Wide table recommendation method and device, equipment, medium and product

Abstract

The invention discloses a wide table recommendation method and device, equipment, a medium and a product, and relates to the technical field of computers.The wide table recommendation method comprises the steps that a wide table data set and a target wide table in the wide table data set are obtained; constructing a wide table graph model according to the wide table data set; and performing message passing based on the wide table graph model, and combining with the target wide table to determine and recommend a to-be-recommended wide table. The wide table graph model is constructed through the wide table data set, the deep dependency relationship between the wide tables is captured, and the to-be-recommended wide table is determined based on the wide table graph model, so that the method is suitable for a large-scale wide table recommendation scene, can dynamically adapt to the change of a data structure, and is particularly suitable for a financial service scene with frequent structure change; and the search speed and the recommendation accuracy are greatly improved.

Description

Technical Field

[0001] This invention relates to the field of computer technology, and in particular to a method, apparatus, device, medium, and product for recommending wide tables. Background Technology

[0002] With the continuous development of big data technology, financial enterprises have accumulated massive amounts of data assets. This vast amount of business data, scattered across various business systems, is accessed by data lake warehouses and is typically stored in the form of wide tables. A wide table is a flat data structure containing numerous fields (columns), suitable for multi-dimensional and multi-indicator analysis scenarios. For example, in the analysis of a bank's customer financial assets, a single wide table may contain information from multiple business themes. While wide tables offer advantages in high integration and ease of analysis, their high-dimensional data characteristics and complex inter-table relationships also present challenges in data management and recommendation.

[0003] Existing wide-table recommendation mechanisms mainly rely on manual annotation, business personnel experience, cross-departmental consultation, keyword retrieval, and simple rule-based and statistical analysis.

[0004] However, due to the sheer number of wide tables and their numerous fields, manual annotation often relies on developers' habits and business knowledge. Users often cannot quickly and accurately locate the wide tables needed for their business scenarios. Methods that rely on predefined rules (keyword matching, table usage frequency) or simple statistical measures (similarity coefficients) can only handle the surface features between wide tables and struggle to leverage the deeper relationships between the data. Summary of the Invention

[0005] This invention provides a method, apparatus, device, medium, and product for wide table recommendations, in order to improve the accuracy of wide table recommendations.

[0006] According to a first aspect of the present invention, a wide table recommendation method is provided, comprising:

[0007] Obtain the wide table dataset and the target wide table in the wide table dataset;

[0008] Based on the wide table dataset, construct a wide table graph model;

[0009] Based on the wide table graph model, message passing is combined with the target wide table to determine the wide table to be recommended and then make a recommendation.

[0010] According to a second aspect of the present invention, a wide table recommendation device is provided, comprising:

[0011] The wide table acquisition module is used to acquire the wide table dataset and the target wide table in the wide table dataset.

[0012] The model building module is used to build a wide table graph model based on the wide table dataset;

[0013] The wide table recommendation module is used to determine the wide table to be recommended and make recommendations based on the target wide table by combining the message passing of the wide table graph model.

[0014] According to a third aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0015] At least one processor; and

[0016] A memory communicatively connected to the at least one processor; wherein,

[0017] The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the wide table recommendation method according to any embodiment of the present invention.

[0018] According to a fourth aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions for causing a processor to execute and implement the wide table recommendation method according to any embodiment of the present invention.

[0019] According to a fifth aspect of the present invention, embodiments of the present invention also provide a computer program product, the computer program product including a computer program that, when executed by a processor, implements the wide table recommendation method of any embodiment of the present invention.

[0020] The technical solution of this invention involves acquiring a wide table dataset and a target wide table within that dataset; constructing a wide table graph model based on the wide table dataset; and using message passing based on the wide table graph model in conjunction with the target wide table to determine and recommend the wide table to be recommended. By constructing a wide table graph model from the wide table dataset, deep dependencies between wide tables are captured, and the wide table to be recommended is determined based on this model. This approach is suitable for large-scale wide table recommendation scenarios and can dynamically adapt to changes in data structure, especially in financial business scenarios with frequent structural changes, significantly improving search speed and recommendation accuracy.

