sample business tables from historical business tables in the target storage area, wherein the historical business tables are business tables whose belonging users have been determined; constructing a sample adjacency matrix corresponding to the plurality of sample business tables based on the association relationship between the plurality of sample business tables; determining a training sample based on the plurality of sample business tables and the sample adjacency matrix corresponding to the plurality of sample business tables.

[0012] In order to achieve the above object, according to another aspect of the present application, a table processing apparatus is provided. The apparatus comprises: a first obtaining module configured to obtain a plurality of service tables newly added in a target storage area, wherein a single service table is configured to record service information of a single user, and part of the plurality of service tables comprises user information of a user, and the remaining service tables do not comprise user information of a user; a first constructing module configured to construct an adjacency matrix based on an association relationship between the plurality of service tables, wherein an element in the adjacency matrix represents an association strength between two service tables, and the association strength represents a probability value that the two service tables belong to the same user; a first processing module configured to process the adjacency matrix and the plurality of service tables by using a target recognition model to obtain a probability value that each service table belongs to a different user, wherein the target recognition model comprises a graph neural network and a table data representation learning model; and a first determining module configured to determine a user associated with each service table according to the probability value that each service table belongs to a different user.

[0013] Further, the first constructing module further comprises: a first constructing submodule configured to take the service tables as nodes, construct edges between the associated nodes based on the association relationship between the plurality of service tables, and obtain a target graph structure; and a second constructing submodule configured to construct the adjacency matrix based on the target graph structure.

[0014] Further, the first processing module further comprises: a first processing submodule configured to process the adjacency matrix and the plurality of service tables by using the graph neural network to obtain a first vector representation of each service table; a second processing submodule configured to process the plurality of service tables by using the table data representation learning model to obtain a second vector representation of each service table; a third processing submodule configured to, for each service table, perform fusion processing on the first vector representation and the second vector representation of the service table to obtain a target vector of the service table; and a fourth processing submodule configured to process the target vector of each service table by using a fully connected layer and a normalization exponential layer in the target recognition model to obtain a probability value that each service table belongs to a different user.

[0015] Further, the third processing submodule further comprises: a determining unit configured to determine a first hyperparameter corresponding to the graph neural network and a second hyperparameter corresponding to the table data representation learning model; and a processing unit configured to perform weighted fusion on the first vector representation and the second vector representation of the service table according to the first hyperparameter and the second hyperparameter to obtain the target vector.

[0016] Further, the first determining module further comprises: a first determining submodule configured to, for each service table, determine a maximum probability value from the probability value corresponding to the service table; and a second determining submodule configured to determine a user corresponding to the maximum probability value as a user associated with the service table.

[0017] Further, the table processing apparatus further comprises: a second obtaining module, configured to obtain a training sample set, wherein each training sample in the training sample set comprises a plurality of sample business tables and a sample adjacency matrix corresponding to the plurality of sample business tables, and a true label of the training sample represents a user to which each sample business table belongs; and a second processing module, configured to train an initial identification model based on the training sample set to obtain a target identification model.

[0018] Further, the table processing apparatus further comprises: a second determining module, configured to determine a plurality of sample business tables from historical business tables in the target storage area, wherein the historical business tables are business tables for which the users have been determined; a second constructing module, configured to construct a sample adjacency matrix corresponding to the plurality of sample business tables based on an association relationship between the plurality of sample business tables; and a third determining module, configured to determine training samples based on the plurality of sample business tables and the sample adjacency matrix corresponding to the plurality of sample business tables.

[0019] In order to achieve the above-mentioned purpose, according to another aspect of the present application, a computer readable storage medium is provided, which comprises a stored executable program, wherein the computer readable storage medium controls the device where the computer readable storage medium is located to execute the table processing method mentioned above when the executable program is running.

[0020] In order to achieve the above-mentioned purpose, according to another aspect of the present application, an electronic device is provided, which comprises a memory storing an executable program, and a processor configured to run the program, wherein the program executes the table processing method mentioned above when running.

[0021] In order to achieve the above-mentioned purpose, according to another aspect of the present application, a computer program product is provided, which comprises computer instructions, and the computer instructions implement the steps of the table processing method mentioned above when executed by a processor.

[0022] In the embodiments of the present application, by constructing an adjacency matrix based on the association relationship between the newly added plurality of business tables, the known complex relationship between the business tables is converted into a quantifiable adjacency matrix. By setting the target identification model to include a graph neural network and a table data representation learning model, and processing the adjacency matrix and the plurality of business tables by using the target identification model, the probability value of each business table belonging to different users is obtained, which effectively obtains the complex relationship between the tables by using the graph neural network, accurately captures the context information and relationship dependency, learns the internal features of the tables by using the table data representation learning model, and thus determines the probability value of the business table belonging to different users based on the internal features of the tables and the cross-table features, thereby improving the accuracy of the determined probability value. By determining the user associated with each business table according to the probability value of each business table belonging to different users, the accuracy of identifying the user associated with the business table can be effectively improved.

