LLM-integrated BI platform self-service method suitable for manufacturing enterprises

Abstract

The embodiment of the invention relates to the technical field of digital enterprise management, and particularly discloses an LLM-integrated BI platform self-service method for a manufacturing enterprise. According to the embodiment of the invention, enterprise business main data is collected and summarized for preprocessing; performing layout configuration, and establishing a BI platform; an LLM big language model is constructed and deployed, and the data warehouse, the BI platform and the LLM big language model are butted and interacted; receiving a use demand of a user, performing analysis and identification through an LLM large language model, and extracting four demand elements of the user; and performing demand matching and interactive feedback processing in a data report, a data instrument disk table and an instrument disk table directory of the BI platform. The problems of high threshold, insufficient personnel and low use efficiency of BI tools used by medium and small enterprises can be solved, so that the enterprise is helped to improve the use efficiency and decision-making ability of production and operation data, the transparency and stability of enterprise management are enhanced, the enterprise is helped to reduce cost and increase efficiency, and the market competitiveness is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of digital enterprise management, and in particular relates to a BI platform self-service method for integrating LLM suitable for manufacturing enterprises. Background Art

[0002] Today, all industries are undergoing digital transformation, and digital business management has become the only way for companies to improve their competitiveness, reduce costs and increase capital.

[0003] On the one hand, BI data visualization and analysis is a very important part of enterprise digital transformation. BI tools have become an essential tool for enterprises to open up internal data connections and carry out digital management transformation. However, many small and medium-sized enterprises, especially manufacturing factories, lack employees who can use BI tools. In the process of digital transformation of manufacturing enterprises, the production, quality, process and supply chain data collected and connected are difficult for managers from various departments to obtain, process and apply, making it difficult to serve the daily production, operation and management of the enterprise conveniently and quickly.

[0004] On the other hand, the development of artificial intelligence for natural language interaction has made breakthrough progress. Many companies in the world have launched commercial LLMs, which lower the threshold for the use of BI tools through the general intelligence of open large language models. However, when it comes to enterprise applications, it still faces the illusion problem of large models, and it is impossible to connect the specialized knowledge and context within the enterprise for implementation, and there is the problem of leakage of enterprise data and business secrets.

[0005] Therefore, in the existing technology, how to effectively integrate LLM and BI systems, develop a BI platform that can be used conveniently and quickly by people who do not understand the BI platform, and provide a feasible implementation method has become an urgent need for small and medium-sized enterprises in data processing applications. Summary of the invention

[0006] The purpose of the embodiment of the present invention is to provide a self-service method for a BI platform integrating LLM for a manufacturing enterprise, aiming to solve the problems raised in the background technology.

[0007] To achieve the above objectives, the embodiments of the present invention provide the following technical solutions:

[0008] A self-service method for a BI platform integrated with LLM for a manufacturing enterprise, the method specifically comprising the following steps:

[0009] In the data warehouse, collect and summarize the enterprise business master data, perform preprocessing, and combine the indicators of concern to multiple departments to generate data warehouse tables, data calculation scheduled tasks and data update directories;

[0010] Layout and configure the data warehouse table, the data calculation scheduled task and the data update directory, deploy BI tools to generate data reports, data dashboards and dashboard catalogs, and form a BI platform;

[0011] Obtain the company's basic business processes, management and evaluation indicators, calculation definition information and relationship information, build and deploy the LLM large language model, and connect and interact with the data warehouse, BI platform and LLM large language model through the developed API interface;

[0012] In the BI platform, the user's usage requirements are received, parsed and identified through the LLM large language model, and the four elements of the user's requirements are extracted;

[0013] Based on the four elements of demand, demand matching and interactive feedback processing are performed in the data reports, data dashboards and dashboard catalogs of the BI platform.