[0021] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 This is a flowchart of a wide table recommendation method provided in Embodiment 1 of the present invention;

[0024] Figure 2 This is a flowchart of a wide table recommendation method provided according to Embodiment 2 of the present invention;

[0025] Figure 3 This is a schematic diagram of a wide table recommendation device provided according to Embodiment 3 of the present invention;

[0026] Figure 4 This is a schematic diagram of the structure of an electronic device that implements an embodiment of the present invention. Detailed Implementation

[0027] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0028] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0029] Example 1

[0030] Figure 1The flowchart illustrates a wide table recommendation method provided in Embodiment 1 of the present invention. This embodiment is applicable to situations involving wide table recommendations. The method can be executed by a wide table recommendation device, which can be implemented in hardware and / or software and can be configured in an electronic device. Figure 1 As shown, the method includes:

[0031] S110. Obtain the wide table dataset and the target wide table in the wide table dataset.

[0032] In this embodiment, the wide table dataset can be understood as a collection of multiple wide table datasets. A wide table can be understood as a logical and consistent definition of enterprise business concepts or information system data within a data lake warehouse, forming a target table with a flat structure and numerous fields, suitable for multi-dimensional aggregation analysis, but also increasing the complexity of management and recommendation. The target wide table can be understood as selecting the most suitable wide table for recommendation, which can be done by relevant personnel.

[0033] Specifically, the processor can obtain the target wide table selected by the relevant personnel, as well as the wide table dataset to which the target wide table belongs.

[0034] S120. Construct a wide table graph model based on the wide table dataset.

[0035] In this embodiment, the wide graph model can be understood as a deep learning model based on graph structure for processing graph data. For example, it can be a graph neural network (GNN) model. Unlike traditional neural networks, GNNs can transmit and aggregate information through the relationships between nodes and edges in the graph, thereby capturing high-dimensional features and complex dependency structures between nodes.

[0036] Specifically, during the formation of wide tables, different data may be generated, processed, and transferred to create wide tables. This means that different wide tables may have processing dependencies and data lineage relationships. These dependencies provide multi-dimensional evidence for similarity evaluation during recommendation. The processor can determine nodes and edges based on the dependencies between each wide table in the wide table dataset, thereby constructing a wide table graph model.

[0037] S130. Based on the wide table graph model, message passing is combined with the target wide table to determine the wide table to be recommended and then make a recommendation.

[0038] In this embodiment, message passing can be understood as the process by which each node in the wide table graph model exchanges information with its neighboring nodes to update its own state. The wide table to be recommended can be understood as the wide table that is most similar to the target wide table and can be recommended.

[0039] Specifically, the processor can perform message passing based on the process by which each node in the wide table graph model updates its own state by exchanging information with its neighboring nodes, thereby determining the embedding vector corresponding to each wide table. By calculating and ranking the similarity between the embedding vector of the target wide table and the embedding vectors of other wide tables, one or more wide tables that are most similar to the target wide table are identified as wide tables to be recommended and then recommended.

[0040] The technical solution of this invention involves acquiring a wide table dataset and a target wide table within that dataset; constructing a wide table graph model based on the wide table dataset; and using message passing based on the wide table graph model in conjunction with the target wide table to determine and recommend the wide table to be recommended. By constructing a wide table graph model from the wide table dataset, deep dependencies between wide tables are captured, and the wide table to be recommended is determined based on this model. This approach is suitable for large-scale wide table recommendation scenarios and can dynamically adapt to changes in data structure, especially in financial business scenarios with frequent structural changes, significantly improving search speed and recommendation accuracy.

[0041] Example 2

[0042] Figure 2 This is a flowchart of a wide table recommendation method provided in Embodiment 2 of the present invention. This embodiment is a further refinement of the above embodiment. Figure 2 As shown, the method includes:

[0043] S201. Obtain the wide table dataset and the target wide table in the wide table dataset.

[0044] S202. Based on the wide table dependency data of each wide table in the wide table dataset, construct the wide table nodes and edges.

[0045] In this embodiment, wide table dependency data can be understood as describing the path and transformations that data undergoes from its source to its final use, reflecting the dependencies between wide table metadata. Analyzing these relationships can reveal the inherent structure, patterns, and evolutionary laws of the data, providing fundamental support for data processing, modeling, and decision-making. Wide table nodes can be understood as representing each wide table in the model. Edges can be understood as representing the relationships or interactions between wide table nodes in the model, serving as key components for transmitting node information and constructing the graph structure. Edges not only carry the connection states between nodes but can also carry weights or attributes, directly affecting the direction, strength, and content of information transmission in the GNN.