[0023] Therefore, the scheme provided in the application achieves the purpose of determining the user associated with the business table according to the internal features and cross-table features of the table, realizes the technical effect of improving the accuracy of identifying the user associated with the business table, and solves the problem of low identification accuracy when identifying the user associated with the business table in the related art. BRIEF DESCRIPTION OF DRAWINGS

[0024] The accompanying drawings, which form a part of this application, are included to provide a further understanding of the application and are incorporated in and constitute a part of this application. The embodiments of the application illustrated in the drawings, and their description, are presented to explain the application and not to limit or define it. In the drawings:

[0025] Figure 1 is a hardware structure block diagram of a computer terminal according to an embodiment of the application;

[0026] Figure 2 is a flowchart of a table processing method according to an embodiment of the application;

[0027] Figure 3 is a schematic diagram of a table processing device according to an embodiment of the application;

[0028] Figure 4 is a structure block diagram of an electronic device according to an embodiment of the application. DETAILED DESCRIPTION

[0029] In order to enable persons skilled in the art to better understand the application scheme, the technical solutions in the embodiments of the application will be clearly and completely described below in conjunction with the drawings in the embodiments of the application. Obviously, the described embodiments are only a part of the embodiments of the application, not all the embodiments. Based on the embodiments in the application, all other embodiments obtained by persons skilled in the art without creative labor should be within the scope of protection of the application.

[0030] It should be noted that the terms "first", "second", etc. in the specification and claims of the application and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily describe a specific order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the application described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device including a series of steps or units does not necessarily limit to those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0031] It should be noted that the information (including but not limited to user equipment information, user personal information, etc.) and data (including but not limited to data for display, analyzed data, etc.) involved in the present application are information and data authorized by the user or authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of related data comply with relevant laws, regulations and standards, necessary security measures are taken, do not violate public order and good customs, and provide corresponding operation portal for user to choose authorization or refusal. For example, the system and the interface between the related users or institutions are provided with the corresponding operation portal for the user to choose to agree or refuse the automatic decision result; if the user chooses to refuse, the expert decision process is entered.

[0032] Embodiment 1

[0033] According to the embodiments of the present application, an embodiment of a form processing method is also provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that here.

[0034] The method embodiment provided by the embodiment of the present application can be executed in a mobile terminal, a computer terminal or a similar computing device. Figure 1 A hardware structure block diagram of a computer terminal (or mobile device) for implementing the form processing method is shown. As shown in Figure 1 , the computer terminal 10 (or mobile device) can include one or more processors 102 (the processor 102 can include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA), a memory 104 for storing data, and a transmission device 106 for communication function. In addition, it can also include a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which can be included as one of the ports of the BUS bus), a network interface, a power supply and / or a camera. Those skilled in the art can understand that Figure 1 The structure shown is only schematic, which does not limit the structure of the above-mentioned electronic device. For example, the computer terminal 10 can include more or fewer components than those shown in Figure 1 , or have a different configuration from Figure 1 .

[0035] It should be noted that the one or more processors 102 and / or other data processing circuitry described above can be referred to herein generally as "data processing circuitry." The data processing circuitry can be embodied in whole or in part as software, hardware, firmware, or any combination thereof. In addition, the data processing circuitry can be a single standalone processing module, or incorporated in whole or in part within any one of the other elements of the computer terminal 10 (or mobile device). As referred to in embodiments of the present application, the data processing circuitry functions as a processor to control, for example, selection of variable resistance terminal paths connected to the interface.

[0036] The memory 104 can be used to store software programs of application software and modules, such as program instructions / data storage means corresponding to the table processing method in embodiments of the present application. The processor 102 executes various functional applications and data processing by running the software programs and modules stored in the memory 104, i.e., implements the table processing method described above. The memory 104 can include a high-speed random access memory, and can further include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 can further include a memory disposed remotely with respect to the processor 102, which can be connected to the computer terminal 10 through a network. Examples of the network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0037] The transmission device 106 is used to receive or send data via a network. Specific examples of the network can include a wireless network provided by a communication provider of the computer terminal 10. In one example, the transmission device 106 includes a network adapter (NIC), which can be connected to other network devices through a base station to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (RF) module, which is used to communicate with the Internet in a wireless manner.

[0038] The display can be, for example, a touch screen type liquid crystal display (LCD), which can enable a user to interact with the user interface of the computer terminal 10 (or mobile device).

[0039] In the above operating environment, the present application provides a table processing method as shown in Figure 2 Figure 2 is a flowchart of the table processing method provided according to embodiments of the present application, as shown in Figure 2 the method includes the following steps:

[0040] ​In step S201, a plurality of newly added business tables in a target storage area are obtained, wherein a single business table is used to record business information of a single user, and some of the plurality of business tables include user information of a user, and the remaining business tables do not include user information of a user.

[0041] Optionally, an electronic device, an application system, a server or the like can be taken as an execution subject of the present application, and in the embodiment, the target processing system is taken as the execution subject to execute the above table processing method.

[0042] Optionally, the target storage area can be a database or a system used to store financial business data, and the area can include various business tables, such as credit record tables, transaction behavior tables and the like. The plurality of newly added business tables refer to tables newly added in the target storage area, and the tables contain the latest business information about different users.

[0043] Optionally, each business table is used to record specific business data of a single user, for example, a business table can record credit history of a user, and the credit history is equivalent to the above business information. Since the business information of a single user can also involve a large amount of data, a single user's business information of a certain type (for example, transaction behavior) can be recorded in a plurality of business tables. However, not all tables directly record the user information of a user, for example, only the first page of the transaction behavior table records the user information of a user, and the other pages of the transaction behavior table omit the user information of the user and only record business information, and therefore, it is necessary to determine the association between the business tables without recording the user information and the user.

