Method and system for generating a data index system based on a large model

Abstract

This application relates to a method for generating a data indicator system based on a large model. The method includes constructing large model prompts for the content of the current business system; using the large model to obtain a list of business domains within the current business system; constructing corresponding large model prompts for each business domain in the business domain list; and using the large model to obtain data indicator-related information for each business domain; importing the data indicator-related information into a data indicator platform to complete the construction of the data indicator system; and managing the data on the data indicator platform. This method can quickly generate a data indicator system using large model knowledge, which business experts can then modify. On the one hand, it changes the traditional reliance on business experts, significantly reducing workload. On the other hand, this method also generates SQL statements on how to extract data indicators from the database, greatly reducing the workload of data engineers in developing data indicators.

Description

Technical Field

[0001] This application relates to the field of computer software technology, and in particular to a method and system for generating a data indicator system based on a large model. Background Technology

[0002] An indicator system is a way to systematically organize business metrics from different dimensions. A data indicator system is a systematic summary of business indicators, used to clarify the scope, dimensions, and data collection logic of indicators, and to quickly obtain relevant information. Data indicators are a summary result obtained through data analysis; they are refined and quantified measures of business units, making business objectives describable, measurable, and decomposable. The essence of systematization is to organize data indicators systematically, specifically by classifying and stratifying different attributes of indicators according to business models and standards. Different data indicators have different definitions and logics; these diverse statistical quantities together constitute a data indicator system, giving it indelible value.

[0003] Large models refer to machine learning models with a massive number of parameters and complex computational structures. These models are typically built from deep neural networks and have billions or even hundreds of billions of parameters. The purpose of large models is to improve their expressive power and predictive performance, enabling them to handle more complex tasks and data. Large models have wide applications in various fields, including natural language processing, computer vision, speech recognition, and recommender systems. By training on massive amounts of data to learn complex patterns and features, large models possess stronger generalization capabilities and can make accurate predictions on unseen data.

[0004] In the process of building and operating enterprise indicators, the indicator system plays a fundamental supporting role in enterprise business analysis and is the cornerstone of successful operation. Building an indicator system has always been a challenging pain point, with enterprises often lacking effective methods. Especially when dealing with management decision-making teams, how to conduct refined data extraction and analysis based on existing indicators becomes a challenge. Business analysis teams need to communicate extensively with data development teams and rely heavily on personal experience to formulate and interpret data indicators. Summary of the Invention

[0005] In view of this, this application proposes a method for generating a data indicator system based on a large model to solve the problems reflected in the background technology.

[0006] This application provides a method for generating a data indicator system based on a large model, including: Build large-scale model prompts for the content of the current business system, and use the large-scale model to obtain the business domain types of the current business system; For each of the aforementioned business domains, a corresponding large model prompt word is constructed, and the large model is used to obtain data indicator-related information for each of the aforementioned business domains; Import the relevant information of the data indicators into the data indicator platform to complete the construction of the data indicator system.

[0007] Optionally, the step of using a large model to obtain the business domain types of the current business system further includes: Based on the basic format of the prompt words, the prompt words for the large model are generated. The prompt words for the business system include task descriptions. The specific prompt is: "You are a top expert in the XXX system and also a data metrics business expert. Please list the main business areas of XXX."

[0008] Optionally, the method for generating a data indicator system based on a large model is characterized by: The basic format of the prompt consists of three parts: task description, input data, and output requirements.

[0009] Optionally, corresponding large model prompt words are constructed for each of the aforementioned business domains, including: For each business domain, construct large model prompt words. The prompt words for each business domain include task descriptions related to data indicators. The specific prompt is: "You are a top expert in the XXX system and a top data indicator business analysis expert. Please define a data indicator system for financial management based on XXX. The data indicators should be comprehensive and detailed, with high business value, and output in tabular form. The first column is the indicator name, the second column is the definition, the third column is the calculation method, the fourth column is the main influencing factors of the data indicator, the fifth column is the data table and column position of the main influencing factors in the fourth column, and the last column is the SQL statement to extract the corresponding indicator from the XXX database."

