Business data processing method and device, computer equipment and readable storage medium

Abstract

The invention relates to a business data processing method and device, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: acquiring data analysis demand information for target business data; performing demand analysis on the data analysis demand information to obtain analysis model description information for the target business data; based on the analysis model description information and a preset analysis model template corresponding to the target business data, constructing a data analysis model for the target business data; updating the target business data into the data analysis model to obtain an updated data analysis model; and performing data analysis based on the updated data analysis model according to a preset data analysis mode to obtain a data analysis result for the target business data. By adopting the method, the accuracy of business data analysis can be improved.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a business data processing method, apparatus, computer equipment, computer-readable storage medium, and computer program product. Background Art

[0002] Business data analysis has become an indispensable part of the corporate decision-making process. Through in-depth mining and analysis of massive business data, companies can gain insight into market trends, optimize operational strategies, and improve financial performance, thereby gaining a favorable position in the fierce market competition. However, the accuracy of the analysis results of traditional business data analysis methods is low, and it is even easy to produce misleading illusions. For example, in financial analysis, if abnormal transactions are not effectively eliminated or the impact of accounting policy changes is not adjusted, profit indicators may be falsely high or low, which may mislead corporate decision makers to make wrong judgments. Summary of the invention

[0003] Based on this, it is necessary to provide a business data processing method, device, computer equipment, computer-readable storage medium and computer program product that can improve the accuracy of business data analysis in response to the above technical problems.

[0004] In a first aspect, the present application provides a business data processing method, comprising:

[0005] Obtain data analysis requirements information for target business data;

[0006] Conduct demand analysis on data analysis demand information to obtain analysis model description information for target business data;

[0007] Based on the analysis model description information and the preset analysis model template corresponding to the target business data, a data analysis model for the target business data is constructed;

[0008] Update the target business data into the data analysis model to obtain an updated data analysis model;

[0009] Perform data analysis based on the updated data analysis model according to the preset data analysis method to obtain data analysis results for the target business data.

[0010] In a second aspect, the present application also provides a business data processing device, including:

[0011] An analysis requirement information acquisition module is used to acquire data analysis requirement information for target business data;

[0012] The demand analysis processing module is used to perform demand analysis on data analysis demand information and obtain analysis model description information for target business data;

[0013] A data analysis model building module is used to build a data analysis model for target business data based on the analysis model description information and the preset analysis model template corresponding to the target business data;

[0014] A data analysis model updating module is used to update the target business data into the data analysis model to obtain an updated data analysis model;

[0015] The data analysis processing module is used to perform data analysis based on the updated data analysis model according to the preset data analysis method to obtain data analysis results for the target business data.

[0016] In a third aspect, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above business data processing method when executing the computer program.

[0017] In a fourth aspect, the present application also provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the steps of the above business data processing method are implemented.

[0018] In a fifth aspect, the present application also provides a computer program product, including a computer program, which implements the steps of the above business data processing method when executed by a processor.

[0019] The above-mentioned business data processing method, device, computer equipment, computer-readable storage medium and computer program product perform demand analysis on the data analysis demand information of the target business data to obtain analysis model description information, build a data analysis model for the target business data based on the analysis model description information and a preset analysis model template, update the target business data into the data analysis model, and then perform data analysis based on the updated data analysis model in accordance with a preset data analysis method to obtain data analysis results of the target business data. In the process of business data processing, after performing demand analysis on the data analysis demand information, build a data analysis model based on the analysis model description information and the analysis model template, update the target business data into the analysis data analysis model, and then perform data analysis in accordance with a preset data analysis method. The data analysis demand of the target business data can be fully mined through the data analysis model, and multi-dimensional data analysis of the target business data can be achieved, thereby improving the accuracy of business data analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0021] Figure 1 An application environment diagram of a business data processing method in an embodiment;

[0022] Figure 2 A flowchart of a method for processing business data in one embodiment;

[0023] Figure 3 A schematic diagram of a process for building a data analysis model in one embodiment;

[0024] Figure 4 A schematic diagram of a process for updating a financial analysis model in one embodiment;

[0025] Figure 5 A schematic diagram of a process for generating a financial analysis report in one embodiment;

[0026] Figure 6 A schematic diagram of a financial report outline in one embodiment;

[0027] Figure 7 is a structural block diagram of a business data processing device in an embodiment;

[0028] Figure 8 is an internal structure diagram of a computer device in one embodiment;

[0029] Fig. 9 FIG. 4 is a diagram showing the internal structure of a computer device in yet another embodiment. DETAILED DESCRIPTION

[0030] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application 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 application and are not used to limit the present application.