[0044] In step S202, an adjacency matrix is constructed based on the association between the plurality of business tables, wherein an element in the adjacency matrix represents an association strength between two business tables, and the association strength represents a probability value that the two business tables belong to a same user.

[0045] In some embodiments, the target processing system can first construct a target graph structure based on the association between the plurality of business tables, and then construct the adjacency matrix based on the target graph structure. In the target graph structure, each node represents a business table, and an edge represents the association between the business tables and the association strength. The association between the business tables is used to represent whether two business tables belong to a same user, and the association strength between the business tables represents a probability value that the two business tables belong to a same user.

[0046] Optionally, the target processing system can determine the association strength between the business forms based on the user information in the business forms. For example, if two business forms record the user information of the same user, the association strength is determined to be 1, and an edge is constructed between the corresponding nodes. If the two business forms do not record the user information of the same user, the association strength is determined to be 0, and no edge is constructed between the corresponding nodes.

[0047] Optionally, the adjacency matrix is a matrix used to represent the connection relationship between nodes in the graph structure, and the elements in the matrix represent the association strength between two nodes (i.e., business forms). In this embodiment, the adjacency matrix is used to quantify the possibility that the business forms belong to the same user. Optionally, the adjacency matrix can be represented as G = (V, E), where V represents a set of nodes, each node v i ∈ V is a business form. E is a set of edges, and each edge e i,j ∈ E represents the association strength between the form v i and the form v j .

[0048] In step S203, the adjacency matrix and the plurality of business forms are processed by using a target identification model to obtain a probability value of each business form belonging to different users, wherein the target identification model includes a graph neural network and a form data representation learning model.

[0049] Optionally, after the adjacency matrix is determined, the adjacency matrix and the plurality of business forms can be processed by using a target identification model to obtain a probability value of each business form belonging to different users. For example, the adjacency matrix and the plurality of business forms are input into a target identification network model, a first vector representation of each business form is determined by a graph neural network according to the input data, and a second vector representation of each business form is determined by a form data representation learning model according to the input data. Thus, the probability value of each business form belonging to different users is determined by the target identification model according to the first vector representation and the second vector representation of each business form.

[0050] The graph neural network is a deep learning model for processing structured graph data, capable of capturing complex relationships between nodes and transmitting information. The graph neural network can encode the adjacency matrix, thereby learning the association features between the business tables. The table data representation learning model can be a hypergraph-enhanced table data learning model. The table data representation learning model can model the table data using a hypergraph structure, such as converting the table data into a hypergraph structure, where each node represents a cell in the table, and the edge represents the relationship between the cells, to capture the invariance and hierarchical structure of the table. The table data representation learning model is used to learn and encode the internal features of the table data, such as cell content, table structure, etc. By combining it with the graph neural network set, the information inside and across tables can be integrated, thereby avoiding the problem of information loss or one-sided understanding caused by focusing on a single table or data source.

[0051] In step S204, the user associated with each business table is determined according to the probability value of each business table belonging to different users.

[0052] Optionally, after obtaining the probability value of each business table belonging to different users, the target processing system can determine the maximum probability value from the probability value corresponding to each business table, thereby determining the user corresponding to the maximum probability value as the user associated with the business table.

[0053] In some embodiments, the target processing system can determine a target business table from a plurality of business tables, wherein the target business table refers to a business table that does not include user information of a user, thereby determining the user associated with the target business table based on the probability value of the target business table belonging to different users. For a business table that includes user information, the user to which the user information in the business table belongs is directly determined as the user associated with the business table.

[0054] Optionally, after determining the user associated with each business table, the target processing system can integrate the business information in the business tables belonging to the same user, and then evaluate the transaction risk value of the user based on the integrated information, thereby facilitating the improvement of the accuracy of transaction risk evaluation.

[0055] In the embodiments of the present application, by constructing an adjacency matrix based on the association relationship between the newly added multiple business forms, the known complex relationship between the business forms is converted into a quantifiable adjacency matrix. By setting the target recognition model to include a graph neural network and a table data representation learning model, and processing the adjacency matrix and the multiple business forms by using the target recognition model, the probability value of each business form belonging to different users is obtained, the complex relationship between the forms is effectively obtained by using the graph neural network, the context information and relationship dependency are accurately captured, the internal features of the forms are learned by using the table data representation learning model, and thus the probability value of the business form belonging to different users is determined based on the internal features of the forms and the cross-form features, the accuracy of the determined probability value is improved, and the user associated with each business form is determined according to the probability value of each business form belonging to different users, so that the accuracy of identifying the user associated with the business form is effectively improved.

[0056] Therefore, the scheme provided in the present application achieves the purpose of determining the user associated with the business form based on the internal features of the form and the cross-form features, realizes the technical effect of improving the accuracy of identifying the user associated with the business form, and solves the problem of low recognition accuracy when identifying the user associated with the business form in the related art.

[0057] Optionally, in the table processing method provided in the embodiments of the present application, the adjacency matrix is constructed based on the association relationship between the multiple business forms, including: regarding the business forms as nodes, constructing edges between the associated nodes based on the association relationship between the multiple business forms to obtain a target graph structure; and constructing the adjacency matrix based on the target graph structure.

[0058] Optionally, before constructing the adjacency matrix, the business forms are first abstracted as nodes in the graph structure. Each business form is regarded as an independent node, and the edges between the nodes reflect the association relationship and the association strength between the business forms.

