Data model construction method and device, equipment, storage medium and product

Abstract

The invention discloses a data model construction method and device, equipment, a storage medium and a product, and relates to the technical field of data processing, and the method comprises the steps: obtaining industry common data provided by a server, and constructing industry models of a plurality of industries according to the industry common data, industry features and business demands, and cross-industry data sharing and fusion are realized, and a data island is broken. According to the method, the industry model is extracted to obtain the multi-industry labels, and the multi-industry labels are added into the central label library corresponding to the server, so that the accuracy and applicability of the data model are ensured. When a label request of a new user is received, the target industry label is obtained from the central label library, the target data model is constructed according to the target industry label, the construction process of the data model is simplified through the use of the central label library, the data access efficiency is improved, meanwhile, the system can flexibly adapt to the requirements of different users and industries, and the user experience is improved. And the method has good expansibility.

Description

Technical Field

[0001] This application relates to the technical field of data processing, and particularly to a method, apparatus, device, storage medium, and product for constructing a data model. Background Art

[0002] With the rapid development of information technology and the explosive growth of data volume, the sharing and integration of data have become important factors driving innovation. However, existing data models are often limited to a single organization or field, making it difficult to achieve cross-organization and cross-industry data sharing and integration, resulting in "data islands".

[0003] Currently, traditional data management systems usually rely on a centralized architecture and are difficult to adapt to changing business requirements and complex industry environments.

[0004] Therefore, how to achieve cross-organization and cross-industry data sharing and integration is an urgent problem to be solved at present. Summary of the Invention

[0005] The main purpose of this application is to provide a method, apparatus, device, storage medium, and product for constructing a data model, aiming to solve the technical problem that current data management is difficult to achieve cross-organization and cross-industry data sharing and integration.

[0006] To achieve the above purpose, this application proposes a method for constructing a data model, and the method includes:

[0007] Obtain industry-common data provided by a server, and construct industry models for multiple industries according to the industry-common data, industry characteristics, and business requirements;

[0008] Refine the industry models to obtain multi-industry tags, and add the multi-industry tags to the corresponding central tag library of the server;

[0009] When receiving a tag request from a new user, obtain target industry tags from the central tag library, and construct a target data model according to the target industry tags.

[0010] In one embodiment, the step of constructing industry models for multiple industries according to the industry-common data, industry characteristics, and business requirements includes:

[0011] Establish a knowledge graph according to the industry-common data, industry characteristics, and business requirements;

[0012] Use a preset convolutional network to update the knowledge graph to obtain an updated knowledge graph;

[0013] Based on the updated knowledge graph, construct industry models for each industry.

[0014] In one embodiment, the step of establishing a knowledge graph based on the industry common data, industry characteristics, and business requirements includes:

[0015] Extract semantic tags from the industry common data, industry characteristics, and business requirements through a semantic analysis model;

[0016] Based on the semantic tags, establish connections between entities through semantic similarity and cosine similarity to obtain a knowledge graph.

[0017] In one embodiment, the step of extracting semantic tags from the industry common data, industry characteristics, and business requirements through a semantic analysis model includes:

[0018] Convert the industry common data, industry characteristics, and business requirements into current features, global features, and key features;

[0019] Decode the current features, global features, and key features through a decoder to obtain semantic tags.

[0020] In one embodiment, the semantic tags include keywords, themes, subjects, sentiment tendencies, and events. The step of establishing connections between entities through semantic similarity and cosine similarity based on the semantic tags to obtain a knowledge graph includes:

[0021] Judge whether the keywords, the subjects, and the sentiment tendencies corresponding to different entities are the same. If they are the same, establish a first connection between the corresponding two entities;

[0022] Calculate the first cosine similarity of the themes corresponding to different entities. If the first cosine similarity reaches a preset threshold, establish a second connection between the corresponding two entities;

[0023] Calculate the second cosine similarity of the events corresponding to different entities. If the second cosine similarity reaches a preset threshold, establish a third connection between the corresponding two entities;

[0024] Based on the first connection, the second connection, and the third connection, obtain a knowledge graph.