[0014] As a further limitation of the technical solution of the embodiment of the present invention, the collection of enterprise business master data, preprocessing, and combining multiple departments' attention indicators to generate a data warehouse, a data warehouse table, a data calculation scheduled task and a data update directory specifically include the following steps:

[0015] Collect data from multiple enterprise systems and aggregate them into enterprise business master data in the data warehouse;

[0016] Perform data cleaning and dimensional modeling on the enterprise business master data to form data wide tables, data application reports and data catalogs of different themes of the enterprise;

[0017] In combination with the indicators of concern to multiple departments, the data wide tables, data application reports and data directories are calculated, stored and automatically updated to generate data warehouses, data warehouse tables, data calculation scheduled tasks and data update directories.

[0018] As a further limitation of the technical solution of the embodiment of the present invention, the multiple enterprise system data include: production workshop equipment system data, MES production execution management system data, quality management data, ERP system data, OA system data, financial management data, warehouse and supply chain management system data, energy management data, CRM sales management system data and environmental security management data.

[0019] As a further limitation of the technical solution of the embodiment of the present invention, the method of combining multiple departments' attention indicators, calculating, storing and automatically updating data wide tables, data application reports and data directories, and generating data warehouses, data warehouse tables, data calculation scheduled tasks and data update directories specifically includes the following steps:

[0020] Obtain information on business needs for daily production and operation of enterprises;

[0021] Analyze the business demand information and determine the indicators that multiple departments should focus on;

[0022] In the data warehouse, multiple departmental focus indicators are combined with data wide tables, data application reports and data directories for calculation, storage and automatic update to generate data warehouses, data warehouse tables, data calculation scheduled tasks and data update directories.

[0023] As a further limitation of the technical solution of the embodiment of the present invention, the layout configuration of the data warehouse table, the data calculation scheduled task and the data update directory, the deployment of BI tools to generate data reports, data dashboard tables and dashboard table directories, and the formation of a BI platform specifically include the following steps:

[0024] Obtaining screening conditions for layout configuration, wherein the screening conditions include a dimension value range and / or a time value range;

[0025] According to the screening conditions, develop and configure the data warehouse table, the data calculation scheduled task and the data update directory, deploy BI tools and configure the generation of data reports, data dashboards and dashboard catalogs;

[0026] A BI platform is formed based on the data report, the data dashboard and the dashboard catalog.

[0027] As a further limitation of the technical solution of the embodiment of the present invention, the acquisition of the company's basic business processes, business management assessment indicators, calculation definition information and association relationship information, the construction and deployment of the LLM large language model, and the docking and interaction of the data warehouse, the BI platform and the LLM large language model through the developed API interface specifically include the following steps:

[0028] Obtain the company's basic business processes, management and evaluation indicators, calculation definition information, and relationship information;

[0029] In the preset local LLM model, the LLM large language model is constructed through the company's basic business processes, management and management assessment indicators, calculation definition information and association relationship information;

[0030] Deploy the LLM large language model;

[0031] Through the developed API interface, the data warehouse, the BI platform and the LLM large language model are connected and interacted.

[0032] As a further limitation of the technical solution of the embodiment of the present invention, in the preset local LLM model, the construction of the LLM large language model through the company's basic business processes, business management assessment indicators, calculation definition information and association relationship information specifically includes the following steps:

[0033] According to the preset division ratio, the company's basic business processes, management and operation assessment indicators, calculation definition information and association relationship information are divided to generate test sets and verification sets;

[0034] In the preset local LLM model, model learning and training are performed through the test set and the validation set to build an LLM large language model.

[0035] As a further limitation of the technical solution of the embodiment of the present invention, in the BI platform, receiving the user's usage requirements, parsing and identifying through the LLM large language model, and extracting the four elements of the user's requirements specifically include the following steps:

[0036] In the BI platform, receiving a user's usage requirement;

[0037] The usage requirements are parsed and identified through the LLM large language model, and four elements of requirements including atomic indicators, statistical cycles, business scope constraints and dimensions / granularity are extracted.