[0059] Optionally, the target processing system can determine the association strength between the business forms based on the user information in the business forms. For example, if the same user information of the same user is recorded in two business forms, the association strength is determined to be 1, and an edge is constructed between the corresponding nodes, if the same user information of the same user is not recorded in the two business forms, the association strength is determined to be 0, and no edge is constructed between the corresponding nodes.

[0060] Optionally, after the target graph structure is determined, the target processing system can construct an adjacency matrix based on the target graph structure, each element in the adjacency matrix representing whether there is an edge between two nodes and the weight of the edge, i.e., the association strength between the business forms. When constructing the adjacency matrix, for two business forms with an association relationship, their corresponding elements will be assigned a non-zero value reflecting the degree of association between them, and the business forms without an association relationship correspond to zero values.

[0061] It should be noted that in the above manner, the complex business form association relationship is effectively converted into a mathematical representation that can be processed by a graph neural network, enabling the model to learn and utilize these relationships, thereby improving the model processing effect and further improving the accuracy of identifying the association relationship between the user and the form.

[0062] Optionally, in the form processing method provided in the embodiments of the present application, the adjacency matrix and the plurality of business forms are processed by using a target recognition model to obtain a probability value of each business form belonging to different users, comprising: processing the adjacency matrix and the plurality of business forms by a graph neural network to obtain a first vector representation of each business form; processing the plurality of business forms by using a form data representation learning model to obtain a second vector representation of each business form; for each business form, performing fusion processing on the first vector representation and the second vector representation of the business form to obtain a target vector of the business form; processing the target vector of each business form by a fully connected layer and a normalization exponential layer in the target recognition model to obtain a probability value of each business form belonging to different users.

[0063] Optionally, after the adjacency matrix and the plurality of business forms are input into the target recognition model, the adjacency matrix and the plurality of business forms can be processed by a graph neural network to obtain a first vector representation of each business form. These vector representations represent the cross-form relationship features carried by the business forms after being associated through the graph structure. The purpose is to introduce a node key information transmission mechanism to learn the business form feature representation containing cross-form relationships by aggregating the features of adjacent nodes.

[0064] Optionally, the convolution layer in the graph neural network can update the feature representation by the following formula:

[0065]

[0066] wherein, H (l+1) represents the node embedding matrix of the L+1th convolution layer; H (l) represents the node embedding matrix of the Lth convolution layer; σ represents an activation function, such as ReLU; is the adjacency matrix of the graph plus the identity matrix, used to consider the self-loop; represents a degree matrix, i.e. The elements on the diagonal are the sum of each row; W (l) denotes the weight matrix of the L-th layer. By stacking multiple convolutional layers, the complex relationship between tables can be learned, and the embedding representation of each table can be obtained.

[0067] Optionally, the table data representation learning model is used to process the business table, extract internal features such as table content and structure, and output a second vector representation. This representation reflects the information of the business table itself and does not include cross-table relationships. The purpose of this step is to capture the details and complexity of the business table using the model and generate a feature vector for each table based on its own content.

[0068] Optionally, after obtaining the first vector representation and the second vector representation, the target processing system can combine the first vector representation (i.e., cross-table relationship features) and the second vector representation (i.e., table internal features) to obtain a more comprehensive feature representation, i.e., a target vector. For example, the target processing system can directly concatenate the first vector representation and the second vector representation to obtain the target vector, or the target processing system can perform weighted fusion on the first vector representation and the second vector representation to obtain the target vector.

[0069] After obtaining the target vector, the target processing system can convert the target vector through a fully connected layer to map it to a space that can represent user classification, and then map the converted vector to a probability value through a normalization exponential layer, thereby obtaining the probability value of each business table belonging to different users.

[0070] It should be noted that through the above process, the internal features of the table and the cross-table relationship features are effectively obtained, and the probability value of each business table belonging to different users is determined based on the features in these two dimensions, thereby improving the accuracy of the determined probability value.

[0071] Optionally, in the table processing method provided in the embodiments of the present application, the first vector representation and the second vector representation of the business table are fused to obtain a target vector of the business table, including: determining a first hyperparameter corresponding to the graph neural network, and determining a second hyperparameter corresponding to the table data representation learning model; according to the first hyperparameter and the second hyperparameter, the first vector representation and the second vector representation of the business table are weighted and fused to obtain the target vector.

[0072] In some embodiments, the target vector can be determined in the following manner:

[0073] F = aQ + bK

[0074] Wherein, F represents a target vector, a represents a first hyperparameter, β represents a second hyperparameter, Q represents a first vector representation, and K represents a second vector representation. The specific values of a and β are learned by the target recognition model in the training process.

[0075] It should be noted that the target vector is determined by using the weighted fusion manner, which improves the accuracy of the determined target vector, and thus the accuracy of identifying the user associated with the business form can be improved.

[0076] Optionally, in the table processing method provided in the embodiments of the present application, the user associated with each business form is determined according to the probability value of each business form belonging to different users, comprising: for each business form, determining the maximum probability value from the probability value corresponding to the business form; and determining the user corresponding to the maximum probability value as the user associated with the business form.

[0077] For example, for a certain business form, assuming that the target recognition model outputs the probability values of the business form belonging to A, B and C three users, wherein the probability value corresponding to A user is 50%, the probability value corresponding to B user is 30%, and the probability value corresponding to C user is 20%, then A user is determined as the user associated with the business form.