[0031] The business data processing method provided in the embodiment of the present application can be applied to Figure 1In the application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data that the server 104 needs to process. The data storage system can be integrated on the server 104, or it can be placed on the cloud or other network servers. The user can configure the corresponding data analysis demand information for the target business data on the terminal 102. The target business data can be data from the business system. Different business systems can correspond to different business data. For example, in the financial management system, the target business data can include various financial data. The terminal 102 can send the data analysis demand information to the server 104, and the server 104 can perform demand analysis on the received data analysis demand information to obtain the analysis model description information. The server 104 builds a data analysis model for the target business data based on the analysis model description information and the preset analysis model template. After updating the target business data to the data analysis model, the data analysis is performed based on the updated data analysis model according to the preset data analysis method to obtain the data analysis results of the target business data. The server 104 can send the data analysis results to the terminal 102 for intuitive display at the terminal 102.

[0032] In some embodiments, the business data processing method can also be executed independently by a computer device such as the terminal 102 or the server 104. For example, after the terminal 102 obtains the data analysis requirement information, it can directly perform business data processing on the target business data based on the data analysis requirement information. Alternatively, the server 104 can directly obtain the data analysis requirement information from the data storage system and then perform business data processing on the target business data, and then store the obtained data analysis results in the data storage system.

[0033] The terminal 102 may be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, IoT devices, and portable wearable devices. The IoT devices may be smart speakers, smart TVs, smart air conditioners, smart car devices, projection devices, etc. Portable wearable devices may be smart watches, smart bracelets, head-mounted devices, etc. Head-mounted devices may be virtual reality (VR) devices, augmented reality (AR) devices, smart glasses, etc. The server 104 may be an independent physical server, a server cluster or distributed system consisting of multiple physical servers, or a cloud server that provides cloud computing services.

[0034] In an exemplary embodiment, Figure 2 As shown, a business data processing method is provided, which is applied to Figure 1 The server in the example is used to illustrate, including the following steps 202 to 210. Among them:

[0035] Step 202: Obtain data analysis requirement information for target business data.

[0036] Among them, the target business data is the business data that needs to be processed in the business system, and specifically may include business data obtained through different interfaces in the business system, such as the target business data may include financial data in the financial system, the target business data may also include shopping transaction data in the transaction system, and the target business data may also include traffic data in the traffic system. The target business data is the data of the business system to which it belongs, but it may include multiple sources. For example, in an ERP (Enterprise Resource Planning) system, the target business data may include various financial data imported from the ERP system, interface or table. The data analysis requirement information is used to describe the demand for data analysis of the target business data, such as can be configured by the user according to actual needs. Through the data analysis requirement information, various requirement information such as the analysis purpose and analysis method for the target business data can be determined.

[0037] Optionally, the server may obtain data analysis requirement information for the target business data, and the data analysis requirement information may be sent to the server by the user after configuring the target business data at the terminal. For example, the user may configure at least one of various data analysis requirement information including industry, analysis purpose, analysis object, and analysis method for the target business data at the terminal, and the analysis method may include at least one of quantitative analysis or qualitative analysis. In some embodiments, corresponding data analysis requirement templates may be pre-configured for different types of business data, and the data analysis requirement templates may include default data analysis requirement information for the corresponding type of business data. The user may configure the data analysis requirement template to obtain the required data analysis requirement information. For example, the user may configure the data analysis requirement template corresponding to the target business data based on the data analysis needs of the target business data to obtain the data analysis requirement information for the target business data, thereby simplifying the configuration processing of the data analysis requirement information and improving the processing efficiency of business data.

[0038] Step 204: perform demand analysis on the data analysis demand information to obtain analysis model description information for the target business data.

[0039] Among them, the analysis model description information is the description information for the data analysis model. Based on the analysis model description information, a corresponding data analysis model can be constructed. The data analysis model is a model framework for performing data analysis on target business data. The data analysis model can include different fields to carry different business data.

