A data visualization method, apparatus and product

Abstract

The application discloses a data visualization method, device and product, relates to the technical field of image display, and aims to realize the capability of autonomously completing dynamic display of a data topology graph based on different display intentions. The method comprises the following steps: determining a display task based on an input target display intention; acquiring data information corresponding to the display task from a first database and determining a data analysis result; acquiring a matched first atomic component from a second database based on the data analysis result; the first database stores application information of product use of all users; the second database stores at least one atomic component; after determining that a second atomic component needs to be generated, constructing a first prompt word based on the data analysis result; generating the second atomic component based on the first prompt word; and determining and outputting a corresponding topology graph based on the first atomic component, the second atomic component and the data processing analysis result.

Description

Technical Field

[0001] This application relates to the field of image modeling technology, and in particular to a data visualization method, apparatus, and product. Background Technology

[0002] With the emergence of natural language models such as Large Language Model (LLM), large models have been widely applied in various fields. For example, traditional operators often use large language models trained on general corpora to handle simple daily reasoning and question answering.

[0003] However, due to the increasing demands of operators, the current large-scale models are often unable to complete tasks related to specific professional fields. For example, when users need to view a specific topology display based on real-time network device distribution and data, the network operations and maintenance team needs to develop and customize the page offline based on the user's display requirements and the monitoring and analysis data of the network devices, and then complete the page development and rendering to provide the user with a specific display. The entire solution has a long cycle and poor intelligence, and cannot meet the user's display needs more flexibly and conveniently according to the actual situation, resulting in a low user experience. Summary of the Invention

[0004] This application provides a data visualization method and apparatus to enable the autonomous dynamic display of data topology diagrams based on different display intentions.

[0005] Firstly, embodiments of this application provide a data visualization method. The method includes:

[0006] The system receives an input target display intent and determines a display task based on the target display intent; it retrieves data information corresponding to the display task from a first database and determines a data analysis result for the display task based on the data information; it retrieves a matching first atomic component from a second database based on the data analysis result; the first database includes one or more of the following: at least one application information corresponding to the product, at least one attribute information corresponding to the product, attribute information corresponding to the user, and behavioral information corresponding to the user; the second database includes information corresponding to at least one atomic component; after determining that a second atomic component needs to be generated, it constructs a first prompt word for generating the second atomic component based on the data analysis result; it generates the second atomic component based on the first prompt word; and it determines and outputs a topology map corresponding to the target display intent based on the first atomic component, the second atomic component, and the data processing and analysis result.

[0007] For example, embodiments of this application provide a scheme for automatic Topo expansion and atomic component addition based on image generation technology. For instance, based on relevant data analysis results (such as network element fault conditions and analysis results), a highly readable Topo display is generated, along with a data summary explanation in natural language.

[0008] Through the above method, the embodiments of this application can, based on the user's input of the target display intent, combine the relevant data information corresponding to the target display intent, such as data related to equipment operation, maintenance and operation as input, and link to perform image rendering and generate new atomic diagrams or diagrams related to equipment operation, maintenance and operation data, such as topology diagrams, etc., which are more adaptable, more flexible, and effectively meet the user's display needs, allowing the user to obtain a more intuitive and faster visualization interface of relevant data.

[0009] As an example, embodiments of this application can integrate at least one large model with image generation capabilities, thereby enabling the generation of a second atomic component based on the first prompt word by invoking the large image generation model. For instance, embodiments of this application can integrate two large image models, namely image model A and image model B. Embodiments of this application can select a suitable image model from image model A and image model B based on the first prompt word to generate the second atomic component. As another example, embodiments of this application can integrate two large image models, namely image model A and image model B. Embodiments of this application can generate second atomic component A based on image model A and second atomic component B based on image model B, respectively, based on the first prompt word. Then, by fusing second atomic component A and second atomic component B, the final second atomic component is obtained.

[0010] As an example, the data visualization system of this application embodiment has an atomic component generation function. This application embodiment can analyze and process based on the first prompt word to obtain the prototype of the relevant second atomic component, and then continuously improve and optimize the prototype of the second atomic component based on the first prompt word and other information to obtain the final second atomic component.

[0011] In one possible implementation, receiving the input of a first target display intent and determining the display task based on the first target display intent includes:

[0012] A second prompt word is constructed based on the target display intent; the second prompt word includes a role for handling the display task, context information for determining the display task, the output format of the display task, and part or all of the requirement question corresponding to the target display intent; the requirement question includes part or all of the target area range to be displayed, target level information, target device information, and target display style; the context information includes part or all of the queryable document information, callable API information, and queryable database information; based on the second prompt word, the display task is determined, and the display task includes at least one subtask; the subtask includes part or all of the knowledge retrieval task, API call task, and database query task.

[0013] Through the above method, this application embodiment provides a way to determine the display task, thereby enabling the visualization system to be more targeted, with a layered task decomposition approach, and to generate a data visualization display interface that better meets user needs.

[0014] As an example, the role used to process the display task in the embodiments of this application may include, but is not limited to, modules, units, other integrated or callable program products, tools, etc., with corresponding processing functions.

[0015] As an example, embodiments of this application can integrate large models with natural language analysis and processing capabilities, such as LLM models, so that the display task corresponding to the target display intent can be determined based on the second prompt word by calling the LLM model.

[0016] As an example, the data visualization system implemented in this application has functions such as natural language analysis and processing. This application embodiment can perform analysis and processing based on the target display intent, such as performing intent decomposition, keyword extraction, and other operations on the target display intent, constructing a first prompt word, and further performing analysis and processing based on the first prompt word to obtain the corresponding display task.

[0017] As an example, the information of the target device shown in the embodiments of this application includes, but is not limited to, attribute information such as the type of the device, and part or all of the application information such as the operation and maintenance and faults of the device, which are not limited herein.

[0018] In one possible implementation, the data analysis results include some or all of the following:

[0019] Information on target devices within the target area and their distribution within the target area; information on target devices at each level in the target hierarchy information and their distribution at each level.

[0020] As an example, the information of the target device within the target area described in the embodiments of this application can be displayed through an information list or other means. For example, the information list may include device fault information, device operation and maintenance information, etc., which are not limited here.

[0021] In one possible implementation, the method further includes:

[0022] When the atomic components in the second database cannot fully meet the matching requirements of the data analysis results, it is determined that the second atomic component needs to be generated; or,

[0023] When the target display intent is found to include keywords indicating the addition of a new atomic component, it is determined that the second atomic component needs to be generated.

[0024] For example, in this application embodiment, the data analysis results can be used to perform matching from the second database. If no matching atomic component is found, it is determined that the second atomic component needs to be generated. As another example, when this application embodiment obtains information that triggers the generation of the second atomic component, such as based on relevant keywords identified in the target display intent (the keywords include but are not limited to "add XX", "extend XX", etc.), it is determined that the second atomic component needs to be generated.

[0025] Through the above methods, the embodiments of this application provide a variety of ways to determine the generation of the second atomic component, which is more flexible and adaptable.

[0026] In one possible implementation, when the second atomic component to be generated includes at least one atomic component icon, a first prompt word for generating the at least one atomic component icon is constructed based on the data analysis results. The first prompt word includes the role used to generate the at least one atomic component icon, descriptive information of the at least one atomic component icon, and part or all of the requirement question corresponding to the target display intent. In another possible implementation, when the second atomic component to be generated includes at least one atomic component display granularity, a first prompt word for generating the at least one atomic component display granularity is constructed based on the data analysis results. The display granularity includes part or all of dimension-based display granularity and scaling-based display granularity. The first prompt word includes the role used to generate the at least one atomic component display granularity, descriptive information of the at least one atomic component display granularity, the original map image of the display area corresponding to the display granularity, and part or all of the requirement question corresponding to the target display intent.

[0027] Through the above methods, this application provides a technology for generating large models based on multimodal images. For example, it realizes multimodal map image rendering, network element device icon and KPI atomic primitive generation, etc., effectively solving the problem that existing Power BI and other natural language-based image arrangement only support chart and report types, and do not support dynamic display of map totems, icons and other types.

[0028] In one possible implementation, based on the first atomic component and the second atomic component, the data processing and analysis results are used for topology orchestration to obtain the display encoding of the topology graph; the display encoding is then rendered to generate a visualized topology graph.

[0029] Through the above method, the embodiments of this application provide a rendering technology for generating Topo atomic components based on natural language understanding, which has stronger applicability.

[0030] In one possible implementation, the method further includes:

[0031] The second atomic component is stored in the second database.

[0032] Understandably, embodiments of this application can update and maintain the atomic component library based on the newly generated second atomic component. Furthermore, embodiments of this application can also update, maintain, and further train and improve the entire data visualization system based on the user's target display intent, relevant data analysis results, and some or all of the second atomic components, thereby better constructing the intelligent, efficient, automatically expandable, accumulative, and maintainable data intelligent visualization system related to equipment operation, maintenance, and management provided by embodiments of this application.

[0033] In one possible implementation, the target presentation intent is described based on natural language.