[0025] In one embodiment, the step of updating the knowledge graph using a preset convolutional network to obtain an updated knowledge graph includes:

[0026] Calculate attention weights according to the current features and the global features;

[0027] Perform weighted summation on the key features through the attention weights to obtain an updated feature representation;

[0028] Update the knowledge graph based on the updated feature representation to obtain an updated knowledge graph.

[0029] In one embodiment, before the step of obtaining target industry tags from the central tag library and constructing a target data model according to the target industry tags, the method further includes:

[0030] Combine private data, and generate new industry tags according to real-time industry characteristics, business requirements, and the industry common data.

[0031] Add the new industry tags to the central tag library to obtain an updated central tag library.

[0032] In addition, to achieve the above object, the present application also proposes a data model construction device, where the data model construction device includes:

[0033] An initial model construction module, configured to obtain industry common data provided by a server, and construct industry models for multiple industries according to the industry common data, industry characteristics, and business requirements.

[0034] An industry tag refinement module, configured to refine the industry models to obtain multi-industry tags, and add the multi-industry tags to the corresponding central tag library of the server.

[0035] A target model construction module, configured to, when receiving a tag request from a new user, obtain target industry tags from the central tag library, and construct a target data model according to the target industry tags.

[0036] In addition, to achieve the above object, the present application also proposes a data model construction device, where the device includes: a memory, a processor, and a computer program stored on the memory and executable on the processor, and the computer program is configured to implement the steps of the data model construction method as described above.

[0037] In addition, to achieve the above object, the present application also proposes a storage medium, where the storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium, and when the computer program is executed by a processor, the steps of the data model construction method as described above are implemented.

[0038] In addition, to achieve the above object, the present application also provides a computer program product, where the computer program product includes a computer program, and when the computer program is executed by a processor, the steps of the data model construction method as described above are implemented.

[0039] One or more technical solutions proposed by the present application have at least the following technical effects:

[0040] Obtain the industry - common data provided by the server, and construct industry models for multiple industries based on the industry - common data, industry characteristics, and business requirements; refine the industry models to obtain multi - industry tags, and add the multi - industry tags to the corresponding central tag library of the server; when receiving a tag request from a new user, obtain the target industry tags from the central tag library, and construct a target data model based on the target industry tags. By obtaining the industry - common data provided by the server and constructing industry models for multiple industries in combination with industry characteristics and business requirements, cross - industry data sharing and integration are achieved, breaking the "data silos". Refining the industry models to obtain multi - industry tags and adding them to the central tag library ensures the accuracy and applicability of the data model, and can better meet the needs of different users. When receiving a tag request from a new user, obtaining the target industry tags from the central tag library and constructing a target data model based on them enables the system to flexibly adapt to the needs of different users and industries, with good scalability. Through the use of the central tag library, the process of constructing the data model is simplified, and the efficiency of data access is improved. Brief Description of the Drawings

[0041] The drawings here are incorporated into the specification and form a part of this specification, showing embodiments consistent with the present application, and are used together with the specification to explain the principles of the present application.

[0042] To more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, for those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0043] Figure 1 It is a schematic flowchart of the first embodiment of the data model construction method of the present application;

[0044] Figure 2 It is a schematic flowchart of the second embodiment of the data model construction method of the present application;

[0045] Figure 3 It is a schematic module structure diagram of the data model construction device in the embodiment of the present application;

[0046] Figure 4 It is a schematic device structure diagram of the hardware operating environment involved in the data model construction method in the embodiment of the present application.

[0047] The realization of the purpose, functional characteristics, and advantages of the present application will be further described with reference to the embodiments and the drawings. Detailed Embodiments

[0048] It should be understood that the specific embodiments described herein are merely used to explain the technical solutions of the present application and are not used to limit the present application.

[0049] To better understand the technical solutions of the present application, the following will be described in detail in conjunction with the accompanying drawings of the specification and specific implementation manners.

[0050] Since traditional data management systems usually rely on a centralized architecture, lack flexibility and scalability, and are difficult to adapt to changing business requirements and complex industry environments. Especially when it comes to data sharing among multiple organizations or industries, data security, privacy, and consistency become major challenges. Existing data models are often limited to a single organization or domain and are difficult to achieve cross-organization and cross-industry data sharing and integration. In addition, existing systems often face performance bottlenecks when dealing with large-scale data and cannot efficiently support real-time data access and analysis. Therefore, how to achieve cross-organization and cross-industry data sharing and integration is an urgent problem to be solved at present.