[0078] It should be noted that through the above process, the user associated with the business form is determined according to the size of the quantitative value output by the model, thereby improving the accuracy of identifying the user associated with the business form.

[0079] Optionally, in the table processing method provided in the embodiments of the present application, the target recognition model is trained by the following manner: obtaining a training sample set, wherein the training sample in the training sample set includes a plurality of sample business forms, a plurality of sample business form corresponding sample adjacency matrices, and the real label of the training sample represents the user to which each sample business form belongs; training an initial recognition model based on the training sample set to obtain the target recognition model.

[0080] Optionally, the sample business form can be understood as a historical business form in the target storage area, and these sample business forms are business forms whose belonging users are known. The sample adjacency matrix is constructed based on a plurality of sample business forms, and the sample adjacency matrix is used to quantify the degree of association between the sample business forms.

[0081] After obtaining the training sample set, the target processing model can train an initial recognition model based on the training sample set to obtain the target recognition model. The loss function used in the training process can be a cross-entropy loss function, and the initial recognition model includes an initial graph neural network and an initial table data representation learning model.

[0082] It should be noted that, by training the initial identification model by using the known data, the target identification model is obtained, so that the initial identification model can effectively learn how to extract key features from the business table and the adjacency matrix in the training process and associate them with the user information, thereby improving the accuracy of the target identification model obtained by training.

[0083] Optionally, in the table processing method provided in the embodiments of the present application, the training sample is obtained by: determining a plurality of sample business tables from historical business tables in the target storage area, wherein the historical business tables are business tables of which the users have been determined; constructing a sample adjacency matrix corresponding to the plurality of sample business tables based on the association relationship between the plurality of sample business tables; and determining the training sample based on the plurality of sample business tables and the sample adjacency matrix corresponding to the plurality of sample business tables.

[0084] Optionally, the historical business table can be a table that has been completed processing and for which the associated user has been determined. The target processing system can randomly select a historical business table from the target storage area as a sample business table, or the target processing system can select a plurality of historical business tables corresponding to a plurality of users from the target storage system as sample business tables, and some of the plurality of sample business tables do not include user information.

[0085] After the plurality of sample business tables are determined, the target processing system can construct a sample adjacency matrix corresponding to the plurality of sample business tables based on the association relationship between the plurality of sample business tables, wherein the process of constructing the sample adjacency matrix is the same as the process of constructing the adjacency matrix, and therefore will not be described again here.

[0086] After the sample adjacency matrix is obtained, the target processing system can combine the plurality of sample business tables and the sample adjacency matrix corresponding to the plurality of sample business tables into a training sample. Among different training samples, the plurality of sample business tables are not completely the same.

[0087] It should be noted that, by the above method, the training sample is effectively constructed, thereby further improving the model training effect.

[0088] As can be seen, the scheme provided in the present application achieves the purpose of determining the user associated with the business table according to the internal features of the table and the cross-table features, realizes the technical effect of improving the accuracy of identifying the user associated with the business table, and solves the problem of low identification accuracy when identifying the user associated with the business table in the related art.

[0089] It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that here.

[0090] Embodiment 2

[0091] The embodiment of the present application also provides a table processing device. It should be noted that the table processing device of the embodiment of the present application can be used to execute the table processing method provided by the embodiment of the present application. The table processing device provided by the embodiment of the present application is introduced as follows.

[0092] According to the embodiment of the present application, a table processing device for implementing the above table processing method is also provided, as shown in Figure 3 The device comprises:

[0093] The first obtaining module 301 is configured to obtain a plurality of newly added business tables in a target storage area, wherein a single business table is used to record the business information of a single user, and part of the plurality of business tables includes user information of a user, and the remaining business tables do not include user information of a user.

[0094] The first construction module 302 is configured to construct an adjacency matrix based on the association relationship between the plurality of business tables, wherein an element in the adjacency matrix represents the association strength between two business tables, and the association strength represents a probability value that the two business tables belong to the same user.

[0095] The first processing module 303 is configured to process the adjacency matrix and the plurality of business tables by using a target recognition model to obtain a probability value that each business table belongs to a different user, wherein the target recognition model includes a graph neural network and a table data representation learning model.

[0096] The first determination module 304 is configured to determine the user associated with each business table according to the probability value that each business table belongs to a different user.

[0097] In the embodiment of the present application, by constructing an adjacency matrix based on the association relationship between the newly added multiple business tables, the known complex relationship between the business tables is converted into a quantifiable adjacency matrix. By setting the target recognition model to include a graph neural network and a table data representation learning model, and processing the adjacency matrix and the multiple business tables by using the target recognition model, the probability value of each business table belonging to different users is obtained, the complex relationship between the tables is effectively obtained by using the graph neural network, the context information and relationship dependency are accurately captured, the internal features of the tables are learned by using the table data representation learning model, and thus the probability value of the business table belonging to different users is determined based on the internal features of the tables and the cross-table features, the accuracy of the determined probability value is improved, and the user associated with each business table is determined according to the probability value of each business table belonging to different users, which can effectively improve the accuracy of identifying the user associated with the business table.

[0098] Therefore, the scheme provided in the present application achieves the purpose of determining the user associated with the business table based on the internal features of the table and the cross-table features, realizes the technical effect of improving the accuracy of identifying the user associated with the business table, and solves the problem of low identification accuracy when identifying the user associated with the business table in the related art.