[0034] Through the above method, this application provides a way to automatically generate Topic displays based on natural language understanding. This method can intelligently and autonomously understand customer intent, thereby customizing the display of network element Topics based on customer intent, achieving zero waiting time and zero manual intervention, and effectively reducing display costs.

[0035] In one possible implementation, based on the first prompt word, the second atomic component is generated by modifying the third atomic component in the second database, wherein the similarity between the third atomic component and the second atomic component is not lower than a first threshold; or,

[0036] Based on the first prompt word, a preliminary architecture of the second atomic component is obtained, and the second atomic component is generated by gradually optimizing the preliminary architecture.

[0037] Through the above methods, this application provides multiple ways to generate the second atomic component. For example, it can be generated based on existing related atomic components, which can obtain the second atomic component more quickly and conveniently. Another example is that it can be generated entirely based on related description information, creating a brand-new atomic component from scratch, which is more operable and flexible.

[0038] Secondly, a data visualization device is provided. The device includes:

[0039] The input / output module is used to receive input and display the intended message.

[0040] A processing module is configured to: determine a display task based on the target display intent; obtain data information corresponding to the display task from a first database and determine the data analysis result of the display task based on the data information; obtain a matching first atomic component from a second database based on the data analysis result; the first database includes one or more of the following: at least one application information corresponding to the product, at least one attribute information corresponding to the product, attribute information corresponding to the user, and behavioral information corresponding to the user; the second database includes information corresponding to at least one atomic component; after determining that a second atomic component needs to be generated, construct a first prompt word for generating the second atomic component based on the data analysis result; generate the second atomic component based on the first prompt word; and determine the topology map corresponding to the target display intent based on the first atomic component, the second atomic component, and the data processing and analysis result.

[0041] The input / output module is used to output the topology map.

[0042] In one possible implementation, the processing module is specifically used for:

[0043] A second prompt word is constructed based on the target display intent; the second prompt word includes a role for handling the display task, context information for determining the display task, the output format of the display task, and part or all of the requirement question corresponding to the target display intent; the requirement question includes part or all of the target area range to be displayed, target level information, target device information, and target display style; the context information includes part or all of the queryable document information, callable API information, and queryable database information; based on the second prompt word, the display task is determined, and the display task includes at least one subtask; the subtask includes part or all of the knowledge retrieval task, API call task, and database query task.

[0044] In one possible implementation, the data analysis result includes the union of some or all of the following:

[0045] Information on the target devices within the target area and the distribution of the target devices within the target area;

[0046] The target hierarchy information includes the target device information corresponding to each level, and the distribution of the target devices corresponding to each level.

[0047] In one possible implementation, the processing module is further configured to:

[0048] When the atomic components in the second database cannot fully meet the matching requirements of the data analysis results, it is determined that the second atomic component needs to be generated; or,

[0049] When the target display intent is found to include keywords indicating the addition of a new atomic component, it is determined that the second atomic component needs to be generated.

[0050] In one possible implementation, the processing module is specifically used for:

[0051] When the second atomic component to be generated includes at least one atomic component icon, a first prompt word for generating the at least one atomic component icon is constructed based on the data analysis results; the first prompt word includes the role for generating the at least one atomic component icon, the descriptive information of the at least one atomic component icon, and part or all of the requirement question corresponding to the target display intent.

[0052] In one possible implementation, the processing module is specifically used for:

[0053] When the second atomic component to be generated includes at least one atomic component display granularity, a first prompt word for generating the at least one atomic component display granularity is constructed based on the data analysis results; the display granularity includes part or all of dimension-based display granularity and scaling-based display granularity, and the first prompt word includes the role for generating the at least one atomic component display granularity, the descriptive information of the at least one atomic component display granularity, the original map image of the display area corresponding to the display granularity, and part or all of the requirement question corresponding to the target display intent.

[0054] In one possible implementation, the processing module is specifically used for:

[0055] Based on the first atomic component and the second atomic component, the data processing and analysis results are used for topological arrangement to obtain the display encoding of the topological graph;

[0056] The display code is rendered to generate a visualized topology map.

[0057] In one possible implementation, the processing module is specifically used for:

[0058] The second atomic component is stored in the second database.

[0059] In one possible implementation, the target presentation intent is described based on natural language.

[0060] In one possible implementation, the processing module is specifically used for:

[0061] Based on the first prompt word, the second atomic component is generated by modifying the third atomic component in the second database, wherein the similarity between the third atomic component and the second atomic component is not lower than a first threshold; or,

[0062] Based on the first prompt word, a preliminary architecture of the second atomic component is obtained, and the second atomic component is generated by gradually optimizing the preliminary architecture.

[0063] Thirdly, embodiments of this application provide a data visualization apparatus, the apparatus including a processor and a memory. The memory is used to store program code. The processor is used to read and execute the program code stored in the memory to implement the method as described in the first aspect or any design of the first aspect.

[0064] Fourthly, this application also provides a computer storage medium. The storage medium stores a software program, which, when read and executed by one or more processors, can implement any of the methods provided in the first aspect.

[0065] Fifthly, embodiments of this application provide a computer program product containing instructions that, when run on a computer, cause the computer to perform the method provided in any of the designs of the first aspect described above.

[0066] Sixthly, embodiments of this application provide a chip including a processor. The processor is configured to perform the methods provided in any of the designs of the first aspect.

[0067] In one possible design, the chip also includes a communication interface that is coupled to the processor.

[0068] In one possible design, the chip is connected to a memory for reading and executing software programs stored in the memory to implement the method provided by any of the designs in the first aspect.

[0069] Based on the implementations provided in the foregoing aspects, this application can be further combined to provide more implementations. Furthermore, the technical effects achievable by any of the second to sixth aspects of this application can be referred to the description of the technical effects achievable by the first aspect and its various implementations. Attached Figure Description

[0070] Figure 1 This application provides a schematic diagram of a first data visualization system architecture;

[0071] Figure 2 A schematic diagram of a UI interaction interface provided in an embodiment of this application;

[0072] Figure 3 This is a schematic diagram of another UI interaction interface provided in an embodiment of this application;

[0073] Figure 4 This is a schematic diagram of a second data visualization system architecture provided in an embodiment of this application;

[0074] Figure 5 This is a schematic diagram of a third data visualization system architecture provided in the embodiments of this application;

[0075] Figure 6 A schematic diagram of the scenario architecture for application based on a data visualization system provided in the embodiments of this application;

[0076] Figure 7 This is a schematic diagram of a data visualization method provided in an embodiment of this application;

[0077] Figure 8 This application provides a schematic diagram of a data visualization scenario process.

[0078] Figure 9 This is a schematic diagram of another data visualization scenario provided in an embodiment of this application;

[0079] Figure 10 A schematic diagram of a data visualization device provided in an embodiment of this application;

[0080] Figure 11 This is a schematic diagram of another data visualization device provided in an embodiment of this application. Detailed Implementation

[0081] The embodiments of this application can be applied to various data model mining scenarios. For example, this application can be used to build intelligent, local adaptive potential customer mining and marketing scenarios.

[0082] To facilitate understanding of the embodiments of this application, the concepts of the technical terms involved are first introduced below:

[0083] (1) Prompt information: In large language models, prompt information is a piece of text used to guide the large language model to generate a specific type of output. By designing and optimizing a set of prompt information, the large language model can perform a specific task.

[0084] (2) Data visualization (DV) refers to the technique of conveying data or information by encoding data or information into visual objects contained in charts.

[0085] (3) A data visualization dashboard is a data display method based on a computer screen. It presents massive amounts of data through graphics, charts, images, etc., making the data more intuitive and easier to understand. It integrates technologies such as data mining, data analysis, and data visualization, and can present complex data in a simple and easy-to-understand way, improving the efficiency of data analysis and decision-making.

[0086] (4) Operator / Customer (User): In this application, operator, customer and user have the same meaning, which are users / consumers of the relevant goods (also known as products).

[0087] (5) Large Language Models (LLMs) are artificial intelligence models designed to understand and generate human language. They are trained on large amounts of text data to learn the ability to understand and generate human language and can perform a wide range of tasks, including text summarization, translation, sentiment analysis, and more.

[0088] The core idea of ​​LLM is to learn patterns and language structures in natural language through large-scale unsupervised training, which can simulate human language cognition and generation processes to some extent. Compared with traditional NLP models, LLM can better understand and generate natural text, while also demonstrating a certain degree of logical thinking and reasoning ability.

[0089] The smallest processing unit in a large model can be called a Token.

[0090] (6) Topology Graph (Topo): A graph that displays the operation, maintenance and related data or information of equipment on a map in the form of points and surfaces.

[0091] In the description of this application, unless otherwise stated, "multiple" refers to two or more. Additionally, " / " indicates that the related objects are in an "or" relationship; for example, A / B can represent A or B. "And / or" in this application merely describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, and B alone, where A and B can be singular or plural. Furthermore, to facilitate a clear description of the technical solutions of the embodiments of this application, the terms "first" and "second" are used in the embodiments to distinguish identical or similar items with essentially the same function and effect. Those skilled in the art will understand that the terms "first" and "second" do not limit the quantity or execution order, and that "first" and "second" are not necessarily different. It should also be noted that, unless specifically stated, the specific description of some technical features in one embodiment can also be used to explain the corresponding technical features mentioned in other embodiments.