[0051] The present application provides a solution. Obtain the industry-common data provided by the server, and construct industry models for multiple industries according to the industry-common data, industry characteristics, and business requirements; refine the industry models to obtain multi-industry tags, and add the multi-industry tags to the corresponding central tag library of the server; when receiving a tag request from a new user, obtain the target industry tags from the central tag library, and construct a target data model according to the target industry tags. By obtaining the industry-common data provided by the server and constructing industry models for multiple industries in combination with industry characteristics and business requirements, cross-industry data sharing and integration are achieved, breaking the "data silos". Refining the industry models to obtain multi-industry tags and adding them to the central tag library ensures the accuracy and applicability of the data model and can better meet the needs of different users. When receiving a tag request from a new user, obtaining the target industry tags from the central tag library and constructing a target data model according to them enables the system to flexibly adapt to the needs of different users and industries and has good scalability. By using the central tag library, the construction process of the data model is simplified and the efficiency of data access is improved.

[0052] Based on this, the embodiments of the present application provide a method for constructing a data model, referring to Figure 1 , Figure 1 is a schematic flowchart of the first embodiment of the method for constructing a data model of the present application.

[0053] In this embodiment, the method for constructing a data model includes steps S10 to S30:

[0054] Step S10, obtain the industry-common data provided by the server, and construct industry models for multiple industries according to the industry-common data, industry characteristics, and business requirements.

[0055] It should be noted that the industry-common data can be data that is common to multiple industries, such as basic statistical data, market trends, consumer behavior, etc. What is manifested in the data can include industry types, document titles, and document contents, etc. Industry characteristics can be unique attributes or features of a specific industry, such as the operation mode of the industry, market demand, competitive environment, etc. Business requirements can be specific requirements or goals that an enterprise or organization needs to meet during the operation process. An industry model can be understood as a model constructed based on industry-common data, industry characteristics, and business requirements, and can be used to analyze and predict various phenomena and trends within the industry.

[0056] Step S20: Refine the industry model to obtain multi-industry tags, and add the multi-industry tags to the corresponding central tag library of the server.

[0057] It should be noted that the multi-industry tags can represent the key characteristics or indicators of different industries. It can be understood that the multi-industry tags can be used to compare and analyze the industry characteristics of different industries. The central tag library can be understood as a database or system that centrally stores multi-industry tags. Exemplarily, the central tag library can provide a unified tag management and access interface, and can support cross-industry data sharing and model construction.

[0058] Step S30: When receiving a tag request from a new user, obtain the target industry tags from the central tag library, and construct a target data model based on the target industry tags.

[0059] It should be noted that the tag request is used to obtain the target industry tags. Exemplarily, the industry information of the target industry can be carried in the tag request. The target industry tags can be understood as tags selected from the central tag library that are related to specific user needs or industries, and are used to construct a target data model that meets the user's needs. The target data model not only fuses various industry data, but also can provide customized market analysis according to the specific industry needs of the user. In addition, it can also conduct personalized promotion for specific industries and target customer groups to improve the marketing effect and customer conversion rate.

[0060] In this embodiment, by obtaining the industry-common data provided by the server and constructing industry models for multiple industries in combination with industry characteristics and business requirements, cross-industry data sharing and integration are achieved, breaking the "data silos". By refining the industry models, multi-industry tags are obtained and added to the central tag library, ensuring the accuracy and applicability of the data model and better meeting the needs of different users. When receiving a tag request from a new user, the target industry tags are obtained from the central tag library and the target data model is constructed based on them, enabling the system to flexibly adapt to the needs of different users and industries and having good scalability. By using the central tag library, the process of constructing the data model is simplified and the efficiency of data access is improved.

[0061] Referring to Figure 2 , Figure 2 FIG. is a schematic flowchart of the second embodiment of the data model construction method of the present application. Based on the first embodiment shown above Figure 1 a second embodiment of the data model construction method of the present application is proposed.