[0040] Exemplarily, for the acquired data analysis demand information, the server can perform demand analysis to determine the various needs for data analysis of the business data, and obtain the analysis model description information for the target business data. The analysis model description information may include natural language text information, so that the data analysis model to be constructed can be accurately defined and described in the form of natural language text, which is convenient for users to understand the meaning. For example, the server can perform demand analysis on the data analysis demand information based on artificial intelligence technology to obtain the analysis model description information in natural language form. For example, the analysis model description information may include descriptive information such as analysis indicators, influencing factors, analysis dimensions, and formula relationships.

[0041] Step 206: construct a data analysis model for the target business data based on the analysis model description information and the preset analysis model template corresponding to the target business data.

[0042] The analysis model template is a template used to build a data analysis model. The analysis model template is used as a model framework to configure and build a corresponding data analysis model according to data analysis requirements. The constructed data analysis model is used to perform data analysis on the target business data, and data analysis can be performed according to each field in the data analysis model.

[0043] Optionally, the server can obtain the preset analysis model template corresponding to the target business data. Different target business data can correspond to different analysis model templates. The server can query the corresponding preset analysis model template according to the business type of the target business data. Different fields can be included in the analysis model template, which can specifically include but are not limited to at least one of the dependent variable, independent variable, analysis dimension, and the formula relationship between the dependent variable and the independent variable. Among them, the independent variable supports user configuration, and specifically can add, delete, and modify branches, add upper and lower levels, etc. The analysis dimension can support filtering operations and combination operations. By updating each field variable in the analysis model template, a data analysis model for data analysis and processing can be constructed. The server can construct a corresponding data analysis model based on the analysis model description information and the analysis model template. For example, the server can configure and update the analysis model template according to the analysis model description information, thereby constructing a data analysis model.

[0044] Step 208: Update the target business data into the data analysis model to obtain an updated data analysis model.

[0045] Exemplarily, for the constructed data analysis model, the server can update it through the target business data, such as the server can add the target business data to the data analysis model to obtain an updated data analysis model. In some embodiments, the server can add the data of each field in the target business data to the corresponding field of the data analysis model respectively, thereby adding the target business data to the data analysis model to obtain an updated data analysis model. The updated data analysis model is constructed based on the analysis model description information and the analysis model template, which can reflect the data analysis requirements for the target business data. The updated data analysis model also carries the target business data itself, and the updated data analysis model can be used to perform data analysis processing on the target business data.

[0046] Step 210, performing data analysis based on the updated data analysis model in accordance with a preset data analysis method to obtain data analysis results for the target business data.

[0047] The data analysis method is a method of performing data analysis and processing on target business data. The data analysis method can be implemented by calling different data analysis algorithms. For example, for financial data, data analysis and processing can be performed by using at least one of the algorithms selected from chain substitution analysis, marginal contribution rate, sensitivity analysis, break-even analysis or regression analysis. The data analysis result is the result obtained by performing data analysis and processing on the target business data according to the data analysis method.

[0048] Optionally, the server can determine a preset data analysis method, and perform data analysis based on the updated data analysis model according to the data analysis method. For example, the server can construct a corresponding data analysis function according to a preset data analysis algorithm, and perform data analysis on the data carried in the updated data analysis model by calling the data analysis function to obtain data analysis results for the target business data.

[0049] In the above-mentioned business data processing method, a demand analysis is performed on the data analysis demand information of the target business data to obtain the analysis model description information, and a data analysis model for the target business data is constructed based on the analysis model description information and the preset analysis model template. After the target business data is updated into the data analysis model, data analysis is performed based on the updated data analysis model in accordance with the preset data analysis method to obtain the data analysis result of the target business data. In the process of business data processing, after a demand analysis is performed on the data analysis demand information, a data analysis model is constructed based on the analysis model description information and the analysis model template, and after the target business data is updated into the analysis data analysis model, data analysis is performed based on the preset data analysis method. The data analysis demand of the target business data can be fully explored through the data analysis model, and multi-dimensional data analysis of the target business data can be realized, thereby improving the accuracy of business data analysis.

[0050] In an exemplary embodiment, obtaining data analysis requirement information for target business data includes: in response to a configuration operation triggered on a data analysis requirement interaction interface, obtaining data analysis requirement information configured by the configuration operation for the target business data.

[0051] Among them, the data analysis demand interaction interface is an interface that supports users to configure data analysis needs. The data analysis demand interaction interface can include data analysis demand configuration items for various business data. Users can trigger corresponding configuration operations for each data analysis demand configuration item, thereby configuring data analysis demand information for the corresponding business data.