[0010] Optionally, the task description related to the data metrics may also include: The data indicator information should include the business domain, indicator name, indicator definition, indicator calculation method, main influencing factors of the indicator, data table and column positions, and the SQL statement for generating the indicator.

[0011] Optionally, the method for generating a data indicator system based on a large model is characterized by: The main influencing factors of the aforementioned indicators are the main business factors that affect the changes in data indicators. The data table and column positions refer to the relevant data table and column names of data indicators and main influencing factors in the business system database; The SQL statement for generating metrics is an SQL statement that generates data metrics in the business system database.

[0012] Optionally, the data metrics platform further includes: Data management operations involving adding, deleting, modifying, and querying information related to the data indicators on the data indicator platform.

[0013] This application also provides software for constructing a data indicator system based on a large model, characterized in that it obtains corresponding large model prompts based on different business systems and generates data indicator information.

[0014] Optionally, the generated data indicator information can be imported into the data indicator platform for CRUD operations.

[0015] This application also provides an electronic device, characterized in that it includes: Memory, processor, and computer programs stored in said memory and executable on said processor The processor is configured to implement the method of any one of claims 1 to 7 when executing the computer program.

[0016] The beneficial effects of this application are as follows: This method can construct large-scale model prompts for different business systems, call the large-scale model to generate data indicator-related information, including business domain, indicator name, indicator definition, indicator calculation method, main influencing factors of the indicator, data table and column positions, and indicator generation SQL statements, etc.; the generated information is then imported into a data indicator platform, where CRUD (Create, Read, Update, Delete) and other management operations are performed, and business experts modify it accordingly. On the one hand, this method changes the traditional reliance on business experts, greatly reducing workload. On the other hand, this method also generates SQL statements on how to extract data indicators from the database, greatly reducing the workload of data engineers in developing data indicators. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the accompanying drawings required in the description of the embodiments or the prior art are briefly introduced below. Obviously, the accompanying drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0018] Figure 1 The flowchart illustrates a method for generating a data indicator system based on a large model, as disclosed in this application. Figure 2 A flowchart illustrating a specific method for generating a list of business domains disclosed in this application is shown; Figure 3This application discloses a flowchart illustrating a specific method for generating data metrics and related information. Detailed Implementation

[0019] Various exemplary embodiments, features, and aspects of this application will now be described in detail with reference to the accompanying drawings. The same reference numerals in the drawings denote elements that have the same or similar functions. Although various aspects of the embodiments are shown in the drawings, they are not necessarily drawn to scale unless specifically indicated otherwise.

[0020] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0021] The term “exemplary” as used herein means “serving as an example, embodiment, or illustration.” Any embodiment illustrated herein as “exemplary” is not necessarily to be construed as superior to or better than other embodiments.

[0022] Furthermore, to better illustrate this application, numerous specific details are provided in the following detailed embodiments. Those skilled in the art should understand that this application can be implemented without certain specific details. In some instances, methods, means, components, and circuits well-known to those skilled in the art have not been described in detail in order to highlight the main points of this application.

[0023] Large models (also known as foundation models) are machine learning models with a large number of parameters and complex computational structures. These models are typically built from deep neural networks and have billions or even hundreds of billions of parameters. Large models are designed to improve the expressive power and predictive performance of models, enabling them to handle more complex tasks and data. Large models also possess emergent capabilities, a phenomenon where many small entities interact to create a large entity that exhibits characteristics not present in its constituent smaller entities. In the context of models, emergent capabilities refer to the sudden emergence of complex abilities and characteristics that were not present in the smaller models when the training data exceeds a certain scale. These abilities allow the model to comprehensively analyze and solve deeper problems, exhibiting characteristics similar to human thinking and intelligence. Emergent capabilities are one of the most significant characteristics of large models. Large models have wide applications in various fields, including natural language processing, computer vision, speech recognition, and recommendation systems. By training on massive amounts of data to learn complex patterns and features, large models have stronger generalization capabilities and can make accurate predictions on unseen data.