[0038] As a further limitation of the technical solution of the embodiment of the present invention, the usage requirement is in voice or text form; the atomic indicator is the indicator vocabulary of product quantity, order amount, quality qualification rate, equipment failure rate and inventory days in the usage requirement; the statistical period is the time range vocabulary in the usage requirement; the business scope constraint is the business limitation condition vocabulary in the usage requirement; the dimension / the granularity is the indicator level vocabulary in the usage requirement.

[0039] As a further limitation of the technical solution of the embodiment of the present invention, the demand matching and interactive feedback processing based on the four elements of demand in the data reports, data dashboards and dashboard catalogs of the BI platform specifically includes the following steps:

[0040] Based on the four elements of the requirements, matching the requirements among the data reports, data dashboards and dashboard catalogs of the BI platform is performed, and determining whether a match is possible;

[0041] If a match is found, the matching related content is obtained, a link list is generated, and displayed in the user interaction page of the BI platform, and the user's feedback selection is received;

[0042] If no match can be made or the feedback is that the selection is useless, a query SQL is generated based on the four elements of the requirement, and the query is passed to the data warehouse for SQL query, the query data result is obtained, a web link is generated, and the link is displayed in the user interaction page of the BI platform.

[0043] Compared with the prior art, the present invention has the following beneficial effects:

[0044] The embodiment of the present invention lowers the daily usage threshold of the BI platform through the access of LLM, which can not only help users quickly find and use existing reports and dashboards, but also flexibly query and use data through the general intelligence of LLM. In the process of interaction with users, it can further learn and adapt to local daily context habits, gradually improve the accuracy and efficiency of data search, and further enhance the applicability and practicality. It can solve the problems of high threshold, insufficient personnel and low efficiency in using BI tools for small and medium-sized enterprises, thereby helping enterprises to improve the efficiency of using production and operation data and decision-making capabilities, and shift from experience-driven to data-driven, effectively enhancing the transparency and stability of enterprise management, and helping enterprises to reduce costs and increase efficiency and improve market competitiveness. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention.

[0046] Figure 1 A flow chart of a method provided by an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0047] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0048] It is understandable that in the existing technology, how to effectively integrate LLM and BI systems, develop a BI platform that can be used conveniently and quickly by people who do not understand the BI platform, and provide a feasible implementation method, has become an urgent need for small and medium-sized enterprises in data processing applications.

[0049] To solve the above problems, the embodiment of the present invention collects and summarizes the enterprise business master data in the data warehouse, performs preprocessing, and combines the indicators of concern of multiple departments to generate data warehouse tables, data calculation scheduled tasks and data update directories; layouts and configures the data warehouse tables, data calculation scheduled tasks and data update directories, generates data reports, data dashboard tables and dashboard table directories, and establishes a BI platform; obtains the company's basic business processes, business management assessment indicators, calculation definition information and association relationship information, builds and deploys the LLM large language model, and connects and interacts the data warehouse, BI platform and LLM large language model through the developed API interface; in the BI platform, receives the user's usage requirements, analyzes and identifies through the LLM large language model, and extracts the four elements of the user's requirements; based on the four elements of requirements, performs demand matching and interactive feedback processing in the data reports, data dashboard tables and dashboard table directories of the BI platform. It can solve the problems of high threshold, insufficient personnel and low efficiency in the use of BI tools by small and medium-sized enterprises, thereby helping enterprises to improve the efficiency and decision-making ability of production and operation data, enhance the transparency and stability of enterprise management, and help enterprises reduce costs and increase efficiency and improve market competitiveness.

[0050] Figure 1 A flow chart of a method provided by an embodiment of the present invention is shown.

[0051] Specifically, a self-service method for a BI platform integrated with LLM for a manufacturing enterprise includes the following steps:

[0052] Step S101, in the data warehouse, collect and summarize the enterprise business master data, perform preprocessing, and combine the indicators of concern to multiple departments to generate data warehouse tables, data calculation scheduled tasks and data update directories.