[0052] Exemplarily, data analysis requirement configuration items for target business data can be displayed in the data analysis requirement interaction interface. Users can trigger configuration operations in the data analysis requirement interaction interface. Configuration operations can act on data analysis requirement configuration items. The server can obtain data analysis requirement information configured by the configuration operation for the target business data.

[0053] Furthermore, a demand analysis is performed on the data analysis demand information to obtain analysis model description information for the target business data, including: generating structured text based on the data analysis demand information; inputting the structured text into a pre-trained demand analysis model to obtain analysis model description information for the target business data output by the demand analysis model.

[0054] The structured text may be a text with a fixed format, and the demand analysis model may be an artificial intelligence model built based on an artificial neural network. The demand analysis model may be used to perform demand analysis on the input text and output corresponding demand analysis results. For example, for the input structured text, the demand analysis model may output analysis model description information for the target business data. The demand analysis model may be constructed based on at least one of a convolutional neural network (CNN) algorithm, a recurrent neural network (RNN) algorithm, a long short-term memory network (LSTM) algorithm, a deep belief network (DBN) algorithm, a generative adversarial network (GAN) algorithm, and a Transformer algorithm.

[0055] Exemplarily, the server can perform structured conversion on the data analysis demand information to generate structured text, and the structured text can have a fixed text format, so that the data analysis demand information can be accurately divided into multiple dimensions. The server can call a pre-trained demand analysis model, such as calling a large language model (LLM) to perform demand analysis on the structured text, so that the large language model can output analysis model description information for the target business data.

[0056] In some exemplary embodiments, a user triggers a configuration operation in a data analysis requirement interaction interface to obtain data analysis requirement information for target business data configuration, generates structured text based on the data analysis requirement information, and performs a requirement analysis on the structured text through a pre-trained requirement analysis model to obtain analysis model description information. The analysis model description information can be accurately and efficiently generated through the requirement analysis model, which is conducive to improving the accuracy and processing efficiency of business data analysis.

[0057] In an exemplary embodiment, Figure 3 As shown, the process of constructing a data analysis model, that is, constructing a data analysis model for target business data based on the analysis model description information and the preset analysis model template corresponding to the target business data, includes steps 302 to 306. Among them:

[0058] Step 302: Obtain a preset analysis model template corresponding to the target business data.

[0059] Among them, different types of business data can preset corresponding analysis model templates, and the analysis model templates can include corresponding fields, such as dependent variables, independent variables, analysis dimensions, formula relationships between dependent variables and independent variables, and other fields.

[0060] Exemplarily, the server may query an analysis model template preset for the target business data. In some embodiments, the server may determine the business type corresponding to the target business data, and query a corresponding analysis model template according to the business type of the target business data.

[0061] Step 304: convert the analysis model description information into a structured query statement.

[0062] Among them, the structured query statement is a statement text used for database process operations. Optionally, the server can perform format conversion on the analysis model description information to convert the analysis model description information into a structured query statement. For example, the server can perform format conversion on the analysis model description information based on NL2SQL (Natural Language to Structured Query Language) technology to obtain the corresponding structured query statement. NL2SQL technology can convert the user's natural statement into an executable SQL (Structured Query Language) statement.

[0063] Step 306: Update the structured query statement into the analysis model template to obtain a data analysis model for the target business data.

[0064] Exemplarily, the server can update the analysis model template through a structured query statement. For example, the server can fill in the data of each field in the structured query statement into the analysis model template respectively, thereby updating the structured query statement into the analysis model template to obtain a data analysis model for the target business data.

[0065] In some exemplary embodiments, the server converts the analysis model description information into a structured query statement, and updates the structured query statement to a preset analysis model template corresponding to the target business data, thereby obtaining a data analysis model for the target business data. A multi-dimensional data analysis model can be constructed based on the analysis model description information, so that data analysis processing can be performed through the data analysis model, which is conducive to improving the accuracy of business data analysis.

[0066] In an exemplary embodiment, the target business data is updated into the data analysis model to obtain an updated data analysis model, including: acquiring the target business data from the business system; mapping and matching the fields of the target business data with the fields in the data analysis model to determine matching field pairs; and adding the target business data to the data analysis model according to the matching field pairs to obtain an updated data analysis model.