[0024] SQL (Structured Query Language) is a database query and programming language used to access, query, update, and manage relational database systems.

[0025] This application describes a method for constructing a data indicator system for a business system using a large-scale model. Furthermore, this method is specifically designed for generating data indicator systems across multiple business domains.

[0026] like Figure 1 As shown in this embodiment, the method for generating a data indicator system based on a large model is described. This method includes the following steps: S100, This application constructs a large model prompt for the business system based on different business systems, and calls the large model to generate a list of business domains.

[0027] Specifically, taking the Oracle EBS system as an example, the prompt for building the large model of the business system is as follows: "You are a top expert in the Oracle EBS system and also a data metrics business expert. Please list the main business areas of Oracle EBS." The large model then generates a list of business areas. The business areas returned by the large model include financial management, supply chain management, manufacturing management, and human resource management, among others.

[0028] The Oracle EBS system is an enterprise-level comprehensive application software suite. This software mainly uses Oracle's database technology and is based on a modular architecture design. It provides comprehensive management tools and data analysis methods for enterprise management, and integrates multiple management modules such as enterprise resource planning, customer relationship management, and supply chain management.

[0029] like Figure 2 The flowchart shown illustrates a specific process for generating a list of business domains. The user constructs and inputs prompts from the Oracle EBS system's large model. Upon receiving the prompts, the large model begins generating a list of business domains based on the prompt content. After generation, the business domain content is returned to the user. The business domains include financial management, supply chain management, manufacturing management, and human resource management.

[0030] The purpose of the large model prompts for building a business system is to obtain the business areas involved from the business system. The prompts can be, but are not limited to, "You are a top expert in the XX system and also a data metrics business expert. Please list the main business areas of XX", where XX is the name of the system used.

[0031] S200, this application constructs a large model prompt for data indicator information for each business domain in the business domain list generated by S100, and calls the large model to generate data indicator information, including indicator name, indicator definition, indicator calculation method, main influencing factors of the indicator, data table and column position, indicator generation SQL statement, etc.

[0032] Specifically, taking the financial management field as an example using the Oracle EBS system, the prompt for building a large model in the financial management field is as follows: "You are a top expert in the Oracle EBS system and a top expert in data indicator business analysis. Please define a data indicator system for financial management based on Oracle EBS. The data indicators should be comprehensive and detailed, with high business value, and output in tabular form. The first column is the indicator name, the second column is the definition, the third column is the calculation method, the fourth column is the main influencing factors of the data indicator, the fifth column is the data table and column position of the main influencing factors in the fourth column, and the last column is the SQL statement for extracting the corresponding indicators from the Oracle EBS database." like Figure 3 The flowchart shown illustrates a specific process for generating data indicator-related information. The user inputs a prompt word from a large model in the field of financial management. After receiving the prompt word, the large model starts generating data indicator-related information for the field of financial management based on the prompt word content. After generation, the data indicator-related information for the field of financial management is returned to the user, including indicator name, indicator definition, indicator calculation method, main influencing factors of the indicator, data table and column positions, and the SQL statement for indicator generation.

[0033] The purpose of constructing a large model with prompts related to data indicators is to obtain relevant data indicator information from a specified business domain. These prompts may include, but are not limited to, phrases such as, "You are a top expert in the XX system and a top data indicator business analysis expert. Please, based on the financial management data indicator system defined in the XX system, ensure that the data indicators are comprehensive and detailed, have high business value, and be output in tabular form. The first column should be the indicator name, the second the definition, the third the calculation method, the fourth the main influencing factors for the level of the data indicator, the fifth the data table and column position of the main influencing factors in the fourth column, and the last column the SQL statement for extracting the corresponding indicator from the XX system database," where XX is the name of the system used.

[0034] The main influencing factors mentioned above refer to the main business factors affecting changes in image data indicators. In the field of financial management, for example, the main influencing factors of profit margin indicators may be the number of orders, sales price, and procurement cost.