[0099] Optionally, in the table processing apparatus provided in the embodiment of the present application, the first construction module further includes: a first construction submodule, configured to take the business table as a node, construct an edge between the associated nodes based on the association relationship between the multiple business tables, and obtain a target graph structure; and a second construction submodule, configured to construct an adjacency matrix based on the target graph structure.

[0100] Optionally, in the table processing apparatus provided in the embodiment of the present application, the first processing module further includes: a first processing submodule, configured to process the adjacency matrix and the multiple business tables by using the graph neural network, and obtain the first vector representation of each business table; a second processing submodule, configured to process the multiple business tables by using the table data representation learning model, and obtain the second vector representation of each business table; a third processing submodule, configured to, for each business table, perform fusion processing on the first vector representation and the second vector representation of the business table, and obtain the target vector of the business table; and a fourth processing submodule, configured to process the target vector of each business table by using the full connection layer and the normalization exponential layer in the target recognition model, and obtain the probability value of each business table belonging to different users.

[0101] Optionally, in the table processing apparatus provided in the embodiment of the present application, the third processing submodule further comprises: a determination unit, configured to determine first hyperparameters corresponding to the graph neural network, and determine second hyperparameters corresponding to the table data representation learning model; and a processing unit, configured to perform weighted fusion on the first vector representation and the second vector representation of the business table according to the first hyperparameters and the second hyperparameters, to obtain a target vector.

[0102] Optionally, in the table processing apparatus provided in the embodiment of the present application, the first determination module further comprises: a first determination submodule, configured to determine, for each business table, a maximum probability value from the probability values corresponding to the business table; and a second determination submodule, configured to determine a user corresponding to the maximum probability value as a user associated with the business table.

[0103] Optionally, in the table processing apparatus provided in the embodiment of the present application, the table processing apparatus further comprises: a second acquisition module, configured to acquire a training sample set, wherein a training sample in the training sample set comprises a plurality of sample business tables and a plurality of sample adjacency matrices corresponding to the plurality of sample business tables, and a real label of the training sample represents a user to which each sample business table belongs; and a second processing module, configured to train the initial identification model based on the training sample set, to obtain the target identification model.

[0104] Optionally, in the table processing apparatus provided in the embodiment of the present application, the table processing apparatus further comprises: a second determination module, configured to determine a plurality of sample business tables from historical business tables in the target storage area, wherein the historical business tables are business tables for which the belonging users have been determined; a second construction module, configured to construct sample adjacency matrices corresponding to the plurality of sample business tables based on the association relationships between the plurality of sample business tables; and a third determination module, configured to determine training samples based on the plurality of sample business tables and the sample adjacency matrices corresponding to the plurality of sample business tables.

[0105] It should be noted that the first acquisition module 301, the first construction module 302, the first processing module 303 and the first determination module 304 correspond to steps S201 to S204 in Embodiment 1, and the four modules have the same instances and application scenarios as the corresponding steps, but are not limited to the content disclosed in Embodiment 1. It should be noted that the above modules or units can be hardware components or software components stored in a memory (for example, the memory 104) and processed by one or more processors (for example, the processors 102a, 102b, …, 102n), and the above modules can also be a part of the apparatus and can run in the computer terminal 10 provided in Embodiment 1.

[0106] Embodiment 3

[0107] The embodiments of the present application can provide an electronic device, Figure 4is a structural block diagram of an electronic device according to an embodiment of the present application. As shown in Figure 4 the electronic device can include one or more (only one is shown in the figure) processors 1002, a memory 1004, a storage controller, and a peripheral interface, wherein the peripheral interface is connected with a radio frequency module, an audio module, and a display. Figure 4

[0108] The memory can be used to store software programs and modules, such as program instructions / modules corresponding to the methods and devices in the embodiments of the present application. The processor executes various functions and data processing by running the software programs and modules stored in the memory, that is, implements the above-mentioned methods. The memory can include a high-speed random access memory, and can further include a non-volatile memory, such as one or more magnetic storage devices, flash memories, or other non-volatile solid-state memories. In some examples, the memory can further include a memory remotely arranged with respect to the processor, which can be connected to the terminal through a network. Examples of the above-mentioned network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0109] The processor can call information and application programs stored in the memory through the transmission device to perform the following steps: acquiring a plurality of business forms newly added in a target storage area, wherein a single business form is used to record the business information of a single user, and part of the plurality of business forms includes user information of the user, and the remaining business forms do not include user information of the user; constructing an adjacency matrix based on the association relationship between the plurality of business forms, wherein the elements in the adjacency matrix represent the association strength between two business forms, and the association strength represents the probability value that the two business forms belong to the same user; processing the adjacency matrix and the plurality of business forms by using a target recognition model to obtain the probability value that each business form belongs to different users, wherein the target recognition model includes a graph neural network and a table data representation learning model; determining the user associated with each business form according to the probability value that each business form belongs to different users.

[0110] The processor can also call information and application programs stored in the memory through the transmission device to perform the following steps: taking the business form as a node, constructing an edge between the associated nodes based on the association relationship between the plurality of business forms to obtain a target graph structure; and constructing an adjacency matrix based on the target graph structure.