[0092] The following description, in conjunction with the accompanying drawings, illustrates exemplary embodiments of this application, including various details to aid understanding. These embodiments should be considered merely exemplary. Therefore, those skilled in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope of this application. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description. It should be noted that certain software, components, models, and other existing industry solutions may be mentioned in the embodiments of this application. These should be considered exemplary, intended only to illustrate the feasibility of implementing the technical solutions of this application, and do not imply that the applicant has already used or necessarily used the aforementioned solutions.

[0093] The acquisition, transmission, storage, and use of data in this application all comply with the requirements of relevant national laws and regulations.

[0094] Existing natural language-based orchestration technologies can arrange and combine existing components based on natural language interactions. For example, users can use the "Ask Data" function to type questions in a common language, and the answers will be presented in a data visualization format based on a fixed template. Alternatively, users can generate charts directly based on preset templates through simple natural language queries (NLP), or even start with smaller components (such as a single KPI or a simple table) and gradually combine them into more complex analytical views based on preset templates.

[0095] In other words, current solutions for orchestrating and displaying large models all rely on preset templates. However, in practical applications, users often have diverse needs. For example, when the displayed topics involve new display dimensions, new device types, new product KPI display requirements, or new layer and map scaling granularity, the existing model system lacks pre-built atomic components for these types, rendering the existing model system's preset templates inadequate for current requirements. Furthermore, current LLM orchestration technology cannot quickly develop and expand atomic components related to display dimensions, device types, KPI metrics, or layer and map scaling granularity of the topic chart, thus failing to achieve the expansion of topic display for different scenarios.

[0096] As can be seen from the above introduction, the current solutions for analysis, drawing, and image display based on large models have poor intelligence and cannot meet users' needs more flexibly and conveniently according to the actual situation, resulting in a low user experience. For example, the current model display system cannot combine relevant data information, such as data related to equipment operation, maintenance, and management, as input to perform dynamic topology display.

[0097] To address the aforementioned issues, this application provides a data visualization method that, based on user display needs and combined with the analysis and processing results of relevant data information, dynamically generates, expands, and renders topologies, providing users with a more intelligent, efficient, automatically expandable, accumulative, and maintainable intelligent image model system.

[0098] The technical solutions in the embodiments of this application will now be described with reference to the accompanying drawings.

[0099] See Figure 1 The diagram shown illustrates the system architecture of a data visualization method provided in this application embodiment. The system architecture may include a UI operation layer 100, a data analysis layer 200, a data storage layer 300, and an AI model layer 400. Through inter-layer interaction, an intelligent data visualization display system is constructed.

[0100] UI operation layer 100 is used to receive user intent.

[0101] As an example, the UI operation layer 100 in this embodiment may include a UI interaction interface, which may include a dialog box for receiving user input of the target display intent.

[0102] As an example, the target display intent input in the embodiments of this application can be based on a natural language description.

[0103] As an example, the target display intent described in this application embodiment may include a requirement issue.

[0104] For example, the requirement problem described in this application embodiment can be understood as the user's need for visualization display, such as the target area range, target hierarchy information, target device information, and part or all of the target display style. The target device information includes, but is not limited to, attribute information such as the device model, and application information such as device operation and maintenance information and fault information. The relevant indicator items of the device to be displayed can be determined according to the target display intent. For example, the relevant indicator items of the device to be displayed can be the operation and maintenance information of a device of type A, that is, the indicator items include type A and operation and maintenance information.

[0105] For example, such as Figure 2 As shown, the interactive interface of the UI operation layer can include multiple options. Users can directly input specific display requirements in the dialog box, or set corresponding parameter requirements (e.g., icon size, interface color, etc.) through various parameter input boxes, or select corresponding display components (e.g., display icons) through component selection boxes. It should be noted that... Figure 2This is merely an interface example of an embodiment of this application and does not constitute a limitation on the embodiment of this application.

[0106] For example, such as Figure 3 As shown, the interactive interface of the UI operation layer can also display related topology diagrams, for example, it can display the topology. Figure 1 , wherein the topology Figure 1 The data visualization system can display the intended meaning based on the user's previously input target. Figure 1 The topology generated through analysis and processing can also be a topology selected by the user from a topology library; there are no restrictions on this, allowing the user to view the currently displayed topology. Figure 1 Perform display intent operations, such as on the currently displayed topology. Figure 1 The system makes modifications and annotations, and then uses the identified modifications and annotations to determine the user's current target display intent.

[0107] As an example, the user's current target display intent described in this application embodiment may be entered once, or it may be obtained by summarizing multiple inputs based on the prompts of the UI interaction interface, or it may be obtained by modifying and updating the previously entered target display intent, etc., and there is no limitation here.

[0108] As an example, after the topology map displayed in the UI interaction interface in this embodiment of the application meets the user's display needs, for example, the user confirms the topology map displayed in the UI interaction interface. At this time, the data visualization system can store the topology map confirmed by the user, associate the topology map with the user ID, and analyze the user's relevant operations for obtaining the topology map, or, based on the content of the topology map, analyze the user's relevant display needs and other experience data content, so as to provide the user with more accurate data visualization display in the future.

[0109] The data processing layer 200 is used to call the data storage layer based on the target display intent obtained from the UI operation layer 100 to obtain the corresponding query results; based on the obtained query results, it interacts with the AI ​​model layer to design and generate a topology map; the generated topology map design is rendered and displayed to the user through the UI operation layer.

[0110] As an example, the data processing layer 200 in this application embodiment may also be referred to as a network management system (Copilot).

[0111] As an example, such as Figure 4As shown, the data processing layer 200 in this embodiment may include an intent perception layer 210, a data query and analysis layer 220, a topology map display and design layer 230, an extended and newly added topology generation layer 240, and a topology rendering layer 250.

[0112] The intent perception layer 210 may include an intent perception and translation engine 211, which is used to perceive and translate the target display intent (e.g., the display intent of a topology map) input by the UI operation layer, and extract relevant information of the target display intent.

[0113] For example, the intent perception and translation engine 211 described in this application embodiment can translate the user intent input from the UI operation layer to generate executable tasks that can obtain data indicator results related to device operation, maintenance and operation. For example, it can obtain a task list (or a task set, which is not limited here) corresponding to the user intent. The task list includes, but is not limited to, data query tasks, API call tasks and knowledge retrieval tasks.

[0114] The data query and analysis layer 220 may include a data query and analysis engine 221, which is used to interact with the data storage layer based on the task list obtained by the intent awareness layer to obtain query results corresponding to the task list, such as data indicators related to device operation, maintenance and operation.

[0115] The topology display design layer 230 is used to design the topology display based on the query results obtained by the data query and analysis layer (e.g., data indicator results corresponding to various tasks).

[0116] For example, when the topology map display design layer 230 in this application embodiment designs the topology map display based on the query results, the design process may include, but is not limited to, Topo atomic component acquisition, atomic component arrangement, and generation of extended and added Topo component images.

[0117] As an example, in this embodiment of the application, when the topology display design layer 230 determines, based on the query results obtained by the data query analysis layer, that there is no pre-generated atomic component that satisfies the target display intent (for example, when it is determined that there is an extended and newly added topology component), it can send a command to the updated topology generation layer 240 to update the topology component so that the updated topology generation layer 240 can obtain the updated topology component.

[0118] As an example, in this embodiment of the application, after receiving the updated topology component sent by the updated topology generation layer 240, the topology display design layer can store the updated topology component in the data storage layer (e.g., the topology component library in the data storage layer) for future use.

[0119] It is understood that the atomic components in the embodiments of this application may include topology components.

[0120] As an example, the topology display design layer described in this application embodiment arranges existing atomic components and newly generated atomic components to generate the encoding for the Topo display, and then sends the encoding to the topology rendering layer 250.

[0121] As an optional example, when the data storage layer stores orchestration schemes, the topology display design layer 220 in this embodiment can also obtain an orchestration scheme corresponding to or related to the query result from the data storage layer based on the query result. In this embodiment, the orchestration scheme related to the query result can be understood as an orchestration scheme whose similarity to the target orchestration scheme to be generated based on the query result is higher than a first threshold. The first threshold can be carried in the target display intent or can be preset by the system.

[0122] For example, embodiments of this application can set corresponding first thresholds based on different topology display requirements and scenarios. When the target orchestration scheme is not present in the data storage layer, an orchestration scheme highly related to the target orchestration scheme can be obtained based on the first threshold. The topology map can then be updated and modified based on this highly related orchestration scheme, leading to a faster acquisition of the target orchestration scheme. For instance, the topology map can be expanded and expanded based on the highly related orchestration scheme to obtain a target orchestration scheme that meets user needs.

[0123] The updated topology generation layer 240 is used to receive the command to update the topology component sent by the Prompt display design layer 230, interact with the constructed Prompt and AI model layers to obtain the updated topology component, and return the updated topology component to the topology display design layer 230.