[0062] In the second embodiment, step S10 includes:

[0063] Step S101, establishing a knowledge graph according to industry-common data, industry characteristics, and business requirements.

[0064] It should be noted that a knowledge graph can be a structured graph for representing knowledge, composed of multiple nodes, where the nodes represent entities (such as specific users, locations, things), and the edges represent the relationships between entities.

[0065] Step S102, using a preset convolutional network to update the knowledge graph to obtain an updated knowledge graph.

[0066] It should be noted that the preset convolutional network can be a deep learning model and can be used to process graph data. Exemplarily, operations such as convolution, pruning, and correction can be performed on the entities themselves and the relationships between entities in the knowledge graph to obtain an updated knowledge graph. Exemplarily, the preset convolutional network can be a Graph Neural Network (GNN), and its calculation formula can be expressed as formula (1).

[0067]

[0068] It should be noted that in formula (1), H m represents the feature (represented by a vector) of the m-th entity before update, represents the feature of the m-th entity after update, and W mdenotes the weight parameter corresponding to the m-th entity, and ReLU denotes the ReLU activation function. denotes the degree matrix of the first adjacency matrix, and the specific calculation formula is shown in Formula (2). denotes the second adjacency matrix, and the specific calculation formula is shown in Formula (3).

[0069]

[0070] In Formula (2), denotes the element value of the i-th row and j-th column of the first adjacency matrix. In Formula (3), A denotes the first adjacency matrix, I denotes the identity matrix, the element values on the diagonal of the identity matrix are 1, and the remaining element values are 0.

[0071] It should be noted that the example of the preset convolutional network here is only used to understand this embodiment, and does not constitute a limitation on the method for updating the knowledge graph using the preset convolutional network in this embodiment. For example, in addition to GNN, the preset convolutional network can also be a Graph Attention Network (GAT).

[0072] Step S103, based on the updated knowledge graph, construct industry models for each industry.

[0073] It should be noted that the updated knowledge graph contains richer and more accurate relationships and information, and the industry models constructed based on the updated knowledge graph ensure the accuracy of the models.

[0074] In this embodiment, by establishing a knowledge graph, industry common data, industry characteristics, and business requirements can be integrated to form a comprehensive knowledge network. Using a preset convolutional network to update the knowledge graph can dynamically optimize and enrich the information in the graph. Constructing industry models based on the updated knowledge graph can more accurately capture industry characteristics and requirements.

[0075] In one implementation manner, the step S101 includes: extracting semantic tags from industry common data, industry characteristics, and business requirements through a semantic analysis model; based on the semantic tags, establishing connections between entities through semantic similarity and cosine similarity to obtain a knowledge graph.

[0076] It should be noted that the semantic analysis model can be used to understand and extract semantic information in the text. The semantic tags can be identifiers extracted from industry common data, industry characteristics, and business requirements, and are used to represent the core content and meaning of the text. Exemplarily, the semantic tags can include keywords, themes, subjects, sentiment tendencies, and events, etc. Among them, the subject can represent the specific things mentioned in the data, such as company names, personal names, and locations, etc. The sentiment tendency can be understood as the sentiment tendency of the text, such as positive, negative, or neutral, etc. Semantic similarity can be understood as the degree of similarity in meaning between two semantic tags or entities. By calculating semantic similarity, it can be determined whether two entities are semantically related. Cosine similarity can be understood as an index for measuring the similarity of two vectors, and is usually used for the calculation of text similarity. Exemplarily, the similarity can be judged by calculating the cosine value of the included angle between two vectors, and the closer the value is to 1, the more similar it is. It can be understood that the knowledge graph can include entities, attributes of the entities, and connections between the entities. The entities can establish data connections with semantic tags, and the attributes of the entities can be represented by the word vectors of the semantic tags.

[0077] It should be noted that the step of extracting semantic tags from industry common data, industry characteristics, and business requirements through the semantic analysis model includes: converting the industry common data, industry characteristics, and business requirements into current features, global features, and key features; decoding the current features, global features, and key features through a decoder to obtain semantic tags.

[0078] It should be noted that the decoder can be used to convert features into semantic tags. Exemplarily, the semantic analysis model can dynamically obtain semantic tags through the attention mechanism.