[0067] Among them, the target business data includes different fields, each field can have corresponding data, that is, the target business data can include the data corresponding to each field. The data analysis model also includes different fields, and different fields can be used to carry different data. The matching field pair is a field pair obtained by combining the field of the target business data with the field in the data analysis model. The matching field pair may include a first field from the target business data and a second field from the data analysis model, wherein the first field and the second field match, such as the first field and the second field have the same meaning or have a subordinate relationship.

[0068] Exemplarily, the server can determine the business system to which the target business data belongs, and obtain the target business data that needs to be analyzed from the business system. The server can map and match the fields of the target business data with the fields in the data analysis model, and obtain matching field pairs based on the matching field combinations. The server can add the target business data to the data analysis model according to the matching field pairs to update the data analysis model and obtain an updated data analysis model. For example, for a matching field pair A, which includes field A1 from the target business data and field A2 from the data analysis model, when updating the data analysis model based on the matching field pair, the server can add data a1 belonging to field A1 in the target business data to field A2 of the data analysis model, and after traversing each matching field pair and updating the data analysis model through each matching field pair, an updated data analysis model is obtained.

[0069] In some exemplary embodiments, the server maps and matches the fields of the target business data with the fields in the data analysis model, and adds the target business data to the data analysis model according to the constructed matching field pairs to obtain an updated data analysis model, so that the target business data can be accurately added to each dimension of the data analysis model, so as to perform data analysis processing through the data analysis model, which is conducive to improving the accuracy of business data analysis.

[0070] In an exemplary embodiment, the business data processing method also includes: when a report generation request for data analysis results is obtained, a report requirement analysis is performed based on the report generation request and the knowledge base data of the business system to which the target business data belongs to obtain report requirement information; and a data analysis report corresponding to the data analysis results is generated based on the report requirement information, knowledge base data and data analysis results through a large language model.

[0071] Among them, the report generation request is used to request the generation of a data analysis report for the data analysis results. The report generation request can be triggered by the user according to actual needs. Knowledge base data is the data in the business system knowledge base (Knowledge Base). The knowledge base is a system or platform that centrally stores, organizes and manages various knowledge in the business system. It includes various types of content, such as documents, articles, FAQs, technical manuals, guides, tutorials, data sets, research reports, historical data analysis results, etc. Report requirement information is used to describe the need to generate a data analysis report for the data analysis results. According to the report requirement information, the required data analysis report can be generated based on the data analysis results. The large language model is an artificial intelligence model trained with a large amount of data, designed to understand and generate natural language text. The large language model is built based on deep learning technology and can capture the complexity and diversity of language. For example, it can process the input data and output the corresponding data analysis report.

[0072] Exemplarily, when the server receives a report generation request for the data analysis result, the server may determine the business system to which the target business data belongs, and obtain the corresponding knowledge base data from the knowledge base of the business system. The knowledge base data may include historical data analysis results and historical data analysis reports. The historical data analysis results may include data analysis results corresponding to business data belonging to the same business type as the target business data. The historical data analysis report is a data analysis report generated for the historical data analysis results. The server may perform a report demand analysis based on the report generation request and the knowledge base data, such as by calling a large language model to perform a report demand analysis and obtain report demand information for the data analysis result. The server may call a pre-trained large language model to generate a data analysis report corresponding to the data analysis result according to the report demand information, the knowledge base data and the data analysis result. For example, the server may use the report demand information, the knowledge base data and the data analysis result as inputs of the large language model, respectively, and the large language model outputs a corresponding data analysis report for the input report demand information, the knowledge base data and the data analysis result. The data analysis report may include, but is not limited to, various formats including word, PPT (PowerPoint, slides).

[0073] In some exemplary embodiments, the server can perform a report requirement analysis based on a report generation request and knowledge base data, and generate a data analysis report corresponding to the data analysis results according to the report requirement information, knowledge base data, and data analysis results through a large language model. The data in the knowledge base can be integrated to fully tap the understanding and retrieval capabilities of the large language model, thereby improving the reliability of the generated data analysis report.

[0074] In an exemplary embodiment, a data analysis report corresponding to the data analysis result is generated through a large language model according to report requirement information, knowledge base data and data analysis results, including: generating a data analysis report template through a large language model according to report requirement information, knowledge base data and data analysis results; generating analysis report data through a large language model according to the data analysis report template, knowledge base data and data analysis results; and obtaining a data analysis report corresponding to the data analysis result according to the analysis report data.