[0035] The data table and column positions mentioned above refer to the names of the data tables and columns related to data indicators and main influencing factors in the business system database.

[0036] The SQL statements used to generate metrics refer to the SQL statements used to generate data metrics in the business system database. These SQL statements are designed to help data developers quickly generate data metrics.

[0037] S300, this application imports the data indicator information generated in S200 into the data indicator platform, and performs management such as adding, deleting, modifying and querying on the data indicator platform.

[0038] Specifically, the process involves entering the indicator names, definitions, calculation methods, main influencing factors, data tables and column positions, and SQL statement information for generating the indicators for the four business areas of financial management, supply chain management, manufacturing management, and human resource management into the data indicator platform. After the information is entered, users can perform data management operations such as creating, reading, updating, and deleting these data on the data indicator platform.

[0039] like Figure 2 As shown, the flowchart for generating the list of business domains includes the following: First, users need to construct prompts for the large model based on the business system they are using. The purpose of these prompts is to extract the relevant business areas from the system. Prompts can include, but are not limited to, phrases like, "You are a top expert in the XX system and also a data metrics business expert. Please list the main business areas of XX," where XX is the name of the system being used. Then, the constructed prompts are input and sent to the large model. The large model will then generate a list of business areas for the system based on the prompts. This list may include financial management, supply chain management, manufacturing management, and human resource management. Finally, the generated list of business areas will be returned to the user.

[0040] like Figure 3 As shown, the flowchart for generating data metric-related information includes the following: First, the user needs to construct prompts for the large model based on the business domain they want to search. The purpose of these prompts is to retrieve data indicator information related to that domain. Prompts can be, but are not limited to, phrases like, "You are a top expert in the XX system and a top data indicator business analysis expert. Please, based on the financial management data indicator system defined in the XX system, provide comprehensive and detailed data indicator coverage with high business value, outputting it in tabular form. The first column should be the indicator name, the second the definition, the third the calculation method, the fourth the main influencing factors for the indicator's level, the fifth the data table and column position of the main influencing factors in the fourth column, and the last column the SQL statement to extract the corresponding indicator from the XX system database," where XX is the name of the system used. Then, the constructed prompts are input and sent to the large model. The large model will then generate data indicator information related to that business domain based on the prompts. This data indicator information includes the indicator name, definition, calculation method, main influencing factors, data table and column position, and the SQL statement for generating the indicator. Finally, the generated data indicator information will be returned to the user.

[0041] The following is the specific implementation process: The specific technical solution is as follows: First, for different business systems, large model prompts are constructed, and the large model is called to generate data indicator-related information, including business domain, indicator name, indicator definition, indicator calculation method, main influencing factors of the indicator, data table and column positions, and indicator generation SQL statement, etc. Then, the generated information is imported into the data indicator platform, and management operations such as adding, deleting, modifying, and querying are performed on the data indicator platform.

[0042] The technical details are as follows: The first step is to build large model prompts for different business systems and then call the large model to generate a list of business domains.

[0043] Taking the Oracle EBS system as an example, the prompt for building the first large model is as follows: "You are a top expert in the Oracle EBS system and also a data metrics business expert. Please list the main business areas of Oracle EBS." The large model then generates a list of business areas. The business areas returned by the large model include financial management, supply chain management, manufacturing management, and human resource management, among others.

[0044] The second step involves constructing large model prompts for each business domain in the business domain list generated in the first step, and calling the large model to generate data indicator-related information, including business domain, indicator name, indicator definition, indicator calculation method, main influencing factors of the indicator, data table and column position, indicator generation SQL statement, etc.

[0045] Taking the financial management field as an example using the Oracle EBS system, "You are a top expert in the Oracle EBS system and a top expert in data indicator business analysis. Please define a data indicator system for financial management based on Oracle EBS. The data indicators should be comprehensive and detailed, with high business value, and output in tabular form. The first column is the indicator name, the second column is the definition, the third column is the calculation method, the fourth column is the main influencing factors of the data indicator, the fifth column is the data table and column position of the main influencing factors in the fourth column, and the last column is the SQL statement for extracting the corresponding indicators from the Oracle EBS database." The third step is to import the data indicator information generated in the second step into the data indicator platform, and then perform management such as adding, deleting, modifying, and querying on the data indicator platform.