[0053] In an embodiment of the present invention, data of multiple enterprise systems including production workshop equipment system data, MES production execution management system data, quality management data, ERP system data, OA system data, financial management data, warehouse and supply chain management system data, energy management data, CRM sales management system data and environmental security management data are collected. In a data warehouse, data of multiple enterprise systems are summarized to generate enterprise business master data, which is then cleaned and dimensionally modeled to form data wide tables, data application reports and data directories of different themes for the enterprise, thereby obtaining business demand information for daily production and operation of the enterprise, analyzing the business demand information, determining indicators of concern to multiple departments, and in a data warehouse, combining the indicators of concern to multiple departments with data wide tables, data application reports and data directories for calculation, storage and automatic update, to generate a data warehouse, data warehouse tables, data calculation timed tasks and data update directories.

[0054] Step S102, layout and configure the data warehouse table, the data calculation scheduled task and the data update directory, deploy BI tools to generate data reports, data dashboard tables and dashboard table directories, and form a BI platform.

[0055] In an embodiment of the present invention, by obtaining filtering conditions for the layout configuration including dimension value ranges and / or time value ranges, and then developing and configuring data warehouse tables, data calculation schedule tasks, and data update directories according to the filtering conditions, BI tools are deployed and BI component configurations are used to generate data reports, data dashboards, and dashboard table directories, and then a BI platform is formed based on the data reports, data dashboards, and dashboard table directories.

[0056] Step S103, obtain the company's basic business processes, management and operation assessment indicators, calculation definition information and association relationship information, build and deploy the LLM large language model, and connect and interact the data warehouse, BI platform and LLM large language model through the developed API interface.

[0057] In an embodiment of the present invention, by obtaining the company's basic business processes, business management assessment indicators, calculation definition information and association relationship information, and then dividing the company's basic business processes, business management assessment indicators, calculation definition information and association relationship information according to a preset division ratio, a test set and a verification set are generated, and in the preset local LLM model, model learning and training are performed through the test set and the verification set to construct an LLM large language model, and then the LLM large language model is deployed for application, and through the developed API interface, the data warehouse, BI platform and LLM large language model are connected and interacted.

[0058] It can be understood that the calculation definition information is the digital definition information of the company's basic business processes and operating management assessment indicators; the association relationship information is the association relationship between the company's basic business processes and operating management assessment indicators.

[0059] It can be understood that API interface is application program interface.

[0060] Step S104, in the BI platform, receiving the user's usage requirements, parsing and identifying them through the LLM large language model, and extracting the four elements of the user's requirements.

[0061] In an embodiment of the present invention, in the BI platform, the user's usage requirements input in the form of voice or text are received, and the usage requirements are parsed and identified through the LLM large language model, and the four elements of the requirements including atomic indicators, statistical cycles, business scope constraints and dimensions / granularity are extracted. Among them, the atomic indicators are indicator vocabulary such as product quantity, order amount, quality qualification rate, equipment failure rate and inventory days in the usage requirements. If they are not identified, the LLM large language model will retrieve semantically related indicators in the data warehouse, list them and push them to the user for selection. The user can only select one atomic indicator at a time for data acquisition. ; Statistical period is the word that indicates the time range in the usage requirements, such as the past week, this month, the past 10 days, etc. If it is not recognized, the user will be replied to clarify the time range of the data requirement. The user can select requirements for multiple time periods, and a corresponding data result will be returned for each time period subsequently; Business scope constraints are the words that indicate the business-limited conditions in the usage requirements to filter indicators (such as the top five, the bottom 10, late attendance (employees), decreased profits (departments or products), etc.); Dimension / granularity is the word that indicates the indicator level in the usage requirements, such as a region, a branch or store, product category, etc.

[0062] Step S105 , based on the four elements of the requirements, demand matching and interactive feedback processing are performed in the data reports, data dashboards and dashboard catalogs of the BI platform.