[0124] As an example, the commands for updating topology components in this embodiment include, but are not limited to, commands for adding or expanding topology components, commands for reducing or increasing the granularity of topology component display, etc. It is understood that if existing atomic components or orchestration templates do not meet user needs, target atomic components or orchestration templates that satisfy the target display intent can be obtained through the updated topology generation layer.

[0125] The topology rendering layer 250 is used to render the Topo image according to the encoding of the Topo display sent by the topology map display design layer 230, generate a visual image, and display it to the user through the UI operation layer.

[0126] The data storage layer 300 is used to store data information.

[0127] As an example, such as Figure 5 As shown, the data storage layer in this embodiment may include multiple databases and knowledge bases, such as device information databases, orchestration template databases, atomic component databases, user information databases (e.g., user preferences), etc., which are not limited here.

[0128] The AI ​​model layer 400 is used to interact with the updated topology generation layer 240 to design and generate the updated Topo component.

[0129] As an example, the AI ​​model layer described in this application embodiment includes, but is not limited to, large image models (ImageGenerative Model) and large language models (such as LLM models); or, the AI ​​model layer described in this application embodiment includes, but is not limited to, modules, units, and neural networks that have image generation and processing and natural language analysis and processing functions respectively, or modules, units, and neural networks that have both image generation and natural language analysis and processing functions.

[0130] In practical applications, the embodiments of this application can utilize the above-mentioned... Figures 1-5 The system architecture described above is used for site operation and maintenance, that is, the above-mentioned Figures 1-5 The operational components of the overall network equipment involved in the design are packaged together for application. Alternatively, embodiments of this application may also, according to the operator's needs, include the aforementioned... Figures 1-5 Some components are packaged and used as an independent application for the operating engine adapted to network devices.

[0131] See Figure 6 The diagram shown is a schematic of a data visualization system scenario application architecture provided in an embodiment of this application. The scenario architecture includes sub-scenarios (also referred to as subsystems) such as intent recognition, task execution, topology generation, and topology rendering. It is used to build an intelligent, efficient, automatically expandable, accumulative, and maintainable intelligent visualization system architecture based on equipment operation, maintenance, and operation-related data.

[0132] In the intent recognition sub-scenario, which corresponds to the UI operation layer and intent perception layer 210, the main function is to receive user intent, translate the user intent, and obtain executable tasks of data indicators related to device operation, maintenance, and operation.

[0133] The task execution sub-scenario, which corresponds to the data query and analysis layer 220 mentioned above, can be used to complete executable tasks in the front-end subsystem by interacting with the data storage layer, and obtain data indicators related to equipment operation, maintenance and operation.

[0134] The Topo generation sub-scene can correspond to the above topology diagram display design layer 230 and the extended new topology generation layer 240. It can be mainly used for the design of Topo display. The design process can include the acquisition of Topo atomic components, the arrangement of atomic components and the generation of extended new Topo component images, as well as the interaction between the constructed Prompt and AI model layers to obtain the newly generated Topo display components, generate the encoding of the final Topo display, and send it to the subsequent rendering module.

[0135] The Topo rendering sub-scene corresponds to the aforementioned topology rendering layer 250. It is mainly used to render Topo images based on the final encoding of the Topo display, and generate visual images to be displayed to users.

[0136] The system described in this application embodiment can run on a cloud computing device system (which may include at least one cloud computing device, such as a server), or on a physical server, or on an edge computing device system (which may include at least one edge computing device, such as a server, desktop computer, etc.), or on various terminal computing devices, such as laptops, personal desktop computers, mobile phones, tablets, ultra-mobile personal computers (UMPCs), netbooks, personal digital assistants (PDAs), etc.

[0137] Exemplary, the system described in this application embodiment can also logically be composed of various parts, and each component of the system can run on different systems or servers. For example, each part of the system can run on any two of a cloud computing device system, an edge computing device system, and a terminal computing device. The cloud computing device system, the edge computing device system, and the terminal computing device are connected by a communication path and can communicate and transmit data with each other. Based on this, the data visualization method provided in this application embodiment combines LLM with a large image generation model to achieve linked expansion of visualization (such as a Topo graph). It can autonomously complete the generation, expansion, and rendering of specific topology graphs on the operator's client, providing users with a more intelligent, efficient, automatically expandable, accumulative, and maintainable intelligent image model system.

[0138] See Figure 7The diagram shown is a schematic representation of a data visualization method provided in an embodiment of this application, which can be applied to... Figures 1 to 6 In the system or architecture shown, the executing entity can be... Figures 1-6 The data visualization system shown includes the following specific steps:

[0139] S701: Receives input indicating the intended display.

[0140] As an example, the target display intent described in this application embodiment is based on natural language. Therefore, it allows parties with display needs to easily and conveniently input their display intent without needing to understand code compilation, making it more widely applicable and suitable for more scenarios.

[0141] As an example, the target display intent described in this application embodiment may include part or all of the target area range to be displayed, target level information, target device information, and target display style, or may be understood as the content included in the requirement problem corresponding to the target display intent.

[0142] For example, assuming that the target display intent input in this application embodiment is "Please analyze and display the fault status of device model A in the provincial level of China" through natural language description, then based on the description, it can be determined that the area to be displayed is the Chinese region, the hierarchical information to be displayed is the provincial level information, the device type to be displayed in the region is device model A, and the device application information to be displayed is the fault information of the relevant device, etc.

[0143] S702: Determine the display task based on the target display intent.

[0144] As an example, embodiments of this application may construct a second prompt word based on the target display intent; then, based on the second prompt word, determine the display task, which includes at least one subtask.

[0145] For example, the subtasks described in this application embodiment may include some or all of the knowledge retrieval task, API call task, and database query task; the second prompt word may include a role for processing the display task, context information for determining the display task, the format of the output of the display task, and some or all of the requirement question corresponding to the target display intent; the requirement question may include some or all of the target area range to be displayed, target level information, target device information, and target display style; the context information may include some or all of the queryable document information, callable API information, and queryable database information.

[0146] For example, in this application embodiment, a second prompt word can be constructed based on the target display intent, and the second prompt word can be sent to the large language model for rewriting and decomposing the target display intent through the large language model to obtain the relevant display task.

[0147] For example, the second prompt word constructed based on the target display intent may include, but is not limited to:

[0148] The role used to process the display task is the "Data Analysis Task Planning Module"; the context information of the display task includes queryable documents, callable API information, and queryable databases; the output format of the display task can be JSON format; the requirement question can be "Analyze and display the fault status of equipment model A in the provincial level of China", etc.

[0149] For example, in this application embodiment, the content of the display task determined by the large language model based on the second prompt word includes, but is not limited to, some or all of the following:

[0150] Task 1: Knowledge Retrieval Task.

[0151] For example, assuming the target display intent is "Please analyze and display the fault status of equipment model A in the provincial level of China", the knowledge retrieval task can be to retrieve fault knowledge related to equipment model A.

[0152] Task 2: API call task.

[0153] For example, assuming the target display intent is "Please analyze and display the fault status of equipment model A in the provincial level of China", the API call task can be to call the API to query the number of equipment of model A in certain regions and related fault data.

[0154] Task 3: Database query task.

[0155] For example, assuming the target display intent is "Please analyze and display the fault status of equipment model A in the provincial level of China", the database query task can be to obtain the fault data of equipment model A from the relevant database.

[0156] It is understood that the display task described in the embodiments of this application may include multiple tasks. For example, the display task may be a list of tasks or a set of tasks, and there is no limitation on this.

[0157] S703: Obtain the data information corresponding to the display task from the first database, and determine the data analysis result of the display task based on the data information.

[0158] As an example, in this application embodiment, the first database may include one or more of the following: at least one application information corresponding to the product, at least one attribute information corresponding to the product, attribute information corresponding to the user, and behavioral information corresponding to the user; the second database includes information corresponding to at least one atomic component.

[0159] As an example, the data analysis results described in this application embodiment include some or all of the following:

[0160] Information on the target devices within the target area and the distribution of the target devices within the target area;

[0161] The target hierarchy information includes the target device information corresponding to each level, and the distribution of the target devices corresponding to each level.

[0162] For example, the data analysis results may include some or all of the following:

[0163] (1) Based on the target display intent, determine all devices within the area range and the distribution of all devices within the area range;

[0164] (2) Based on the device types included in the target display intent, determine all devices under the device type;

[0165] (3) Based on the hierarchical information included in the target display intent, determine all devices under the hierarchical level and the distribution of all devices under the hierarchical level;

[0166] (4) A list of all information corresponding to the device application information determined based on the device application information included in the target display intent.

[0167] Understandably, the data analysis results may be obtained by integrating relevant information. For example, when the target display intent includes regional scope, device type, and device application information, the data analysis results are obtained by integrating relevant information about the regional scope, relevant information about the device type, and relevant information about the device application information. In this case, the data analysis results are the device application information under the device type within the regional scope.

[0168] For example, assuming the target display intent is "Please analyze and display the fault status of equipment model A at the provincial level in China", this application embodiment obtains the data information corresponding to the display task from the first database, and after processing and analyzing the data information, the obtained data analysis results include, but are not limited to, some or all of the following:

[0169] Analysis Result 1: Equipment information and distribution within each province in China.

