Data visualization analysis method, device and equipment based on large model, medium and program product

Abstract

The invention discloses a data visualization analysis method and device based on a large model, a medium and a program product. The method comprises the steps that configuration information of permissions, cue words, file ranges and field rules configured according to application nodes is received; after monitoring that the user triggers the question and answer interface in the target application node, receiving a natural language question of the user; calling the large model to perform problem analysis, extracting problem dimensions and indexes, and generating preliminary query logic in combination with preset information of a knowledge base; acquiring an archive range configured by the application node, and recalling archives; constructing a semantic model according to the node type, and performing secondary processing on semantic model field information in combination with a field mapping rule; and constructing an SQL (Structured Query Language) and executing query, and presenting a query result in a visual form. The application is an enterprise-level application in an ERP system. The analysis method provided by the invention can adapt to complicated business scenes and role requirements of enterprises, and realizes low threshold, high flexibility, strong system compatibility and high accuracy.

Description

Technical Field

[0001] This invention relates to the field of AI-driven enterprise data visualization and analysis technology, and in particular to a data visualization and analysis method, apparatus, equipment, medium, and program product based on a large model. Background Technology

[0002] In the current enterprise data analysis context, the limitations of traditional reporting tools are quite prominent: First, they have a high operational threshold. Users must be familiar with modeling logic and SQL syntax, making it difficult for non-technical personnel to conduct analysis independently. Specifically, users need to model the entire process from data acquisition to result presentation, making the workflow extremely cumbersome. Without a deep understanding of the system, it is difficult to produce analysis that meets their specific needs. Second, data display is rigid and cannot meet the needs of users for personalized secondary analysis. Third, the system has poor compatibility. Analytical functions need to be developed separately for different nodes (such as free reports, document lists, etc.), resulting in low development efficiency. Similar data analysis needs exist in various nodes of different systems, and each need requires separate modeling, which is costly. Furthermore, existing analysis tools have weak integration capabilities with business systems, data synchronization is lagging, and there is a lack of flexible access control and personalized configuration functions. The accuracy of file retrieval is insufficient, and files with the same name are easily confused, which reduces the accuracy of analysis results. It is difficult to adapt to the complex business scenarios and role requirements of enterprises.

[0003] Therefore, there is an urgent need for a data visualization and analysis method based on large models, which can adapt to the complex business scenarios and role requirements of enterprises, and achieve enterprise data analysis with low threshold, high flexibility, strong system compatibility and high accuracy. Summary of the Invention

[0004] In view of this, the present invention provides a data visualization analysis method, apparatus, device, medium and program product based on a large model, which can adapt to the complex business scenarios and role requirements of enterprises, and realize enterprise data analysis with low threshold, high flexibility, strong system compatibility and high accuracy, so as to at least partially solve the problems existing in the prior art.

[0005] To achieve the above objectives, the present invention adopts the following technical solution:

[0006] In a first aspect, the present invention provides a data visualization and analysis method based on a large model, comprising the following steps:

[0007] Receive configuration information for permissions, prompts, file ranges, and field rules configured by application node;

[0008] After detecting that the user has triggered the Q&A interface in the target application node, the system receives the user's natural language question.

[0009] The large model is invoked to analyze the problem, extract the problem dimensions and indicators, and generate preliminary query logic by combining the pre-set information in the knowledge base; the pre-set information in the knowledge base includes prompt words, field mapping rules and file comparison tables.

[0010] Obtain the file range configured for the application node and perform file retrieval;

[0011] A semantic model is constructed based on node type, and the semantic model field information is further processed by combining field mapping rules.

[0012] Construct SQL and execute the query, then present the query results in a visual format.

[0013] Furthermore, the application is an enterprise-level application within an ERP system, and the data visualization and analysis method is based on the page metadata of the enterprise-level application.

[0014] Furthermore, the file retrieval is performed from a vector library, which synchronously receives incremental files.

[0015] Furthermore, the step of obtaining the file range configured by the application node and performing file retrieval includes: obtaining the file range configured by the application node, matching key information containing the primary key, and performing file retrieval.

[0016] Furthermore, the visualization includes presentation in the form of graphics and text, while generating data summaries, insight analysis, and / or attribution analysis.

[0017] Furthermore, the presentation in graphic and textual form includes automatically adapting various chart visualization formats based on data dimensions.