[0063] In an embodiment of the present invention, based on the four elements of demand, demand matching is performed in the data reports, data dashboards and dashboard catalogs of the BI platform, and it is determined whether they can be matched, and then different processing is performed. Specifically: if they can be matched, the matching related content is obtained, and a link list is generated, which is displayed on the user interaction page of the BI platform, and the user's feedback selection is received, and the feedback selection includes "useful" and "useless"; if they cannot be matched or the feedback selection is useless, a query SQL is generated according to the four elements of demand, and passed to the data warehouse for SQL query, the query data results are obtained, and a web page link is generated and displayed on the user interaction page of the BI platform.

[0064] It is understandable that during the use of the LLM large language model, the user's needs and selection behaviors will be recorded and later learned and optimized in order to provide more accurate data recommendations and data output.

[0065] It should be understood that, although each step in the flow chart of each embodiment of the present invention is shown in sequence according to the indication of the arrow, these steps are not necessarily performed in sequence according to the order indicated by the arrow. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be performed in other orders. Moreover, at least a portion of the steps in each embodiment may include a plurality of sub-steps or a plurality of stages, and these sub-steps or stages are not necessarily performed at the same time, but can be performed at different times, and the execution order of these sub-steps or stages is not necessarily performed in sequence, but can be performed in turn or alternately with at least a portion of other steps or sub-steps or stages of other steps.

[0066] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0067] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0068] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

[0069] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A self-service method for the BI platform with integrated LLM for manufacturing enterprises, characterized by: The method specifically comprises the following steps: In the data warehouse, collect and summarize the enterprise business master data, perform preprocessing, and combine the indicators of concern to multiple departments to generate data warehouse tables, data calculation scheduled tasks and data update directories; Layout and configure the data warehouse table, the data calculation scheduled task and the data update directory, deploy BI tools to generate data reports, data dashboards and dashboard catalogs, and form a BI platform; Obtain the company's basic business processes, management and evaluation indicators, calculation definition information and relationship information, build and deploy the LLM large language model, and connect and interact with the data warehouse, BI platform and LLM large language model through the developed API interface; In the BI platform, the user's usage requirements are received, parsed and identified through the LLM large language model, and the four elements of the user's requirements are extracted; Based on the four elements of demand, demand matching and interactive feedback processing are performed in the data reports, data dashboards and dashboard catalogs of the BI platform.

2. The self-service method for the BI platform integrated with LLM applicable to manufacturing enterprises according to claim 1, characterized in that: The process of collecting the enterprise business master data, preprocessing it, and combining the indicators of interest of multiple departments to generate a data warehouse, a data warehouse table, a data calculation scheduled task, and a data update directory specifically includes the following steps: Collect data from multiple enterprise systems and aggregate them into enterprise business master data in the data warehouse; Perform data cleaning and dimensional modeling on the enterprise business master data to form data wide tables, data application reports and data catalogs of different themes of the enterprise; In combination with the indicators of concern to multiple departments, the data wide tables, data application reports and data directories are calculated, stored and automatically updated to generate data warehouses, data warehouse tables, data calculation scheduled tasks and data update directories.

3. The self-service method for the BI platform integrated with LLM applicable to manufacturing enterprises according to claim 2 is characterized in that: The multiple enterprise system data include: production workshop equipment system data, MES production execution management system data, quality management data, ERP system data, OA system data, financial management data, warehouse and supply chain management system data, energy management data, CRM sales management system data and environmental security management data.

4. The self-service method for the BI platform integrated with LLM applicable to manufacturing enterprises according to claim 2, characterized in that: The method of combining the indicators of concern of multiple departments, calculating, storing and automatically updating the data wide table, data application report and data directory, and generating the data warehouse, data warehouse table, data calculation scheduled task and data update directory specifically includes the following steps: Obtain information on business needs for daily production and operation of enterprises; Analyze the business demand information and determine the indicators that multiple departments should focus on; In the data warehouse, multiple departmental focus indicators are combined with data wide tables, data application reports and data directories for calculation, storage and automatic update to generate data warehouses, data warehouse tables, data calculation scheduled tasks and data update directories.