[0170] Analysis Result 2: Specifically regarding the fault information of device model A.

[0171] S704: Based on the data analysis results, obtain the matching first atomic component from the second database.

[0172] As an example, in this embodiment of the application, the second database stores at least one atomic component.

[0173] For example, based on the data analysis results obtained from the target display intent of "Please analyze and display the fault status of equipment model A in the provincial level of China", this application embodiment performs atomic component matching from the second atomic library. For example, the atomic components stored in the second database include atomic components with equipment signal B, atomic components corresponding to fault status, and atomic components within the China region. Then, the first atomic component obtained by matching from the second database in this application embodiment may include the atomic component corresponding to the fault status and the atomic component within the China region.

[0174] S705: After determining that a second atomic component needs to be generated, construct a first prompt word for generating the second atomic component based on the data analysis results.

[0175] In this embodiment of the application, the second atomic component can be understood as an atomic component that is not generated in the current system or does not exist in the atomic component library.

[0176] As an example, embodiments of this application may determine the need to generate the second atomic component in various ways, and are not limited to the following two methods:

[0177] Method 1: When the atomic components in the second database cannot fully meet the matching requirements of the data analysis results, it is determined that the second atomic component needs to be generated.

[0178] For example, based on the example content in S704 above, the first atomic component obtained by matching from the second database based on the target display intent of "Please analyze and display the fault status of equipment model A in the provincial level of China" may include the atomic component corresponding to the fault status and the atomic component of the China region. However, no atomic component related to the equipment model A or the atomic component related to the provincial display granularity was matched. Therefore, it is determined that the second atomic component needs to be generated.

[0179] Method 2: When the target display intent is found to include keywords indicating the addition of a new atomic component, it is determined that the second atomic component needs to be generated.

[0180] For example, the keywords used to indicate the need to add extended atomic components in the embodiments of this application include, but are not limited to, words such as "add," "extend," and "add." For instance, the target display intent could be "add display of equipment fault information based on provincial granularity," etc.

[0181] As an example, the content of the second atomic component in this application embodiment includes, but is not limited to, the icon of the atomic component, and part or all of the display granularity of the atomic component. The display granularity includes, but is not limited to, dimension-based display granularity and scaling-based display granularity. Therefore, the construction of the first prompt word based on the content of the second atomic component is not specifically limited to the following two:

[0182] Construction Scenario 1: When the second atomic component to be generated includes at least one atomic component icon, a first prompt word for generating the at least one atomic component icon is constructed based on the data analysis results; the first prompt word includes the role for generating the at least one atomic component icon, the description information of the at least one atomic component icon, and part or all of the requirement question corresponding to the target display intent.

[0183] For example, the second prompt word for the atomic component icon constructed based on the data analysis results may include, but is not limited to:

[0184] The role used to construct the atomic component icon is "topology component diagram drawing module"; the drawing instructions for the atomic component icon include drawing the icon based on the icon description information; the description information of the atomic component icon; and the requirement issue.

[0185] Construction Scenario 2: When the second atomic component to be generated includes at least one atomic component display granularity, a first prompt word for generating the at least one atomic component display granularity is constructed based on the data analysis results; the display granularity includes part or all of the dimension-based display granularity and the scaling-based display granularity, and the first prompt word includes the role for generating the at least one atomic component display granularity, the descriptive information of the at least one atomic component display granularity, the original map image of the display area corresponding to the display granularity, and part or all of the requirement question corresponding to the target display intent.

[0186] For example, the second cue words for displaying the granularity of atomic components constructed based on the data analysis results may include, but are not limited to:

[0187] The role used to construct the display granularity of the atomic components is "granular drawing module of topology map component"; the drawing instructions for the display granularity of the atomic components include drawing the components based on the description information of the display granularity; the description information of the display granularity of the atomic components; the original map image corresponding to the display area involved in the data analysis results; and the requirement issue.

[0188] S706: Generate the second atomic component based on the first prompt word.

[0189] As an example, there are multiple ways to generate the second atomic component based on the first prompt word in this application embodiment, and these are not limited to the following:

[0190] Generation Method 1: Based on the first prompt word, the second atomic component is generated by modifying the third atomic component in the second database, and the similarity between the third atomic component and the second atomic component is not lower than the first threshold.

[0191] Generation Method 2: Based on the first prompt word, obtain the preliminary architecture of the second atomic component, and gradually optimize it to generate the second atomic component.

[0192] As an example, the image generation model described in this application embodiment can generate relevant atomic component icons based on the first prompt word, and perform transformation rendering on the map layout image to generate a map layout that conforms to a specific geographical range and the specified dimensions, scaling granularity, and user question style.

[0193] As an example, embodiments of this application may store the newly generated second atomic component into an existing atomic component library for future use.

[0194] S707: Based on the first atomic component, the second atomic component, and the data processing and analysis results, determine and output the topology map corresponding to the target display intent.

[0195] As an example, embodiments of this application can construct a third prompt word based on the first atomic component, the second atomic component, and the data processing and analysis results, and input the third prompt word into the large language model to obtain the topological image information of the target display.

[0196] For example, the third prompt word described in the embodiments of this application includes, but is not limited to, some or all of the following:

[0197] (1) The first atomic component and the second atomic component.

[0198] For example, usable map tuple components, fault information icons, device icons, etc.

[0199] (2) The relationship between data information and maps.

[0200] For example, the corresponding data that links equipment status and fault status with the map.

[0201] (3) Detailed data information.

[0202] For example, detailed fault data.

[0203] As an example, the large language model described in this application outputs the encoding of the topological image based on the third prompt word, and the encoding information of the topological image includes, but is not limited to, the following:

[0204] (1) Layout.

[0205] For example, the layout described in the embodiments of this application can be understood as an arrangement scheme. The applied arrangement scheme can be obtained by training and analysis based on empirical data, or it can be selected from existing arrangement schemes, or it can be generated based on the display requirements. There are no limitations on this.

[0206] (2) Customized scaling granularity of map and landmark location information.

[0207] (3) Device type icons (including existing and new ones) and their corresponding landmark locations.

[0208] (4) Relevant information for each device.

[0209] For example, KPI indicators such as fault / maintenance information for each device (including existing and new ones).

[0210] As an example, embodiments of this application can render the encoded topology image through a WebUI interface to generate the final topology display result.

[0211] For example, rendering of complete regional equipment, fault and maintenance information and geographical distribution based on Topo format may include, but is not limited to, device icons rendered based on equipment fault information, new metrics rendered based on maintenance information KPI types, and maps with specific scaling granularity rendered based on regional range.

[0212] As an example, the topology image information displayed to the user in this application embodiment can be viewed by clicking, and the operation and maintenance status of related equipment can be viewed by clicking, the fault analysis report of related equipment can be viewed by clicking, the display dimension can be zoomed by clicking, and the data status of related areas can be displayed by clicking, etc., without limitation.

[0213] Understandably, when the existing atomic components can satisfy the target display intent, step 706 above can be omitted, and the topology map corresponding to the target display intent can be determined and output directly based on the first atomic component and the data analysis results.

[0214] Furthermore, in order to better understand the embodiments of this application, the embodiments of this application are illustrated with examples of the data visualization solutions provided by the embodiments of this application in combination with different user needs scenarios, and are not limited to the following three scenarios.

[0215] Scenario 1: This application embodiment takes a specific product equipment operation failure information related Topic display requirement as an example to demonstrate how the present invention adaptively generates a Topic display image that meets the user's requirement to add display icons.

[0216] For example, suppose the user's intended message is "Please help me analyze the faults of device type A at the provincial level in China." If the atomic components already stored in the data visualization system described in this embodiment cannot meet the display requirements for device type A (i.e., the stored atomic components do not contain an icon corresponding to device type A), then this embodiment can analyze the fault situation of device type A at the provincial level in China for the user in scenario 1. Based on the analysis results, it adaptively generates an icon for device type A and dynamically displays the fault situation of device type A at the provincial level in China based on a topology map, making it more intuitive, specific, and flexible, allowing users to quickly obtain the required information.

[0217] See Figure 8 The diagram shown is a flowchart illustrating a data visualization method provided in an embodiment of this application.

[0218] Step 801: Input the target display intent.

[0219] As an example, in this application embodiment, the user can input the target display intention in the interactive interface; or, the user can input the target display intention through voice recognition via a microphone or other voice device, etc., without limitation.

[0220] For example, in this embodiment of the application, when a user uses the product to perform data analysis on the product and customizes the Topo display image based on the analysis results, the user can describe the target display intent in natural language. The target display intent may include display requirements related to device fault information, display requirements related to area range, hierarchical information, device type, etc., which are not limited here.

[0221] Step 802: The network management system constructs prompt word A based on the target display intent and sends prompt word A to the large language model.

[0222] For example, the network management system described in this application embodiment can construct a prompt word A based on the target display intent, and send an intent rewriting and intent decomposition request to the large language model based on the prompt word A. The intent rewriting and intent decomposition request includes the prompt word A.