[0079] It should be noted that the step of establishing connections between entities based on semantic tags through semantic similarity and cosine similarity to obtain the knowledge graph includes: judging whether the keywords, subjects, and sentiment tendencies corresponding to different entities are the same. If they are the same, a first connection is established between the corresponding two entities; calculating the first cosine similarity of the themes corresponding to different entities. If the first cosine similarity reaches a preset threshold, a second connection is established between the corresponding two entities; calculating the second cosine similarity of the events corresponding to different entities. If the second cosine similarity reaches a preset threshold, a third connection is established between the corresponding two entities; based on the first connection, the second connection, and the third connection, the knowledge graph is obtained.

[0080] It should be noted that the first connection can be understood as the connection established between the corresponding two entities when the keywords, subjects, and sentiment tendencies of different entities are the same. Similarly, the two entities connected by the second connection are similar in theme, and the two entities connected by the third connection are similar in event.

[0081] Exemplarily, the connection methods between entities include but are not limited to the following. When at least one pair of the keywords, subjects, or sentiment tendencies of two entities are the same, a connection relationship can be established between the corresponding entities; when the similarity value of the events or themes of two entities is greater than or equal to a preset threshold, a connection relationship can be established between the corresponding entities, where the similarity value can be calculated by cosine similarity. Exemplarily, the preset threshold can be set to 0.6, and the cosine similarity can be expressed as formula (4).

[0082]

[0083] In formula (4), x i and y i respectively represent the word vectors of the semantic tags of two entities, similarity represents the similarity value, and n represents the dimension of the semantic tags.

[0084] In this embodiment, by using a semantic analysis model to extract semantic tags from industry-common data, industry characteristics, and business requirements, the core information in the text can be captured more accurately. By establishing connections between entities through semantic similarity and cosine similarity, the multi-level relationships between entities can be effectively identified and displayed. This method not only considers the surface similarity of entities but also deeply analyzes their semantic and thematic relevance. Further, through the first connection, the second connection, and the third connection, the knowledge graph can display the multi-dimensional relationships between entities, providing strong support for complex query and analysis tasks.

[0085] In one embodiment, the step S102 includes: calculating the attention weight according to the current feature and the global feature; performing weighted summation on the key features through the attention weight to obtain an updated feature representation; and updating the knowledge graph based on the updated feature representation to obtain an updated knowledge graph.

[0086] Exemplarily, the semantic analysis model can be expressed as formula (5).

[0087]

[0088] In formula (4), Attention (attention) can be represented as the updated feature output by the semantic analysis model, which is used to capture important information in the data. Query (query) can be represented as the current feature, and Key (key) can be represented as the global feature, serving as a key vector for matching with the Query. Value (value) can be represented as the key feature, serving as a value vector. T represents the transpose operation, which is used to transpose the Key so as to perform matrix multiplication with the Query, d kcan represent the dimensionality of industry text data. The attention mechanism performs a weighted sum of Values based on the matching results of Query and Key to generate updated features. Among them, d k can take a constant value of 4, and softmax represents the softmax activation function, which is used to convert the attention scores into a probability distribution.

[0089] In this embodiment, by calculating the attention weights, the most important parts of the current features and global features can be identified and emphasized. By performing a weighted sum of the key features, an updated feature representation is obtained, which can more effectively update the knowledge graph, ensuring that the knowledge graph can timely reflect the latest data changes and feature information. By adjusting the degree of attention to different features through the attention mechanism, the updated knowledge graph can more flexibly adapt to different data and tasks.

[0090] In one embodiment, before the step of obtaining the target industry label from the central label library and constructing the target data model according to the target industry label, it further includes: combining private data, and generating new industry labels according to real-time industry features, business requirements, and industry common data; adding the new industry labels into the central label library to obtain an updated central label library.

[0091] It should be noted that private data can be data proprietary to an enterprise or organization. Exemplarily, private data can include access control lists, encrypted data, and encapsulated information, etc. Real-time industry features can be understood as market dynamics, technological progress, policy changes, etc. Business requirements can be the conditions or goals that an enterprise or organization needs to meet during operation, involving aspects such as product development, marketing, and customer service. The new industry labels are labels generated based on private data, real-time industry features, business requirements, and industry common data, reflecting the latest dynamics and specific requirements of the industry.