[0046] Specifically, the wide table dataset includes not only the wide tables themselves but also the wide table dependency data for each wide table. The processor can analyze the wide table dependency data to treat each wide table as a wide table node and determine the relationships between each wide table to obtain edges.

[0047] For example, the format of wide table data can be [table name, field sequence, next level table]. The "next level table" can be used to determine the wide table dependency data between different wide tables. The wide table dependency data can be used to determine the hierarchical relationship between different wide tables and the edges between nodes of different wide tables.

[0048] S203. Construct word frequency vectors based on the wide table field data of each wide table.

[0049] In this embodiment, the wide table field data can be understood as the specific values ​​of the field sequence in the wide table. The word frequency vector can be understood as a feature vector generated based on the field content, reflecting the frequency of the field's occurrence in the wide table, and is used to initialize the features of nodes in the graph neural network.

[0050] Specifically, the processor can generate word frequency vectors based on the data content of the wide table fields of each wide table.

[0051] S204. Based on each wide table node, each edge, and each word frequency vector, construct the wide table graph model corresponding to the wide table dataset.

[0052] Specifically, the processor can construct a graph model based on each wide table node and its corresponding edge, and initialize the node features of each wide table node with word frequency vectors to obtain the wide table graph model.

[0053] S205. Based on the wide table graph model, perform message passing and determine the embedding vector of each wide table in the wide table dataset.

[0054] In this embodiment, the embedding vector can be understood as a low-dimensional dense vector representation, which typically maps high-dimensional sparse features to a low-dimensional space to facilitate similarity calculation.

[0055] Specifically, the processor can perform message passing based on the wide table graph model to determine the embedding vector of each wide table in the wide table dataset.

[0056] Furthermore, based on the above embodiments, the steps for determining the embedding vectors of each wide table in the wide table dataset through message passing based on the wide table graph model can be refined as follows:

[0057] Based on the graph structure in the wide table graph model, the adjacency matrix is ​​determined; message passing is performed on the adjacency matrix using the graph convolutional layer of the wide table graph model to obtain the embedding vector of each wide table.

[0058] In this embodiment, a graph structure can be understood as a data structure composed of nodes and edges. An adjacency matrix can be understood as a numerical representation of a graph structure, used to describe the connection relationships between nodes. A graph convolutional layer can be understood as a processing layer used for feature extraction and information aggregation of graph structure data.

[0059] Specifically, the adjacency matrix is ​​determined based on the graph structure in the wide table graph model; message passing is performed on the adjacency matrix using the graph convolutional layer of the wide table graph model to obtain the embedding vector for each wide table.

[0060] For example, the adjacency matrix A can be defined as: A∈R M×M Represents a graph structure. A ij This represents the edge between wide table i and wide table j. The processing of a graph convolutional layer may include: first, normalizing the adjacency matrix: Where A represents the original adjacency matrix, D represents the degree matrix, and I represents the identity matrix. Next, a graph convolution layer is applied to the normalized adjacency matrix. The graph convolution formula is: in, It is the normalized adjacency matrix, H (L) W is the node feature of layer L. (L) These are learnable parameters, and σ is a non-linear activation function. After L layers of graph convolution, the embedding vector H of the wide table can be obtained. (L) ∈R M×T .

[0061] S206. Based on each embedding vector and the target embedding vector of the target wide table, determine the wide table to be recommended and make recommendations.

[0062] In this embodiment, the target embedding vector can be understood as the embedding vector corresponding to the target wide table after graph convolutional layer processing.

[0063] Specifically, the processor can determine the similarity between each embedding vector and the target embedding vector of the target wide table and sort them. It can also extract the top one or several wide tables in the sorted list as the wide tables to be recommended according to a set number, and then make recommendations based on the selected wide tables.

[0064] Furthermore, based on the above embodiments, the steps of determining the recommended wide table and making recommendations based on each embedding vector and the target embedding vector of the target wide table can be refined as follows:

[0065] Construct similarity matrices for each embedding vector and the target embedding vector; determine the similarity between each wide table and the target wide table; and based on the similarity scores, determine the wide table to be recommended and make recommendations.

[0066] In this embodiment, the similarity matrix can be understood as a quantitative representation of the similarity between each wide table in matrix form. The similarity score can be understood as a numerical value used to characterize the degree of similarity between two matrices.