[0111] ​The processor can further call information and application programs stored in the memory through the transmission device to perform the following steps: processing the adjacency matrix and the plurality of business tables through the graph neural network to obtain first vector representations of the business tables; processing the plurality of business tables using the table data representation learning model to obtain second vector representations of the business tables; for each business table, performing fusion processing on the first vector representation and the second vector representation of the business table to obtain a target vector of the business table; and processing the target vector of each business table through a fully connected layer and a normalization exponential layer in the target identification model to obtain probability values of the business table belonging to different users.

[0112] The processor can call information and application programs stored in the memory through the transmission device to perform the following steps: determining first hyperparameters corresponding to the graph neural network and second hyperparameters corresponding to the table data representation learning model; and performing weighted fusion on the first vector representation and the second vector representation of the business table according to the first hyperparameters and the second hyperparameters to obtain the target vector.

[0113] The processor can call information and application programs stored in the memory through the transmission device to perform the following steps: for each business table, determining a maximum probability value from the probability values corresponding to the business table; and determining a user corresponding to the maximum probability value as a user associated with the business table.

[0114] The processor can call information and application programs stored in the memory through the transmission device to perform the following steps: obtaining a training sample set, wherein a training sample in the training sample set includes a plurality of sample business tables and sample adjacency matrices corresponding to the sample business tables, and a real label of the training sample represents a user to which each sample business table belongs; training an initial identification model based on the training sample set to obtain the target identification model.

[0115] The processor can call information and application programs stored in the memory through the transmission device to perform the following steps: determining a plurality of sample business tables from historical business tables in the target storage area, wherein the historical business tables are business tables for which the belonging users have been determined; constructing sample adjacency matrices corresponding to the plurality of sample business tables based on association relationships between the plurality of sample business tables; and determining training samples based on the plurality of sample business tables and the sample adjacency matrices corresponding to the plurality of sample business tables.

[0116] In the embodiment of the present application, by constructing an adjacency matrix based on the association relationship between the newly added multiple business tables, the known complex relationship between the business tables is converted into a quantifiable adjacency matrix. By setting the target recognition model to include a graph neural network and a table data representation learning model, and processing the adjacency matrix and the multiple business tables by using the target recognition model, the probability value of each business table belonging to different users is obtained, the complex relationship between the tables is effectively obtained by using the graph neural network, the context information and relationship dependency are accurately captured, the internal features of the tables are learned by using the table data representation learning model, and thus the probability value of the business table belonging to different users is determined based on the internal features of the tables and the cross-table features, the accuracy of the determined probability value is improved, and the user associated with each business table is determined according to the probability value of each business table belonging to different users, which can effectively improve the accuracy of identifying the user associated with the business table.

[0117] Therefore, the scheme provided in the present application achieves the purpose of determining the user associated with the business table based on the internal features of the table and the cross-table features, realizes the technical effect of improving the accuracy of identifying the user associated with the business table, and solves the problem of low identification accuracy when identifying the user associated with the business table in the related art.

[0118] Those skilled in the art can understand that Figure 4 The structure shown is only schematic, and the electronic device can also be a smart phone (such as an Android phone, an iOS phone, etc.), a tablet computer, a palm computer, a Mobile Internet Device (MID), a PAD, or the like. Figure 4 It does not limit the structure of the electronic device. For example, the electronic device can further include more or fewer components (such as a network interface, a display device, etc.) than those shown in the figure, or have a different configuration from that shown in the figure. Figure 4 It does not limit the structure of the electronic device. For example, the electronic device can further include more or fewer components (such as a network interface, a display device, etc.) than those shown in the figure, or have a different configuration from that shown in the figure. Figure 4 It does not limit the structure of the electronic device. For example, the electronic device can further include more or fewer components (such as a network interface, a display device, etc.) than those shown in the figure, or have a different configuration from that shown in the figure.

[0119] Those skilled in the art can understand that all or part of the steps in the various methods of the above embodiments can be instructed by a program to complete the related hardware of the terminal device, and the program can be stored in a computer readable storage medium, which can include a flash disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, etc.

[0120] Embodiment 4

[0121] The embodiment of the present application further provides a storage medium. Optionally, in the embodiment, the storage medium can be used to save the program code executed by the table processing method provided in the first embodiment.

[0122] Optionally, in the embodiment, the storage medium can be located in any one of computer terminals in a computer terminal group in a computer network, or in any one of mobile terminals in a mobile terminal group.

[0123] The present application further provides a computer program product, which is adapted to execute the steps of the table processing method when executed on a data processing device.

[0124] The serial numbers of the embodiments of the present application are only for description, and do not represent the advantages or disadvantages of the embodiments.

[0125] In the above embodiments of the present application, the description of each embodiment has its own focus, and the parts not described in detail in a certain embodiment can be referred to the relevant description of other embodiments.

[0126] In the several embodiments of the present application, it should be understood that the disclosed technology can be implemented in other ways. Of course, the above-described device embodiments are only schematic. For example, the division of units is only a logical function division, and there can be another division manner in actual implementation. For example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed mutual coupling or direct coupling or communication connection between units can be indirect coupling or communication connection through some interface, unit or module, and can be electrical or other forms.

[0127] The units described as separate components can or can not be physically separate, and the components displayed as units can or can not be physical units, that is, they can be located in one place, or can be distributed on a plurality of network units. According to actual needs, part or all of the units can be selected to achieve the purpose of the embodiment.

[0128] In addition, each functional unit in each embodiment of the present application can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of software functional unit.