[0092] In this embodiment, by combining private data, real-time industry features, business requirements, and industry common data to generate new industry labels, it can ensure that the data model is closer to the actual business requirements and industry dynamics, improving the accuracy and relevance of the model. Dynamically generating and updating industry labels enables the central label library to timely reflect the latest changes in the industry, enhancing the flexibility and adaptability of data processing. The updated central label library can more effectively guide resource allocation and management, realizing cross-industry data sharing and integration, simplifying the process of constructing the data model, and improving the efficiency of data access.

[0093] To make the description of the above embodiments clearer, in one implementation, a platform construction method is proposed. The platform can be a server for managing a central tag library. Exemplarily, the platform can adopt a microservices architecture and use a distributed database to store tag data. The distributed database can support large-scale data storage and fast query. In addition, the platform further includes a tag management module, a real-time and offline platform construction module, a tag assembly and scheduling module, and a query engine module. Among them, the tag management module can manage features (i.e., tags at the atomic level) and combined tags through a system, support the creation, update, deletion, and version management of tags, and ensure the accuracy and consistency of tag data. The real-time and offline platform construction module can support real-time and offline feature processing. The real-time processing module is used to quickly respond to user requests and provide real-time data analysis and decision support; the offline processing module is used for batch data processing and in-depth analysis. The tag assembly and scheduling module can help users assemble and schedule tags. The query engine module can help users efficiently retrieve and analyze tag data through a distributed query engine.

[0094] In this implementation, through the combination of the microservices architecture and the distributed database, the platform can efficiently manage and store large-scale tag data, support fast query and data processing, and improve the overall performance of the system. Through a systematic management method, it supports the creation, update, deletion, and version management of tags, ensuring the accuracy and consistency of tag data, enabling users to flexibly define and adjust tags to adapt to changing business needs. The combination of the real-time and offline platform construction modules provides real-time data analysis and decision support, and also supports batch data processing and in-depth analysis. The dual processing enables the platform to quickly respond to user requests. The query engine module, through the distributed query engine, supports complex queries, data filtering, and aggregation analysis, providing fast query responses. The modularly constructed platform realizes efficient, flexible, and scalable central tag library management, and can meet the diverse needs of different industries and users.

[0095] It should be noted that the above examples are only for understanding this application and do not constitute a limitation on the data model construction method of this application. Based on this technical concept, more forms of simple transformations are within the protection scope of this application.

[0096] This application also provides a data model construction device. Please refer to Figure 3 , the data model construction device includes:

[0097] An initial model construction module 10, configured to obtain industry-common data provided by a server, and construct industry models for multiple industries according to the industry-common data, industry characteristics, and business requirements;

[0098] The industry label extraction module 20 is used to extract the industry model to obtain multiple industry labels and add the multiple industry labels to the central label library corresponding to the server;

[0099] The target model construction module 30 is used to obtain target industry labels from the central label library when receiving a label request from a new user, and construct a target data model according to the target industry labels.

[0100] The data model construction device provided in this application adopts the data model construction method in the above embodiment, and can solve the technical problem that it is difficult to achieve cross-organization and cross-industry data sharing and integration in current data management. Compared with the prior art, the beneficial effects of the data model construction device provided in this application are the same as those of the data model construction method provided in the above embodiment, and other technical features in the data model construction device are the same as those disclosed in the method of the above embodiment, which will not be elaborated here.

[0101] This application provides a data model construction device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein, the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the data model construction method in the first embodiment above.

[0102] Refer to the following Figure 4 , which shows a schematic structural diagram of a data model construction device suitable for implementing the embodiments of this application. The data model construction device in the embodiments of this application may include, but is not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 4 The data model construction device shown is only an example and should not impose any limitation on the functions and usage scope of the embodiments of this application.