[0018] Furthermore, while receiving users' natural language questions, it also supports receiving and parsing uploaded attachments.

[0019] In a second aspect, the present invention provides a computer device, the device comprising: a processor and a memory;

[0020] The memory is used to store one or more program instructions;

[0021] The processor is configured to run one or more program instructions to perform the steps of a large-model-based data visualization and analysis method as described above.

[0022] Thirdly, the present invention provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the above-described data visualization and analysis method based on a large model.

[0023] Fourthly, the present invention provides a computer program product comprising computer program instructions that, when executed by a processor, implement the steps of the data visualization analysis method based on a large model as described above.

[0024] This invention provides a data visualization and analysis method based on a large model. Through natural language interaction and AI-driven data analysis capabilities, it effectively solves the pain points in traditional enterprise data analysis scenarios, bringing multi-dimensional beneficial technical effects, which are reflected in:

[0025] 1. Significantly lowers the operational threshold and improves analysis efficiency: Users do not need to master SQL syntax or modeling logic. They can quickly obtain data results and graphical displays by asking questions in everyday language, allowing non-technical personnel to complete data analysis independently. Compared with traditional reporting tools, it shortens the analysis preparation time and greatly improves the efficiency of enterprise data utilization.

[0026] 2. Adapt to personalized needs in multiple scenarios and reduce enterprise costs: Supports flexible configuration of permissions, prompts, file ranges and field rules by application node, adapting to different business scenarios to reduce costs, while avoiding the waste of resources from building multiple systems repeatedly.

[0027] 3. Optimize data presentation and insight capabilities to support scientific decision-making: Automatically adapt chart types according to data dimensions and generate data summaries, insightful analyses, or attribution analyses to help users quickly grasp the core information of the data.

[0028] 4. Enhanced System Compatibility and Maintainability: This system can be embedded into various enterprise-level applications within ERP systems, such as documents and reports. It leverages semantic models to achieve unified data acquisition, eliminating the need for separate analysis modules for different systems. This system is suitable for ERP systems with numerous business entities and frequent data analysis needs, particularly in scenarios such as procurement management, sales statistics, and financial analysis. It provides efficient, low-cost, and highly accurate technical support for data-driven decision-making, helping enterprises improve their digital operation capabilities. Attached Figure Description

[0029] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings in the following description are merely exemplary, and those skilled in the art can derive other embodiments based on the provided drawings without creative effort.

[0030] Figure 1 This is an overall architecture diagram of a data visualization and analysis system based on a large model according to an embodiment of the present invention;

[0031] Figure 2 This is a schematic diagram of the data preparation principle in a large-model-based data visualization and analysis system according to an embodiment of the present invention;

[0032] Figure 3 This is a diagram illustrating the questioning principle in a data visualization and analysis system based on a large model, according to an embodiment of the present invention.

[0033] Figure 4 This is a flowchart of a data visualization and analysis method based on a large model according to an embodiment of the present invention. Detailed Implementation

[0034] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0035] This invention primarily relates to AI-driven enterprise data visualization and analysis, which can be embedded into various enterprise-level applications within ERP systems. It supports the analysis needs of multiple data carriers, such as free report nodes and document nodes, helping users without professional modeling capabilities to quickly gain data insights and generate graphical displays. The entire system is built upon shared metadata and semantic models, targeting enterprise data analysis scenarios and providing users with natural language interactive data analysis capabilities.

[0036] Figure 1 This is an overall architecture diagram of a data visualization and analysis system based on a large model, according to an embodiment of the present invention. This system is essentially a lightweight intelligent analysis assistant system for enterprise data visualization and analysis, comprising:

[0037] The foundational capability layer consists of a large model component, a knowledge base component, a service call component, and a workflow component. The large model component parses user natural language queries, extracts dimensions (such as "supplier" and "quarterly") and metrics (such as "total price including tax"), and generates SQL logic. The knowledge base component stores pre-defined prompts, field mapping rules, and a file lookup table, providing contextual support for the large model. The data service component retrieves data from the semantic model based on the SQL logic. The file synchronization service establishes file associations through a lookup table and adds a dynamic listener class to achieve incremental file synchronization to the vector library, realizing dynamic file synchronization. The workflow component manages the entire assistant execution process—"questioning → parsing → data querying → rendering"—ensuring smooth transitions between each stage.