5. The self-service method for the BI platform integrated with LLM applicable to manufacturing enterprises according to claim 1, characterized in that: The layout configuration of the data warehouse table, the data calculation scheduled task and the data update directory, the deployment of BI tools to generate data reports, data dashboard tables and dashboard table directories, and the formation of a BI platform specifically include the following steps: Obtaining screening conditions for layout configuration, wherein the screening conditions include a dimension value range and / or a time value range; According to the screening conditions, develop and configure the data warehouse table, the data calculation scheduled task and the data update directory, deploy BI tools and configure the generation of data reports, data dashboards and dashboard catalogs; A BI platform is formed based on the data report, the data dashboard and the dashboard catalog.

6. The self-service method for the BI platform integrated with LLM applicable to manufacturing enterprises according to claim 1, characterized in that: The acquisition of the company's basic business processes, management and evaluation indicators, calculation definition information and association relationship information, construction and deployment of the LLM large language model, and the docking and interaction of the data warehouse, BI platform and LLM large language model through the developed API interface specifically include the following steps: Obtain the company's basic business processes, management and evaluation indicators, calculation definition information, and relationship information; In the preset local LLM model, the LLM large language model is constructed through the company's basic business processes, management and management assessment indicators, calculation definition information and association relationship information; Deploy the LLM large language model; Through the developed API interface, the data warehouse, the BI platform and the LLM large language model are connected and interacted.

7. The self-service method for the BI platform integrated with LLM applicable to manufacturing enterprises according to claim 6, characterized in that: In the preset local LLM model, the construction of the LLM large language model includes the following steps through the company's basic business processes, business management assessment indicators, calculation definition information and association relationship information: According to the preset division ratio, the company's basic business processes, management and operation assessment indicators, calculation definition information and association relationship information are divided to generate test sets and verification sets; In the preset local LLM model, model learning and training are performed through the test set and the validation set to build an LLM large language model.

8. The self-service method for the BI platform integrated with LLM applicable to manufacturing enterprises according to claim 1, characterized in that: In the BI platform, receiving the user's usage requirements, parsing and identifying through the LLM large language model, and extracting the four elements of the user's requirements specifically include the following steps: In the BI platform, receiving a user's usage requirement; The usage requirements are parsed and identified through the LLM large language model, and four elements of the requirements including atomic indicators, statistical cycles, business scope constraints and dimensions / granularity are extracted.

9. The self-service method for the BI platform integrated with LLM applicable to manufacturing enterprises according to claim 8, characterized in that: The usage requirements are in voice or text form; the atomic indicators are indicator vocabulary such as product quantity, order amount, quality qualification rate, equipment failure rate and inventory days in the usage requirements; the statistical period is the time range vocabulary in the usage requirements; the business scope constraint is the business limitation condition vocabulary in the usage requirements; the dimension / the granularity is the indicator level vocabulary in the usage requirements.

10. The self-service method for the BI platform integrated with LLM applicable to manufacturing enterprises according to claim 1, characterized in that: Based on the four elements of the requirements, the requirements matching and interactive feedback processing in the data reports, data dashboards and dashboard catalogs of the BI platform specifically includes the following steps: Based on the four elements of the requirements, matching the requirements among the data reports, data dashboards and dashboard catalogs of the BI platform is performed, and determining whether a match is possible; If a match is found, the matching related content is obtained, a link list is generated, and displayed in the user interaction page of the BI platform, and the user's feedback selection is received; If no match can be made or the feedback is that the selection is useless, a query SQL is generated based on the four elements of the requirement, and the query is passed to the data warehouse for SQL query, the query data result is obtained, a web link is generated, and the link is displayed in the user interaction page of the BI platform.

Images