[0103] As Figure 4As shown, the data model construction device may include a processing device 1001 (such as a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1002 or a program loaded from a storage device 1003 into a random access memory (RAM) 1004. In the RAM 1004, various programs and data required for the operation of the data model construction device are also stored. The processing device 1001, the ROM 1002, and the RAM 1004 are connected to each other through a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Generally, the following systems may be connected to the I / O interface 1006: an input device 1007 including, for example, a touch screen, a touch pad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; an output device 1008 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 1003 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 1009. The communication device 1009 can allow the data model construction device to communicate with other devices wirelessly or wiredly to exchange data. Although Figure 4 a data model construction device with various systems is shown, it should be understood that it is not required to implement or have all the shown systems. Instead, more or fewer systems may be implemented or had.

[0104] In particular, according to the embodiments disclosed in the present application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains program codes for executing the methods shown in the flowcharts. In such an embodiment, the computer program can be downloaded and installed from a network through the communication device, or installed from the storage device 1003, or installed from the ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the methods of the embodiments disclosed in the present application are executed.

[0105] The data model construction device provided by the present application adopts the data model construction method in the above embodiments, and can solve the technical problem that it is difficult to achieve cross-organization and cross-industry data sharing and integration in current data management. Compared with the prior art, the beneficial effects of the data model construction device provided by the present application are the same as those of the data model construction method provided by the above embodiments, and other technical features in the data model construction device are the same as those disclosed in the method of the previous embodiment, and will not be elaborated here.

[0106] It should be understood that each part disclosed in this application can be implemented by hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in a suitable manner in any one or more embodiments or examples.

[0107] As described above, the above are only specific embodiments of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed in this application, and all of them should be covered by the protection scope of this application. Therefore, the protection scope of this application should be subject to the protection scope of the claims.

[0108] This application provides a computer-readable storage medium with computer-readable program instructions (i.e., computer programs) stored thereon, and the computer-readable program instructions are used to execute the data model construction method in the above embodiments.

[0109] The computer-readable storage medium provided by this application can be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections with 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 of the above. In this embodiment, the computer-readable storage medium can be any tangible medium that contains or stores a program, and this program can be used by or in combination with an instruction execution system, device, or device. The program code contained on the computer-readable storage medium can be transmitted by any appropriate medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0110] The above computer-readable storage medium can be included in the data model construction device; it can also exist separately without being assembled into the data model construction device.

[0111] The above computer-readable storage medium carries one or more programs, which, when executed by a data model construction device, cause the data model construction device to: obtain industry-common data provided by a server, and construct industry models for multiple industries according to the industry-common data, industry characteristics, and business requirements; refine the industry models to obtain multi-industry tags, and add the multi-industry tags to the corresponding central tag library of the server; when receiving a tag request from a new user, obtain target industry tags from the central tag library, and construct a target data model according to the target industry tags.

[0112] Computer program code for performing the operations of the present application may be written in one or more programming languages or combinations thereof. The programming languages include object-oriented programming languages such as Java, Smalltalk, C++, and also include conventional procedural programming languages such as the "C" language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, execute as a stand-alone software package, partly on the user's computer and partly on a remote computer, or execute entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (for example, by using an Internet service provider to connect through the Internet).

[0113] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present application. In this regard, each block in the flowchart or block diagram may represent a module, a program segment, or a part of code that contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than marked in the accompanying drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, and combinations of blocks in the block diagram and / or flowchart, may be implemented by a dedicated hardware-based system for performing the specified functions or operations, or may be implemented by a combination of dedicated hardware and computer instructions.

[0114] The modules involved in the embodiments of the present application can be implemented in software or in hardware. In some cases, the name of the module does not constitute a limitation on the unit itself.

[0115] The readable storage medium provided by the present application is a computer-readable storage medium. The computer-readable storage medium stores computer-readable program instructions (i.e., computer programs) for executing the above data model construction method, and can solve the technical problem that it is difficult to achieve cross-organization and cross-industry data sharing and integration in current data management. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided by the present application are the same as those of the data model construction method provided by the above embodiments, and will not be elaborated here.

[0116] The present application also provides a computer program product, including a computer program, and when the computer program is executed by a processor, the steps of the data model construction method as described above are implemented.

[0117] The computer program product provided by the present application can solve the technical problem that it is difficult to achieve cross-organization and cross-industry data sharing and integration in current data management. Compared with the prior art, the beneficial effects of the computer program product provided by the present application are the same as those of the data model construction method provided by the above embodiments, and will not be elaborated here.