[0067] Specifically, the processor can construct a similarity matrix between each embedding vector and the target embedding vector, and determine the similarity between each wide table and the target wide table using methods such as cosine similarity algorithm. The processor can sort the wide tables according to their similarity, and extract the recommended wide tables from the sorting results according to a set number of outputs. For example, it can output the TOP-N sorting results as the recommended wide tables, and find one or more wide tables that are most similar to the target wide table as the recommended wide tables.

[0068] The technical solution of this invention constructs a wide table graph model based on the wide table's dependent data, transforming the wide table data into a graph structure. This effectively captures the content features and structural information of the wide table, providing automated assistance to users for efficient wide table recommendation. The overall algorithm flow involves constructing a wide table graph model using the graph structure, performing feature learning and embedding optimization on the wide table nodes using this model, and using graph embedding techniques to perform graph convolution operations, passing messages and updating the node embedding vectors, jointly learning the features of the wide table nodes and edge relationships. Similarity is calculated based on the embedding vectors, a similarity score is calculated, and the wide table to be recommended is obtained. This achieves efficient encoding and similarity calculation of the wide table, significantly improving search speed and recommendation accuracy. It is suitable for large-scale wide table recommendation scenarios and can dynamically adapt to changes in data structure, especially in financial business scenarios with frequent structural changes.

[0069] Example 3

[0070] Figure 3 This is a schematic diagram of a wide-table recommendation device provided in Embodiment 3 of the present invention. Figure 3 As shown, the device includes: a wide table acquisition module 31, a model construction module 32, and a wide table recommendation module 33.

[0071] Wide table acquisition module 31 is used to acquire wide table dataset and target wide table in wide table dataset;

[0072] Model building module 32 is used to build a wide table graph model based on the wide table dataset;

[0073] The wide table recommendation module 33 is used to determine the wide table to be recommended and make recommendations based on the target wide table by combining the message passing of the wide table graph model.

[0074] The technical solution of this invention involves acquiring a wide table dataset and a target wide table within that dataset; constructing a wide table graph model based on the wide table dataset; and using message passing based on the wide table graph model in conjunction with the target wide table to determine and recommend the wide table to be recommended. By constructing a wide table graph model from the wide table dataset, deep dependencies between wide tables are captured, and the wide table to be recommended is determined based on this model. This approach is suitable for large-scale wide table recommendation scenarios and can dynamically adapt to changes in data structure, especially in financial business scenarios with frequent structural changes, significantly improving search speed and recommendation accuracy.

[0075] Furthermore, the model building module 32 is specifically used for:

[0076] Based on the wide table dependency data of each wide table in the wide table dataset, construct the wide table nodes and edges;

[0077] Construct word frequency vectors based on the wide table field data of each wide table;

[0078] Based on each wide table node, each edge, and each word frequency vector, construct a wide table graph model corresponding to the wide table dataset.

[0079] Furthermore, the wide table recommendation module 33 includes:

[0080] The first determining unit is used to perform message passing based on the wide table graph model and determine the embedding vector of each wide table in the wide table dataset;

[0081] The second determining unit is used to determine the wide table to be recommended and make recommendations based on each of the embedding vectors and the target embedding vector of the target wide table.

[0082] Specifically, the first determining unit is used for:

[0083] Determine the adjacency matrix based on the graph structure in the wide graph model;

[0084] The adjacency matrix is ​​message-passed using the graph convolutional layer of the wide table graph model to obtain the embedding vector for each wide table.

[0085] Specifically, the second determining unit is used for:

[0086] Construct a similarity matrix between each of the embedding vectors and the target embedding vector;

[0087] Determine the similarity between each of the wide tables and the target wide table;

[0088] Based on the aforementioned similarity, a wide table to be recommended is determined and then recommended.

[0089] The wide table recommendation device provided in the embodiments of the present invention can execute the wide table recommendation method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the method execution.

[0090] Example 4

[0091] Figure 4 A schematic diagram of an electronic device 40 that can be used to implement embodiments of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0092] like Figure 4 As shown, the electronic device 40 includes at least one processor 41 and a memory, such as a read-only memory (ROM) 42 or a random access memory (RAM) 43, communicatively connected to the at least one processor 41. The memory stores computer programs executable by the at least one processor. The processor 41 can perform various appropriate actions and processes based on the computer program stored in the ROM 42 or loaded into the RAM 43 from storage unit 48. The RAM 43 may also store various programs and data required for the operation of the electronic device 40. The processor 41, ROM 42, and RAM 43 are interconnected via a bus 44. An input / output (I / O) interface 45 is also connected to the bus 44.