[0046] The beneficial effects of this application are that this method can construct large-scale model prompts for different business systems, call the large-scale model to generate data indicator-related information, including business domain, indicator name, indicator definition, indicator calculation method, main influencing factors of the indicator, data table and column positions, and indicator generation SQL statements, etc.; the generated information is then imported into a data indicator platform, where CRUD (Create, Read, Update, Delete) and other management operations are performed, and business experts modify it accordingly. On the one hand, this changes the traditional reliance on business experts, greatly reducing workload. On the other hand, this method also generates SQL statements on how to extract data indicators from the database, greatly reducing the workload of data engineers in developing data indicators.

[0047] The various embodiments of this application have been described above. These descriptions are exemplary and not exhaustive, nor are they limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen to best explain the principles, practical application, or improvement of the technology in the market, or to enable others skilled in the art to understand the embodiments disclosed herein.

Claims

1. A large model-based method for generating a data index system, characterized in that, include: Build large-scale model prompts for the content of the current business system, and use the large-scale model to obtain the business domain types of the current business system; For each of the aforementioned business domains, a corresponding large model prompt word is constructed, and the large model is used to obtain data indicator-related information for each of the aforementioned business domains; Import the relevant information of the data indicators into the data indicator platform to complete the construction of the data indicator system.

2. The method of claim 1, wherein the method further comprises: The method of using a large model to obtain the types of business domains in the current business system includes: Based on the basic format of the prompt words, the prompt words for the large model are generated. The prompt words for the business system include task descriptions. The specific prompt is: "You are a top expert in the XXX system and also a data metrics business expert. Please list the main business areas of XXX." 3.The method of claim 2, wherein, The basic format of the prompt consists of three parts: task description, input data, and output requirements. 4.The method of claim 1, wherein, For each of the aforementioned business domains, construct corresponding large model prompt words, including: For each business domain, construct large model prompt words. The prompt words for each business domain include task descriptions related to data indicators. The specific prompt is: "You are a top expert in the XXX system and a top data indicator business analysis expert. Please define a data indicator system for financial management based on XXX. The data indicators should be comprehensive and detailed, with high business value, and output in tabular form. The first column is the indicator name, the second column is the definition, the third column is the calculation method, the fourth column is the main influencing factors of the data indicator, the fifth column is the data table and column position of the main influencing factors in the fourth column, and the last column is the SQL statement to extract the corresponding indicator from the XXX database." 5. The method of claim 4, wherein the method further comprises: The task description related to the data metrics includes: The data indicator information should include the business domain, indicator name, indicator definition, indicator calculation method, main influencing factors of the indicator, data table and column positions, and the SQL statement for generating the indicator.

6. The method for generating a data indicator system based on a large model as described in claim 5, characterized in that: The main influencing factors of the aforementioned indicators are the main business factors that affect the changes in data indicators. The data table and column positions refer to the relevant data table and column names of data indicators and main influencing factors in the business system database; The SQL statement for generating metrics is an SQL statement that generates data metrics in the business system database.

7. The method of claim 1, wherein the method further comprises: determining a plurality of data indicators based on the plurality of data sets; and determining a plurality of data indicators based on the plurality of data sets. The data metrics platform includes: Data management operations involving adding, deleting, modifying, and querying information related to the data indicators on the data indicator platform.

8. A software for constructing a data index system based on a large model, characterized by, Based on different business systems, obtain corresponding large model prompts and generate data indicator information.

9. The method of claim 8, wherein, The generated data metrics information can also be imported into the data metrics platform for CRUD operations.

10. An electronic device, comprising: include: Memory, processor, and computer programs stored in said memory and executable on said processor The processor is configured to implement the method of any one of claims 1 to 7 when executing the computer program.

Images