[0223] As an example, the content of prompt word A constructed by the network management system based on the target display intent in this embodiment may include, but is not limited to, some or all of the following:

[0224] Content 1: Characters

[0225] For example, the role described is "Data Analysis Task Planning Expert".

[0226] Content 2: Contextual information.

[0227] For example, the context information may include queryable documents, callable API information, or part or all of a queryable database.

[0228] Content 3: Output format requirements.

[0229] For example, the input format is specified as JSON.

[0230] Content 4: The issue of the requirement to indicate the intended purpose of the target display.

[0231] Step 803: The large language model returns a list of executable tasks based on the prompt word A.

[0232] As an example, step 803 described in the embodiments of this application can be implemented based on technologies such as mind chain or ToC, and is not limited here.

[0233] As an example, the task list returned by the large language model based on the prompt word A in this application embodiment includes, but is not limited to, one or more of the following tasks:

[0234] Task 1: Knowledge Retrieval Task.

[0235] For example, the knowledge retrieval task described in this application embodiment can be a fault knowledge-related retrieval task, such as retrieving fault knowledge related to a device with device type A.

[0236] Task 2: API call task.

[0237] For example, the API call task described in this application embodiment can be an API call task related to fault information, such as calling an API to query the number and type of fault alarms of devices of type A in certain regions.

[0238] Task 3: Database query task.

[0239] For example, the database query task described in this application embodiment can be a database query task related to fault data, such as calling a relevant database to query the number of devices of device type A and fault data in certain regions.

[0240] Step 804: The network management system completes the tasks in the task list one by one and obtains the data analysis results.

[0241] For example, the network management system described in this application embodiment can return the data analysis results in the form of text and / or lists.

[0242] For example, the data analysis results include, but are not limited to, some or all of the following:

[0243] Task analysis result 1: Complete equipment and geographical distribution in the region.

[0244] Task analysis result 2: Specific equipment types and a list of faults for each equipment.

[0245] Step 805: The network management system matches existing atomic components based on the data analysis results.

[0246] For example, the network management system described in this application embodiment can perform matching and retrieval from the atomic component library in the data storage layer based on the geographical range, device type, and fault type information in the data analysis results to obtain the corresponding atomic component, such as the first atomic component.

[0247] Step 806: The network management system determines that an atomic component of device type A needs to be added, and then executes the subsequent step 807.

[0248] For example, in this embodiment of the application, when the network management system determines, based on the data analysis results, that there are existing atomic components that cannot match the data analysis results, for example, it determines that the atomic component (icon) of device type A cannot be matched from the existing atomic components, then the network management system determines that it is necessary to expand and add a Topo component, and the network management system can generate the Topo component through subsequent steps.

[0249] For example, in this embodiment of the application, when a user is familiar with the atomic components stored in the data visualization system and knows that the current atomic components or arrangement templates can no longer meet the display requirements when making intent input, the user can directly input the need to add an extended component, or input the content of the extended component to be added, thereby omitting step 806 and directly generating the extended component through subsequent steps.

[0250] Step 807: The network management system constructs a prompt word B based on the icon of the device type A that needs to be added, and sends the prompt word B to the image generation model.

[0251] As an example, the network management system described in this application embodiment can obtain the icon of the atomic component to be added by interacting with the constructed prompt word B and the large image model based on the device type A that needs to be added.

[0252] For example, the prompt word B described in the embodiments of this application includes, but is not limited to, some or all of the following:

[0253] Content 1: Characters.

[0254] For example, the role could be described as a "Topo component diagram drawing master".

[0255] Content 2: Instructions.

[0256] For example, the instruction could be an instruction to generate an image, such as "Generate the most appropriate icon for device type A based on the following description".

[0257] Content 3: Description of the icon for device type A.

[0258] For example, the icon description of device type A in the embodiments of this application can be obtained by the system through analysis and processing of the relevant data of device type A. For example, if the analysis shows that the main function of the device corresponding to device type A is a base station, then a related icon description such as "transmission tower used for signal transmission or reception" can be generated. Alternatively, the icon description can be an icon description input by the user, such as the icon color, icon shape, icon size, icon type, etc., as requested by the user.

[0259] Content 4: The aforementioned requirements.

[0260] Step 808: The image generation model generates an icon for device type A based on prompt word B and sends it to the network management system.

[0261] Step 809: The network management system stores the obtained icon of device type A.

[0262] For example, the network management system described in this application embodiment can store the newly generated Topo component image into an existing Topo component image library for future use.

[0263] Step 810: The network management system constructs a prompt word D and sends the prompt word D to the large language model.

[0264] For example, the prompt word D includes, but is not limited to, some or all of the following:

[0265] Content 1: Available map tuple components, fault information icons, and device icons.

[0266] Content 2: Fault, equipment and map corresponding data.

[0267] Content 3: Detailed fault data information.

[0268] Step 811: The large language model obtains the topology display image information based on the prompt word D, and sends the topology display image information to the network management system.

[0269] For example, the image information for generating the topology display includes, but is not limited to, some or all of the following:

[0270] Content 1: Layout.

[0271] For example, the layout described in the embodiments of this application can be an arrangement scheme, such as which icons are generally placed in which positions.

[0272] Content 2: Customizable zoom granularity for map and landmark location information.

[0273] Content 3: Device type icon and corresponding landmark location.

[0274] At this point, the device type icon can be understood as all required device type icons, including both existing and newly added device type icons.

[0275] Content 4: Fault information for each device.

[0276] Step 812: The network management system renders the generated topology display image information to generate the final topology display map.

[0277] As an example, the network management system described in this application embodiment can generate the final Topic display result through WebUI interface rendering.

[0278] For example, the generated final topology display result includes, but is not limited to, the following:

[0279] Content 1: Complete equipment, fault and maintenance information, and geographical distribution in Topo format for each region.

[0280] For example, Content 1 may include a device icon rendered based on device fault information, and part or all of a map with a specific scaling granularity rendered based on a region range.

[0281] Content 2: Click to view the equipment's operation and maintenance status.

[0282] Content 3: Click to view the equipment failure analysis report.

[0283] Step 813: Display the final fault information result to the user in the form of a Topo diagram on the UI display interface.

[0284] It should be noted that the above embodiments are merely illustrative examples of this application and do not constitute a limitation on the embodiments of this application. For example, embodiments of this application may also obtain other indicator (KPI) topology display information of product equipment operation, etc., which are not limited here.

[0285] Scenario 2: This embodiment takes a specific product equipment operation information related Topic display requirement as an example to demonstrate how the present invention adaptively generates a Topic display image that meets the user's new display granularity requirements.

[0286] For example, suppose the user's intended content is "Please help me analyze the operational status of device type A in various cities of province A in China." If the atomic components already stored in the data visualization system described in this embodiment cannot meet the city-level display granularity requirements (i.e., the stored atomic components do not contain the granularity corresponding to the city-level display), then this embodiment can analyze the operational status of device type A in various cities of province A in China for the user in scenario 2. Based on the analysis results, it adaptively generates atomic components with city-level display granularity and dynamically displays the operational status of device type A in various cities of province A in China based on a topology graph. Users can use zooming and other operations to more intuitively, specifically, and flexibly browse the required information such as the operational status of relevant devices at the regional, provincial, and city levels.

[0287] See Figure 9 The diagram shown is a flowchart illustrating a data visualization method provided in an embodiment of this application.

[0288] Step 901: Input the target display intent.

[0289] Step 902: The network management system constructs a prompt word A based on the target display intent and sends the prompt word A to the large language model.

[0290] Step 903: The large language model returns a list of executable tasks based on the prompt word A.

[0291] As an example, the task list returned by the large language model based on the prompt word A in this application embodiment includes, but is not limited to, one or more of the following tasks:

[0292] Task 1: Knowledge Retrieval Task.

[0293] For example, the knowledge retrieval task described in this application embodiment can be a retrieval task related to operational knowledge, such as retrieving operational knowledge related to device type A.

[0294] Task 2: API call task.

[0295] For example, the API call task described in this application embodiment can be an API call task related to operation information, such as calling an API to query the number of devices of device type A and operation data of each city in province A in China.

[0296] Task 3: Database query task.

[0297] For example, the database query task described in the embodiments of this application can be a database query task related to operational data.

[0298] Step 904: The network management system completes the tasks in the task list one by one and obtains the data analysis results.

[0299] For example, the network management system described in this application embodiment can return the data analysis results in the form of text and / or lists.

[0300] For example, the data analysis results include, but are not limited to, some or all of the following:

[0301] Task Analysis Result 1: Complete equipment distribution in various cities of China region and province A, and equipment distribution within each city.

[0302] Task Analysis Result 2: Equipment Type A and the corresponding operation list for each type A equipment.

[0303] Step 905: The network management system matches existing atomic components based on the data analysis results.

[0304] Step 906: The network management system determines that it is necessary to add atomic components with city-level display granularity, and then executes the subsequent step 907.

[0305] Step 907: The network management system constructs a prompt word C based on the atomic components of the city-level display granularity that need to be added, and sends the prompt word C to the image generation big model.

[0306] As an example, the network management system described in this application embodiment can obtain atomic components that satisfy the display dimensions based on the display dimensions (e.g., the dimensions and scaling granularity of the Topo map) that require the addition of atomic components, by constructing a prompt word C and interacting with a large image model.