[0118] The above are only some embodiments of the present application, and do not limit the patent scope of the present application accordingly. Any equivalent structural transformation made by using the content of the specification and drawings of the present application under the technical concept of the present application, or any direct / indirect application in other related technical fields, is included in the patent protection scope of the present application.

Claims

1. A method for constructing a data model, characterized in that The method described above includes: Obtaining industry-common data provided by a server, and constructing industry models for multiple industries based on the industry-common data, industry characteristics, and business requirements; Refining the industry models to obtain multi-industry labels, and adding the multi-industry labels to the corresponding central label library of the server; When receiving a label request from a new user, obtaining target industry labels from the central label library, and constructing a target data model based on the target industry labels.

2. The method according to claim 1, wherein The step of constructing industry models for multiple industries based on the industry-common data, industry characteristics, and business requirements includes: Establishing a knowledge graph based on the industry-common data, industry characteristics, and business requirements; Updating the knowledge graph using a preset convolutional network to obtain an updated knowledge graph; Based on the updated knowledge graph, constructing industry models for each industry.

3. The method according to claim 2, wherein The step of establishing a knowledge graph based on the industry-common data, industry characteristics, and business requirements includes: Extracting semantic labels from the industry-common data, industry characteristics, and business requirements through a semantic analysis model; Based on the semantic labels, establishing connections between entities through semantic similarity and cosine similarity to obtain a knowledge graph.

4. The method according to claim 3, wherein The step of extracting semantic labels from the industry-common data, industry characteristics, and business requirements through a semantic analysis model includes: Converting the industry-common data, industry characteristics, and business requirements into current features, global features, and key features; Decoding the current features, global features, and key features through a decoder to obtain semantic labels.

5. The method according to claim 3, wherein The semantic labels include keywords, themes, subjects, sentiment tendencies, and events. The step of establishing connections between entities through semantic similarity and cosine similarity based on the semantic labels to obtain a knowledge graph includes: Judging whether the keywords, the subjects, and the sentiment tendencies corresponding to different entities are the same. If they are the same, establishing a first connection between the corresponding two entities; Calculating the first cosine similarity of the themes corresponding to different entities. If the first cosine similarity reaches a preset threshold, establishing a second connection between the corresponding two entities; Calculating the second cosine similarity of the events corresponding to different entities. If the second cosine similarity reaches a preset threshold, establishing a third connection between the corresponding two entities; Based on the first connection, the second connection, and the third connection, obtaining a knowledge graph.

6. The method according to claim 4, wherein The step of updating the knowledge graph using a preset convolutional network to obtain an updated knowledge graph includes: Calculating attention weights based on the current features and the global features; Performing weighted summation on the key features through the attention weights to obtain an updated feature representation; Based on the updated feature representation, updating the knowledge graph to obtain an updated knowledge graph.

7. The method according to any one of claims 1 to 6, characterized in that, Before the step of obtaining target industry labels from the central label library and constructing a target data model based on the target industry labels, it further includes: Combining private data, and generating new industry labels according to real-time industry characteristics, business requirements, and the industry-common data. Add the new industry tags into the central tag library to obtain an updated central tag library.

8. A data model construction device, characterized in that The device includes: An initial model construction module, configured to obtain industry common data provided by a server, and construct industry models for multiple industries according to the industry common data, industry characteristics, and business requirements; An industry tag refinement module, configured to refine the industry models to obtain multi-industry tags, and add the multi-industry tags into the corresponding central tag library of the server; A target model construction module, configured to, when receiving a tag request from a new user, obtain target industry tags from the central tag library, and construct a target data model according to the target industry tags.

9. A data model construction device, characterized in that, The device includes: a memory, a processor, and a computer program stored on the memory and executable on the processor, where the computer program is configured to implement the steps of the data model construction method according to any one of claims 1 to 7.

10. A storage medium, characterized in that, The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the data model construction method according to any one of claims 1 to 7 are implemented.

11. A computer program product, characterized in that, The computer program product includes a computer program. When the computer program is executed by a processor, the steps of the data model construction method according to any one of claims 1 to 7 are implemented.

Images