[0075] The data analysis report template is a reference template for the data analysis report. The data analysis report template may specifically include an outline of the data analysis report. By configuring the data analysis report template, the corresponding data analysis report can be obtained. The analysis report data is metadata for generating the data analysis report, that is, after the analysis report data is laid out, typesetted and rendered, the required data analysis report can be generated.

[0076] Optionally, the server can generate a data analysis report template by calling a large language model, such as the server can input report requirement information, knowledge base data and data analysis results into the large language model for template generation, and the large language model outputs the generated data analysis report template. The server can generate analysis report data based on the data analysis report template, knowledge base data and data analysis results. For example, the server can input the data analysis report template, knowledge base data and data analysis results into the large language model for metadata analysis, and the large language model outputs the generated analysis report data. The server can obtain a data analysis report corresponding to the data analysis result based on the analysis report data. For example, the server can perform layout, typesetting and other processing based on the analysis report data to obtain a data analysis report corresponding to the data analysis result. In some embodiments, the server can also obtain a data analysis report corresponding to the data analysis result according to the analysis report data through a large language model, such as the server can input the analysis report data into the large language model, and the large language model performs layout, typesetting and other processing, and outputs a data analysis report that meets the preset format requirements, such as outputting a data analysis report in PPT format.

[0077] In some exemplary embodiments, the server generates a data analysis report template based on report requirement information, knowledge base data and data analysis results through a large language model, and generates analysis report data based on the data analysis report template, knowledge base data and data analysis results, and then obtains a data analysis report based on the analysis report data. The data of the knowledge base can be integrated to fully tap the understanding and retrieval capabilities of the large language model, thereby improving the reliability of the generated data analysis report.

[0078] The present application also provides an application scenario, and the application scenario applies the above-mentioned business data processing method. Specifically, the application of the business data processing method in the application scenario is as follows:

[0079] This application scenario is a financial system scenario, and the target business data that needs to be processed is financial data. Specifically, the existing solution only supports passing financial data and prompt words to the AI ​​(Artificial Intelligence) model, and using the AI ​​model to directly analyze the financial data and draw conclusions. Therefore, the financial data source in traditional technology is single, resulting in poor data analysis accuracy; and the lack of LLM layer pre-data preprocessing leads to unstable analysis results and easy hallucinations; in addition, traditional technology cannot integrate traditional algorithms for data processing, and at the current stage, AI capabilities cannot meet the complex requirements of financial analysis for data calculation.

[0080] Based on this, this application scenario applies the above-mentioned business data processing method. In the data processing stage, it supports the introduction of data imported from the ERP (Enterprise Resource Planning) system, interface or Excel and stores it as the data source layer; after cleaning and sorting, it is stored as the fact data layer; after selecting the dimension field in the fact data layer, the dimension field is used as the query constraint condition and stored together with other data as the dimension data layer. The dimension data in the dimension data layer is used as the target business data that needs to be analyzed.

[0081] Furthermore, in the model processing stage, the first step is that the user configures the data analysis demand information. Specifically, the user configures the financial analysis model name, industry, analysis purpose, analysis object, etc. according to the required fields of the "AI Model Outline" interface, and selects "decomposition by formula" or "decomposition by business factors" (corresponding to quantitative analysis and qualitative analysis in the analysis method, respectively). The second step is to conduct a demand analysis on the data analysis demand information configured by the user to obtain the analysis model description information. Specifically, a structured text is generated according to the content filled in by the user, and is passed into the AI ​​model as a prompt, requiring the AI ​​model to output a natural language of the financial analysis model containing analysis indicators, formula relationships, influencing factors (layered decomposition into Markdown format), and analysis dimensions. Among them, Prompt refers to a given text or sentence, which is used to start and guide the large model to generate output of a specific type, theme or format; Markdown format is a lightweight markup language (MarkupLanguage). Step 3: Based on the analysis model description information and the preset analysis model template corresponding to the target business data, a data analysis model for the target business data is constructed. Specifically, the natural language output by the AI ​​model is parsed using NL2SQL technology, and filled into the "financial analysis model" framework, that is, filled into the analysis model template, thereby obtaining a data analysis model for the target business data, specifically a financial analysis model. The financial analysis model is an analysis tool that uses quantitative and qualitative methods to evaluate a company's financial status, operational efficiency, and profitability.