[0307] For example, the prompt word C described in the embodiments of this application includes, but is not limited to, some or all of the following:

[0308] Content 1: Characters.

[0309] For example, the role could be described as "the master of drawing dimensions and scaling granularity for the Topo map component".

[0310] Content 2: Instructions.

[0311] For example, the instruction could be an instruction to generate an image, such as "Render the existing image according to the following description 'Generate the most appropriate Topo component image'".

[0312] Content 3: The original map images corresponding to the regions involved in the analysis and processing of user intent by the network management system.

[0313] Content 4: The aforementioned requirements.

[0314] Step 908: The image generation model performs transformation rendering on the map layout image according to the prompt word C, generates a map layout that meets the requirements, and sends it to the network management system.

[0315] Step 909: The network management system stores the obtained map layout that meets the requirements.

[0316] Step 910: The network management system constructs a prompt word D and sends the prompt word D to the large language model.

[0317] Step 911: The large language model obtains the topology display image information based on the prompt word D, and sends the topology display image information to the network management system.

[0318] Step 912: The network management system renders the generated topology display image information to generate the final topology display map.

[0319] Step 913: Display the final operational status to the user in the form of a Topo chart on the UI display interface.

[0320] It is understood that the content of Scenario 1 and Scenario 2 above is only an example of this application. The embodiments of this application can also meet the needs of multiple new extended components at the same time. For example, it can meet the needs of adding new icons and the needs of adding new display granularity. For another example, the embodiments of this application can also meet the needs of displaying multiple indicators of products and devices, or the needs of multiple display granularities, and generate icons of multiple atomic components, and / or granularities of multiple atomic components.

[0321] For example, suppose the user intent is "Please help me analyze the equipment maintenance information, operation information, and fault information of provinces in China"; or, by setting multiple display granularities, the user intent could be "Please help me analyze the equipment operation information of province A in China and the equipment fault information of various cities in province B".

[0322] At this point, the large language model can obtain a task group based on the prompt word A constructed by the network management system through the user's displayed intent. The task group can include a task list corresponding to multiple tasks, such as the task list corresponding to task A, the task list corresponding to task B, the task list corresponding to task C, etc.

[0323] For example, the task list corresponding to task A includes, but is not limited to, one or more of the following tasks:

[0324] Task 1: RAG retrieval task corresponding to Task A.

[0325] Task 2: The API call task corresponding to Task A.

[0326] Task 3: Database query task corresponding to Task A.

[0327] As an example, in this embodiment of the application, when it is necessary to add or expand multiple atomic component icons, a corresponding prompt word can be constructed based on each atomic component icon that needs to be added or expanded, and a corresponding atomic component icon can be generated based on each prompt word through the image large model.

[0328] As an example, in this embodiment of the application, when it is necessary to add or expand multiple atomic component granularities, corresponding prompt words can be constructed based on each atomic component granularity that needs to be added or expanded, and a map corresponding to the atomic component granularity can be generated based on each prompt word through the image large model.

[0329] For the sake of brevity, when the embodiments of this application involve multiple atomic component icons and / or multiple atomic component granularities, the content of Scenario 1 and Scenario 2 above can be combined, and will not be elaborated here.

[0330] Figure 10 This is a structural diagram of a data visualization device provided in an embodiment of this application. The device can be implemented via software and may be part or all of the device itself. The device provided in this embodiment of this application can implement the embodiments of this application. Figures 7-9 In any of the aforementioned processes, the apparatus includes: an input / output module 1010 and a processing module 1020, wherein:

[0331] Input / output module 1010 is used to receive the target display intent of the input;

[0332] The processing module 1020 is configured to: determine a display task based on the target display intent; obtain data information corresponding to the display task from a first database and determine the data analysis result of the display task based on the data information; obtain a matching first atomic component from a second database based on the data analysis result; the first database includes one or more of the following: at least one application information corresponding to the product, at least one attribute information corresponding to the product, attribute information corresponding to the user, and behavioral information corresponding to the user; the second database includes information corresponding to at least one atomic component; after determining that a second atomic component needs to be generated, construct a first prompt word for generating the second atomic component based on the data analysis result; generate the second atomic component based on the first prompt word; and determine the topology map corresponding to the target display intent based on the first atomic component, the second atomic component, and the data processing and analysis result.

[0333] The input / output module 1010 is used to output the topology map.

[0334] In one possible implementation, the processing module 1020 is specifically used for:

[0335] A second prompt word is constructed based on the target display intent; the second prompt word includes a role for handling the display task, context information for determining the display task, the output format of the display task, and part or all of the requirement question corresponding to the target display intent; the requirement question includes part or all of the target area range to be displayed, target level information, target device information, and target display style; the context information includes part or all of the queryable document information, callable API information, and queryable database information; based on the second prompt word, the display task is determined, and the display task includes at least one subtask; the subtask includes part or all of the knowledge retrieval task, API call task, and database query task.

[0336] In one possible implementation, the data analysis result includes the union of some or all of the following:

[0337] Information on the target devices within the target area and the distribution of the target devices within the target area;

[0338] The target hierarchy information includes the target device information corresponding to each level, and the distribution of the target devices corresponding to each level.

[0339] In one possible implementation, the processing module 1020 is further configured to:

[0340] When the atomic components in the second database cannot fully meet the matching requirements of the data analysis results, it is determined that the second atomic component needs to be generated; or,

[0341] When the target display intent is found to include keywords indicating the addition of a new atomic component, it is determined that the second atomic component needs to be generated.

[0342] In one possible implementation, the processing module 1020 is specifically used for:

[0343] When the second atomic component to be generated includes at least one atomic component icon, a first prompt word for generating the at least one atomic component icon is constructed based on the data analysis results; the first prompt word includes the role for generating the at least one atomic component icon, the descriptive information of the at least one atomic component icon, and part or all of the requirement question corresponding to the target display intent.

[0344] In one possible implementation, the processing module 1020 is specifically used for:

[0345] When the second atomic component to be generated includes at least one atomic component display granularity, a first prompt word for generating the at least one atomic component display granularity is constructed based on the data analysis results; the display granularity includes part or all of dimension-based display granularity and scaling-based display granularity, and the first prompt word includes the role for generating the at least one atomic component display granularity, the descriptive information of the at least one atomic component display granularity, the original map image of the display area corresponding to the display granularity, and part or all of the requirement question corresponding to the target display intent.

[0346] In one possible implementation, the processing module 1020 is specifically used for:

[0347] Based on the first atomic component and the second atomic component, the data processing and analysis results are used for topological arrangement to obtain the display encoding of the topological graph;

[0348] The display code is rendered to generate a visualized topology map.

[0349] In one possible implementation, the processing module 1020 is specifically used for:

[0350] The second atomic component is stored in the second database.

[0351] In one possible implementation, the target presentation intent is described based on natural language.

[0352] In one possible implementation, the processing module 1020 is specifically used for:

[0353] Based on the first prompt word, the second atomic component is generated by modifying the third atomic component in the second database, wherein the similarity between the third atomic component and the second atomic component is not lower than a first threshold; or,

[0354] Based on the first prompt word, a preliminary architecture of the second atomic component is obtained, and the second atomic component is generated by gradually optimizing the preliminary architecture.

[0355] The module division in this embodiment is illustrative and represents only one logical functional division. In actual implementation, other division methods are possible. Furthermore, the functional modules in each embodiment of this application can be integrated into a single processor, exist as separate physical entities, or be integrated into a single module. The integrated modules described above can be implemented in hardware or as software functional modules.

[0356] If the integrated module is implemented as a software functional module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a terminal device (which may be a personal computer, network device, etc.) or processor to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0357] Another data visualization device is also provided in this application embodiment. Figure 11 An example of a possible architecture diagram for a data visualization device is provided.

[0358] The data visualization device includes a memory 1101, a processor 1102, a communication interface 1103, and a bus 1104. The memory 1101, processor 1102, and communication interface 1103 are interconnected via the bus 1104.

[0359] The memory 1101 can be a ROM, a static storage device, a dynamic storage device, or RAM. The memory 1101 can store programs. When the program stored in the memory 1101 is executed by the processor 1102, the processor 1102 and the communication interface 1103 are used to execute the aforementioned program. Figures 7-9 Any of the data visualization methods shown. The memory 1101 can also store information such as multiple databases.

[0360] The processor 1102 may be a general-purpose central processing unit (CPU), microprocessor, application-specific integrated circuit (ASIC), graphics processing unit (GPU), or one or more integrated circuits.

[0361] The processor 1102 can also be an integrated circuit chip with signal processing capabilities. In implementation, some or all of the functions of the modeling device of this application can be completed by the integrated logic circuits in the hardware of the processor 1102 or by instructions in software form. The processor 1102 described above can also be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the above embodiments of this application. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this application can be directly embodied in the execution of a hardware decoding processor, or can be executed by a combination of hardware and software modules in the decoding processor. The software modules can be located in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art.