[0038] Custom settings layer components: These include node permission control components, node prompt word components, node file retrieval components, and node field rule components. The node permission control component restricts whether the current node can use the analytics assistant; the node prompt word component optimizes the accuracy of AI responses; the node file retrieval component limits the scope of application files, reducing collisions with files of the same name; and the node field rule component configures the information of required fields and controls the enabling and disabling of fields.

[0039] Interaction layer components: These include a dialogue component, a chart display component, and an attachment component. The dialogue component provides a natural language input box to receive user questions and display AI responses; the chart display component supports various visualization formats such as pie charts, bar charts, and indicator cards, and can automatically adapt to data dimensions (e.g., indicator cards for single indicators, bar charts for multiple dimensions); the attachment component allows users to upload and parse attachment content while asking questions, further facilitating data analysis.

[0040] Figure 2 and Figure 3 This is a diagram illustrating the main working principle of a data visualization and analysis system based on a large model, consisting of a data preparation principle diagram and a questioning principle diagram, including:

[0041] (1) Pre-configuration: Administrators complete application permission allocation, prompt word optimization, file scope definition and field rule configuration through the custom settings layer;

[0042] (2) Interface activation: The user opens the target application node (such as purchase order maintenance) and summons the assistant via a button;

[0043] (3) Natural language questioning: The user inputs a question (such as "Zhang San's order quantity and total order price and tax in the first quarter"), and the dialog component submits the request;

[0044] (4) Problem Analysis: The large model component extracts the problem dimensions and indicators, and generates preliminary query logic by combining the pre-set information in the knowledge base;

[0045] (5) File retrieval: Obtain the file range configured by the application from the vector library, match key information (such as the primary key corresponding to "Zhang San"), and avoid confusion of files with the same name;

[0046] (6) Semantic modeling: The data service component constructs a model based on the node type and performs secondary processing of the model field information in conjunction with field mapping rules;

[0047] (7) Data query and rendering: Construct SQL and execute the query. The chart display component presents the results in the form of graphics and text (such as a dual indicator bar chart + text description), and generates insight analysis, attribution analysis, etc.

[0048] (8) Log recording: The process component records the execution process, which is convenient for subsequent analysis of data anomalies through the analysis nodes.

[0049] Figure 4 This is a flowchart of a data visualization and analysis method based on a large model according to an embodiment of the present invention. (Combined with...) Figure 4 As shown, the method includes the following steps:

[0050] Step S100: Receive configuration information for permissions, prompts, file ranges, and field rules configured according to application nodes.

[0051] The applications are enterprise-level applications within ERP systems. Because they can be embedded into various enterprise-level applications within the ERP system, they offer system compatibility. Furthermore, they can utilize metadata information from the enterprise-level application pages, eliminating the need for redundant development and reducing enterprise costs. Additionally, they can be customized at the application node level (such as configuring permissions, prompts, file scope, and field rules) to meet the needs of different business scenarios.

[0052] Step S200: After detecting that the user has triggered the question and answer interface in the target application node, receive the user's natural language question.

[0053] Step S300: Call the large model to analyze the problem, extract the problem dimensions and indicators, and generate preliminary query logic by combining the knowledge base pre-set information; the knowledge base pre-set information includes prompt words, field mapping rules and file comparison table.

[0054] By using a large model to decompose the problem, low-threshold data analysis is achieved, solving the problem that traditional tools require professional skills, allowing users to obtain analysis results simply by asking questions in natural language.

[0055] Step S400: Obtain the file range configured for the application node and perform file retrieval.

[0056] Specifically, file retrieval is performed from a vector library, which can synchronously receive incremental files, enabling dynamic file synchronization. During retrieval, key information containing the primary key is matched to retrieve files. File retrieval is highly accurate, avoiding confusion with files of the same name and improving the precision of the analysis results.

[0057] Step S500: Construct a semantic model based on the node type, and then perform secondary processing on the semantic model field information in conjunction with field mapping rules;

[0058] Step S600: Construct SQL and execute the query, and present the query results in a visual format.

[0059] The visualization formats include both text and graphics, with the ability to automatically adapt to various chart visualization styles based on data dimensions, displaying a suitable mix of charts and text. Additionally, it can simultaneously generate data summaries, insightful analyses, and / or attribution analyses to help users quickly gain insights.