[0093] Multiple components in electronic device 40 are connected to I / O interface 45, including: input unit 46, such as keyboard, mouse, etc.; output unit 47, such as various types of monitors, speakers, etc.; storage unit 48, such as disk, optical disk, etc.; and communication unit 49, such as network card, modem, wireless transceiver, etc. Communication unit 49 allows electronic device 40 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0094] Processor 41 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 41 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 41 performs the various methods and processes described above, such as the wide table recommendation method.

[0095] In some embodiments, the wide table recommendation method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 48. In some embodiments, part or all of the computer program may be loaded and / or mounted on electronic device 40 via ROM 42 and / or communication unit 49. When the computer program is loaded into RAM 43 and executed by processor 41, one or more steps of the wide table recommendation method described above may be performed. Alternatively, in other embodiments, processor 41 may be configured to perform the wide table recommendation method by any other suitable means (e.g., by means of firmware).

[0096] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0097] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0098] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0099] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0100] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or computing systems that include middleware components (e.g., application servers), or computing systems that include frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0101] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0102] In one embodiment, the present invention further includes a computer program product comprising a computer program that, when executed by a processor, implements the wide table recommendation method of any embodiment of the present invention.

[0103] In implementing the computer program product, computer program code for performing the operations of this invention can be written in one or more programming languages ​​or a combination thereof. Programming languages ​​include object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0104] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0105] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A wide-table recommendation method, characterized in that, include: Obtain the wide table dataset and the target wide table in the wide table dataset; Based on the wide table dataset, construct a wide table graph model; Based on the wide table graph model, message passing is combined with the target wide table to determine the wide table to be recommended and then make a recommendation.

2. The method according to claim 1, characterized in that, The step of constructing a wide table graph model based on the wide table dataset includes: Based on the wide table dependency data of each wide table in the wide table dataset, construct the wide table nodes and edges; Construct word frequency vectors based on the wide table field data of each wide table; Based on each wide table node, each edge, and each word frequency vector, construct a wide table graph model corresponding to the wide table dataset.

3. The method according to claim 1, characterized in that, The step of combining the target wide table with message passing based on the wide table graph model to determine the wide table to be recommended and then recommending it includes: Message passing is performed based on the wide table graph model to determine the embedding vector of each wide table in the wide table dataset; Based on the embedding vectors and the target embedding vector of the target wide table, the wide table to be recommended is determined and recommended.

4. The method according to claim 3, characterized in that, The step of message passing based on the wide table graph model, and determining the embedding vector of each wide table in the wide table dataset, includes: Determine the adjacency matrix based on the graph structure in the wide graph model; The adjacency matrix is ​​message-passed using the graph convolutional layer of the wide table graph model to obtain the embedding vector for each wide table.

5. The method according to claim 3, characterized in that, The step of determining and recommending a wide table based on each of the embedding vectors and the target embedding vector of the target wide table includes: Construct a similarity matrix between each of the embedding vectors and the target embedding vector; Determine the similarity between each of the wide tables and the target wide table; Based on the aforementioned similarity, a wide table to be recommended is determined and then recommended.

6. A wide-table recommendation device, characterized in that, include: The wide table acquisition module is used to acquire the wide table dataset and the target wide table in the wide table dataset. The model building module is used to build a wide table graph model based on the wide table dataset; The wide table recommendation module is used to determine the wide table to be recommended and make recommendations based on the target wide table by combining the message passing of the wide table graph model.

7. The apparatus according to claim 6, characterized in that, The model building module is specifically used for: Based on the wide table dependency data of each wide table in the wide table dataset, construct the wide table nodes and edges; Construct word frequency vectors based on the wide table field data of each wide table; Based on each wide table node, each edge, and each word frequency vector, construct a wide table graph model corresponding to the wide table dataset.

8. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor to enable the at least one processor to perform the wide table recommendation method according to any one of claims 1-5.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that cause a processor to execute the wide table recommendation method according to any one of claims 1-5.

10. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the wide table recommendation method according to any one of claims 1-5.

Images