[0362] The communication interface 1103 uses a transceiver module, such as, but not limited to, a transceiver, to enable communication between the modeling device and other devices or communication networks. For example, point cloud data can be acquired through the communication interface 1103.

[0363] Bus 1104 may include a pathway for transmitting information between various components of the modeling apparatus (e.g., memory 1101, processor 1102, communication interface 1103).

[0364] The descriptions of the processes corresponding to the above-mentioned figures each have their own emphasis. For parts of a process that are not described in detail, please refer to the relevant descriptions of other processes.

[0365] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. The computer program product includes one or more computer instructions. When these computer program instructions are loaded and executed on a server or terminal, they generate all or part of the processes or functions described in the embodiments of this application. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic cable, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium accessible to the server or terminal, or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, and magnetic tape), an optical medium (e.g., digital video disk (DVD), etc.), or a semiconductor medium (e.g., solid-state drive).

[0366] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0367] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A data visualization method, characterized in that, include: Receive the input target display intent, and determine the display task based on the target display intent; The data information corresponding to the display task is obtained from the first database, and the data analysis result of the display task is determined based on the data information. Based on the data analysis results, a matching first atomic component is obtained from the second database; the first database includes one or more of the following: at least one application information corresponding to the product, at least one attribute information corresponding to the product, attribute information corresponding to the user, and behavioral information corresponding to the user; the second database includes information corresponding to at least one atomic component. After determining that a second atomic component needs to be generated, a first prompt word for generating the second atomic component is constructed based on the data analysis results. The second atomic component is generated based on the first prompt word; Based on the first atomic component, the second atomic component, and the data processing and analysis results, the topology map corresponding to the target display intent is determined and output.

2. The method as described in claim 1, characterized in that, The received target display intent, and the determination of the display task based on the target display intent, include: A second prompt word is constructed based on the target display intent; the second prompt word includes a role for handling the display task, context information for determining the display task, the output format of the display task, and part or all of the requirement question corresponding to the target display intent; the requirement question includes part or all of the target area range to be displayed, target hierarchy information, target device information, and target display style; the context information includes part or all of the queryable document information, callable API information, and queryable database information. Based on the second prompt word, the display task is determined, and the display task includes at least one sub-task; the sub-task includes some or all of the knowledge retrieval task, API call task, and database query task.

3. The method as described in claim 2, characterized in that, The data analysis results include some or all of the following: Information on the target devices within the target area and the distribution of the target devices within the target area; The target hierarchy information includes the target device information corresponding to each level, and the distribution of the target devices corresponding to each level.

4. The method according to any one of claims 1 to 3, characterized in that, The method further includes: When the atomic components in the second database cannot fully meet the matching requirements of the data analysis results, it is determined that the second atomic component needs to be generated; or, When the target display intent is found to include keywords indicating the addition of a new atomic component, it is determined that the second atomic component needs to be generated.

5. The method according to any one of claims 1 to 4, characterized in that, After determining that a second atomic component needs to be generated, a first prompt word for generating the second atomic component is constructed based on the data analysis results, including: When the second atomic component to be generated includes at least one atomic component icon, a first prompt word for generating the at least one atomic component icon is constructed based on the data analysis results; the first prompt word includes the role for generating the at least one atomic component icon, the descriptive information of the at least one atomic component icon, and part or all of the requirement question corresponding to the target display intent.

6. The method according to any one of claims 1 to 5, characterized in that, After determining that a second atomic component needs to be generated, a first prompt word for generating the second atomic component is constructed based on the data analysis results, including: When the second atomic component to be generated includes at least one atomic component display granularity, a first prompt word for generating the at least one atomic component display granularity is constructed based on the data analysis results; the display granularity includes part or all of dimension-based display granularity and scaling-based display granularity, and the first prompt word includes the role for generating the at least one atomic component display granularity, the descriptive information of the at least one atomic component display granularity, the original map image of the display area corresponding to the display granularity, and part or all of the requirement question corresponding to the target display intent.

7. The method according to any one of claims 1 to 6, characterized in that, The step of outputting the topology map corresponding to the target display intent based on the first atomic component, the second atomic component, and the data processing and analysis results includes: Based on the first atomic component and the second atomic component, the data processing and analysis results are used for topological arrangement to obtain the display encoding of the topological graph; The display code is rendered to generate a visualized topology map.

8. The method according to any one of claims 1 to 7, characterized in that, The method further includes: The second atomic component is stored in the second database.

9. The method according to any one of claims 1 to 8, characterized in that, The target presentation intent is described using natural language.

10. The method according to any one of claims 1 to 9, characterized in that, The generation of the second atomic component based on the first prompt word includes: Based on the first prompt word, the second atomic component is generated by modifying the third atomic component in the second database, wherein the similarity between the third atomic component and the second atomic component is not lower than a first threshold; or, Based on the first prompt word, a preliminary architecture of the second atomic component is obtained, and the second atomic component is generated by gradually optimizing the preliminary architecture.

11. A data visualization device, characterized in that, include: The input / output module is used to receive input and display the intended message. The processing module is used to determine the display task based on the target display intent; The system retrieves data information corresponding to the display task from a first database and determines the data analysis result of the display task based on the data information; it then retrieves a matching first atomic component from a second database based on the data analysis result; the first database includes one or more of the following: at least one application information corresponding to the product, at least one attribute information corresponding to the product, attribute information corresponding to the user, and behavioral information corresponding to the user; the second database includes information corresponding to at least one atomic component. After determining that a second atomic component needs to be generated, a first prompt word for generating the second atomic component is constructed based on the data analysis results. The second atomic component is generated based on the first prompt word; the topology map corresponding to the target display intent is determined based on the first atomic component, the second atomic component, and the data processing and analysis results. The input / output module is used to output the topology map.

12. The apparatus as claimed in claim 11, characterized in that, The processing module is specifically used for: A second prompt word is constructed based on the target display intent; the second prompt word includes a role for processing the display task, context information for determining the display task, the format of the output of the display task, and part or all of the requirement question corresponding to the target display intent; The requirements include the target area to be displayed, target hierarchy information, target device information, and some or all of the target display style; the context information includes some or all of the queryable document information, callable API information, and queryable database information. Based on the second prompt word, the display task is determined, and the display task includes at least one sub-task; the sub-task includes some or all of the knowledge retrieval task, API call task, and database query task.

13. The apparatus as claimed in claim 11 or 12, characterized in that, The data analysis results include the union of some or all of the following: Information on the target devices within the target area and the distribution of the target devices within the target area; The target hierarchy information includes the target device information corresponding to each level, and the distribution of the target devices corresponding to each level.

14. The apparatus as claimed in any one of claims 11 to 13, characterized in that, The processing module is also used for: When the atomic components in the second database cannot fully meet the matching requirements of the data analysis results, it is determined that the second atomic component needs to be generated; or, When the target display intent is found to include keywords indicating the addition of a new atomic component, it is determined that the second atomic component needs to be generated.

15. The apparatus according to any one of claims 11 to 14, characterized in that, The processing module is specifically used for: When the second atomic component to be generated includes at least one atomic component icon, a first prompt word for generating the at least one atomic component icon is constructed based on the data analysis results; The first prompt includes the role used to generate the at least one atomic component icon, the descriptive information of the at least one atomic component icon, and part or all of the requirement question corresponding to the target display intent.

16. The apparatus according to any one of claims 11 to 15, characterized in that, The processing module is specifically used for: When the second atomic component to be generated includes at least one atomic component display granularity, a first prompt word for generating the at least one atomic component display granularity is constructed based on the data analysis results; The display granularity includes some or all of the dimension-based display granularity and the scaling-based display granularity. The first prompt includes the role used to generate the display granularity of the at least one atomic component, the descriptive information of the display granularity of the at least one atomic component, the original map image of the display area corresponding to the display granularity, and some or all of the requirement question corresponding to the target display intent.

17. The apparatus according to any one of claims 11 to 16, characterized in that, The processing module is specifically used for: Based on the first atomic component and the second atomic component, the data processing and analysis results are used for topological arrangement to obtain the display encoding of the topological graph; The display code is rendered to generate a visualized topology map.

18. The apparatus as claimed in any one of claims 11 to 17, characterized in that, The processing module is specifically used for: Based on the first prompt word, the second atomic component is generated by modifying the third atomic component in the second database, wherein the similarity between the third atomic component and the second atomic component is not lower than a first threshold; or, Based on the first prompt word, a preliminary architecture of the second atomic component is obtained, and the second atomic component is generated by gradually optimizing the preliminary architecture.

19. A data visualization device, characterized in that, include: Memory, used to store program instructions; A processor, configured to be coupled to the memory, to invoke program instructions in the memory, and to execute the method as described in any one of claims 1-10.

20. A chip, characterized in that, The chip is connected to a memory and is used to read and execute program code stored in the memory to implement the method as described in any one of claims 1-10.

21. A computer storage medium, characterized in that, The computer storage medium stores a computer program, which, when executed by a computer, causes the computer to perform the method as described in any one of claims 1-10.

22. A computer program product containing instructions, characterized in that, When the computer program product is run on a computer, it causes the computer to perform the method as described in any one of claims 1-10.

Images