[0060] In addition, the above method supports receiving and parsing uploaded attachments while receiving users' natural language questions.

[0061] Furthermore, this embodiment also provides a computer device, the device including: a processor and a memory; the memory for storing one or more program instructions; the processor for running one or more program instructions to perform the steps of the data visualization analysis method based on a large model as described above.

[0062] In addition, this embodiment also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the data visualization and analysis method based on a large model as described above.

[0063] Furthermore, this embodiment also provides a computer program product, which includes computer program instructions that, when executed by a processor, implement the steps of the data visualization analysis method based on a large model as described above.

[0064] In this embodiment of the invention, the processor can be an integrated circuit chip with signal processing capabilities. The processor can 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 this embodiment of the invention. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the methods disclosed in this embodiment of the invention can be directly implemented by a hardware decoding processor, or implemented 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. The processor reads information from the storage medium and, in conjunction with its hardware, completes the steps of the above methods. The storage medium can be memory, for example, volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. Non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. Volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of RAM are available, such as Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced Synchronous DRAM (ESDRAM), Synchlink DRAM (SLDRAM), and Direct Rambus RAM (DRRAM).The storage media described in the embodiments of this invention are intended to include, but are not limited to, these and any other suitable types of memory. Those skilled in the art will recognize that the functions described in one or more of the above examples can be implemented using a combination of hardware and software. When applied software, the corresponding functions can be stored in a computer-readable medium or transmitted as one or more instructions or code on a computer-readable medium. Computer-readable media include computer storage media and communication media, wherein communication media include any medium that facilitates the transmission of computer programs from one place to another. Storage media can be any available medium accessible to general-purpose or special-purpose computers. Although the invention has been described in detail above with general description and specific embodiments, modifications or improvements can be made to it, which will be apparent to those skilled in the art. Therefore, all such modifications or improvements made without departing from the spirit of the invention are within the scope of protection claimed by this invention.

[0065] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Any simple modifications, equivalent changes, or alterations made by those skilled in the art using the disclosed technical content shall fall within the protection scope of the present invention.

Claims

1. A data visualization and analysis method based on a large model, characterized in that, Includes the following steps: Receive configuration information for permissions, prompts, file ranges, and field rules configured by application node; After detecting that the user has triggered the Q&A interface in the target application node, the system receives the user's natural language question. The large model is invoked to analyze the problem, extract the problem dimensions and indicators, and generate preliminary query logic by combining the pre-set information in the knowledge base; the pre-set information in the knowledge base includes prompt words, field mapping rules and file comparison tables. Obtain the file range configured for the application node and perform file retrieval; A semantic model is constructed based on node type, and the semantic model field information is further processed by combining field mapping rules. Construct SQL and execute the query, then present the query results in a visual format.

2. The data visualization and analysis method based on a large model according to claim 1, characterized in that, The application is an enterprise-level application within an ERP system, and the data visualization and analysis method is based on the page metadata of the enterprise-level application.

3. The data visualization and analysis method based on a large model according to claim 1, characterized in that, The file retrieval is performed from a vector database, which synchronously receives incremental files.

4. The data visualization and analysis method based on a large model according to claim 1, characterized in that, The process of obtaining the file range configured by the application node and retrieving files includes: obtaining the file range configured by the application node, matching key information containing the primary key, and retrieving files.

5. The data visualization and analysis method based on a large model according to claim 1, characterized in that, The visualization includes presentation in the form of graphics and text, while generating data summaries, insight analysis, and / or attribution analysis.

6. The data visualization and analysis method based on a large model according to claim 5, characterized in that, The presentation in graphic and textual form includes automatically adapting various chart visualization formats based on data dimensions.

7. The data visualization and analysis method based on a large model according to claim 1, characterized in that, While receiving users' natural language questions, it also supports receiving and parsing uploaded attachments.

8. A computer device, characterized in that, The device includes: a processor and a memory; The memory is used to store one or more program instructions; The processor is configured to run one or more program instructions to perform the steps of a large-model-based data visualization analysis method as described in any one of claims 1 to 7.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of a data visualization and analysis method based on a large model as described in any one of claims 1 to 7.

10. A computer program product, characterized in that, The computer program product includes computer program instructions that, when executed by a processor, implement the steps of a data visualization and analysis method based on a large model as described in any one of claims 1 to 7.

Images