[0129] The integrated unit, if implemented in the form of a software function unit and sold or used as an independent product, can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application essentially or say the part that contributes to the prior art or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server or a network device, etc.) to execute all or part of the steps of the method described in the various embodiments of the present application. The aforementioned storage medium includes: a U disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a mobile hard disk, a magnetic disk or an optical disk, and various media that can store program codes.

[0130] The above is only the preferred embodiment of the present application, and it should be pointed out that for those skilled in the art, without departing from the principles of the present application, a number of improvements and refinements can be made, which should be considered as the protection scope of the present application.

Claims

1. A table processing method, characterized in that: include: Acquire multiple newly added business tables in the target storage area, wherein a single business table is used to record business information of a single user, some of the multiple business tables include user information of the user, and the remaining business tables do not include user information of the user; Constructing an adjacency matrix based on the association relationships between the multiple business tables, including: using the business tables as nodes, and constructing edges between the associated nodes based on the association relationships between the multiple business tables to obtain a target graph structure; constructing the adjacency matrix based on the target graph structure, wherein elements in the adjacency matrix represent the association strength between two business tables, and the association strength represents the probability value that the two business tables belong to the same user; The adjacency matrix and the plurality of business tables are processed using a target recognition model to obtain a probability value that each business table belongs to a different user, including: Processing the adjacency matrix and the plurality of business tables through a graph neural network to obtain a first vector representation of each business table, wherein the target recognition model includes the graph neural network and a table data representation learning model; Processing the plurality of business tables using the table data representation learning model to obtain a second vector representation of each business table; For each business table, determining a first hyperparameter corresponding to the graph neural network and a second hyperparameter corresponding to the table data representation learning model; performing weighted fusion on the first vector representation and the second vector representation of the business table based on the first hyperparameter and the second hyperparameter to obtain a target vector for the business table; The target vectors of each business table are processed by the fully connected layer and the normalized index layer in the target recognition model to obtain the probability value of each business table belonging to a different user; The user associated with each business table is determined based on the probability value that each business table belongs to a different user.

2. The method according to claim 1, characterized in that Based on the probability values ​​of each business table belonging to different users, the user associated with each business table is determined, including: For each business table, determining a maximum probability value from the probability values ​​corresponding to the business table; The user corresponding to the maximum probability value is determined as the user associated with the business table.

3. The method according to claim 1, characterized in that The target recognition model is trained in the following way: Acquire a training sample set, wherein the training samples in the training sample set include a plurality of sample business tables and a sample adjacency matrix corresponding to the plurality of sample business tables, and the true labels of the training samples represent the users to which each sample business table belongs; An initial recognition model is trained based on the training sample set to obtain the target recognition model.

4. The method according to claim 3, characterized in that The training samples are obtained in the following way: Determine a plurality of sample business tables from the historical business tables in the target storage area, wherein the historical business tables are business tables of users to whom the tables have been determined to belong; Constructing a sample adjacency matrix corresponding to the plurality of sample business tables based on the association relationship between the plurality of sample business tables; The training samples are determined based on the multiple sample business tables and sample adjacency matrices corresponding to the multiple sample business tables.

5. A table processing device, characterized in that: include: a first acquisition module, configured to acquire a plurality of newly added business tables in a target storage area, wherein a single business table is used to record business information of a single user, some of the plurality of business tables include user information of the user, and the remaining business tables do not include user information of the user; A first construction module, configured to construct an adjacency matrix based on the association relationships between the multiple business tables, comprising: a first construction submodule, configured to use the business tables as nodes and construct edges between the associated nodes based on the association relationships between the multiple business tables to obtain a target graph structure; a second construction submodule, configured to construct the adjacency matrix based on the target graph structure, wherein an element in the adjacency matrix represents the strength of association between two business tables, and the strength of association represents the probability value that the two business tables belong to the same user; The first processing module is configured to process the adjacency matrix and the plurality of business tables using a target recognition model to obtain a probability value of each business table belonging to a different user, including: a first processing submodule, configured to process the adjacency matrix and the plurality of business tables through a graph neural network to obtain a first vector representation of each business table, wherein the target recognition model includes the graph neural network and the table data representation learning model; a second processing submodule, configured to process the plurality of business tables using the table data representation learning model to obtain a second vector representation of each business table; a determination unit configured to determine, for each business table, a first hyperparameter corresponding to the graph neural network and a second hyperparameter corresponding to the table data representation learning model; a processing unit configured to perform weighted fusion on the first vector representation and the second vector representation of the business table based on the first hyperparameter and the second hyperparameter to obtain a target vector for the business table; a fourth processing submodule, configured to process the target vectors of each business table through the fully connected layer and the normalized index layer in the target recognition model to obtain a probability value that each business table belongs to a different user; The first determining module is configured to determine the user associated with each business table according to the probability value that each business table belongs to a different user.

6. A computer-readable storage medium, characterized in that The computer-readable storage medium includes a stored executable program, wherein when the executable program is run, the device where the computer-readable storage medium is located is controlled to execute the table processing method according to any one of claims 1 to 4.

7. An electronic device, characterized in that: include: a memory storing an executable program; A processor is used to run the program, wherein the program executes the table processing method according to any one of claims 1 to 4 when running.

8. A computer program product comprising computer instructions, characterized in that When the computer instructions are executed by a processor, the steps of the table processing method according to any one of claims 1 to 4 are implemented.

Images