[0082] The corresponding relationship between each content and field can be shown in Table 1 below.

[0083] Table 1

[0084]

[0085] Among them, independent variables support setting up branches for addition, deletion and modification, support adding superiors and subordinates, support setting up labels such as income, fixed costs, variable costs, and profits; analysis dimensions refer to the angles or aspects from which the financial status of an enterprise is evaluated when conducting financial analysis, and they support filtering (such as filtering by product category) and combinations (such as region + product category).

[0086] Step 4: Update the target business data to the data analysis model to obtain the updated data analysis model. Specifically, map and match the variables and dimension fields in the filled-in "Financial Analysis Model" with the financial data in the dimension data layer to obtain the updated financial analysis model. Among them, mapping matching supports automatic matching based on AI, that is, obtaining the financial analysis model fields and dimension data table fields, passing them into the AI ​​model, outputting the matching results, generating a one-to-one or one-to-many matching relationship, and writing it back to the financial analysis model. In addition to automatic matching of the AI ​​model, the following auxiliary mapping paths are also supported: select data table fields to match fields in the financial analysis model, support setting expressions; select ERP system fields to match fields in the financial analysis model, support setting expressions; select variable fields of other models.

[0087] In a specific application, take the inventory turnover analysis of a mobile phone manufacturing company as an example. Figure 4 As shown, users can fill in the "AI model outline" and pass it into the AI ​​model. Specifically, you can enter the content of "Model name: Mobile phone manufacturing company inventory turnover analysis model Analysis purpose: locate the cause of abnormal inventory turnover Analysis object: inventory turnover industry: 3C manufacturing industry" into the AI ​​model, and the AI ​​model outputs natural language, which can be used to describe the financial analysis model. The AI ​​model can fill the output content into the "Financial Analysis Model" and map the fields in the "Financial Analysis Model" with the fields of the dimension data table to obtain an updated financial analysis model.

[0088] In the data attribution analysis phase, you can preset common financial analysis functions in the business system. Low-order functions include chain substitution analysis, marginal contribution rate, sensitivity analysis, and break-even analysis; advanced functions include regression analysis. When performing data analysis, first select the financial analysis model, then select the independent variable, dependent variable, and analysis dimension of this analysis, and finally select the financial analysis function (if the function is empty, no data processing is performed, and only the variables are counted and listed according to the analysis dimension). The algorithm processes and generates the analysis results in the form of a table, thereby obtaining the data analysis results for the financial data.

[0089] In the financial report generation stage, we use RAG (Retrieval-Augmented Generation) technology + LLM's understanding ability, combined with knowledge base knowledge recall to realize NL (Natural Language) to financial report templates to financial analysis reports readable by financial personnel, and support word / PPT format output. Specifically, Figure 5As shown, financial personnel can initiate a report generation request, and the AI ​​model acts as an agent (AI-Agent) to perform intent recognition and analysis scope recognition and analysis in turn to find the corresponding knowledge base data from the server of the business system. The AI ​​model can generate a financial report outline based on the AI ​​output RAG (historical report generation outline and system configuration outline). After the user confirms the outline, the AI ​​model reads the financial data from the business system to generate the report and renders the result into a data analysis report in ppt / word format.

[0090] The financial report generation process can include 4 stages:

[0091] 1. NL (Natural Language) input and recognition: The user inputs NL and combines it with the knowledge base data of the ERP system to realize the recall of demands for basic data filtering, so as to realize the summoning of different basic demands in different enterprise scenarios and different business scenarios. This part of demands will be used as input for the next stage, that is, as report demand information. The sample data is as follows:

[0092] {

[0093] "Request": "Generate an annual financial analysis report for xxx enterprise",

[0094] "Enterprise Stage":"Listing","Year":2023,

[0095] "Period":"January 2023 to December 2023",

[0096] "Analysis scope": "Three major report status, xxxxx"

[0097] }

[0098] 2. LLM generates financial report outline: The report requirement information recalled by NL is submitted to the big model again, combined with the prompt word engineering + knowledge base knowledge, to generate the specified financial report template outline. Among them, the financial report template is a data format generated by LLM and can be understood by LLM. Based on this format, financial data analysis and interpretation are performed and visual word / ppt is generated. The sample data is as follows:

[0099] {

[0100] 1. Title 1, xxxx

[0101] 1.1 Title 2xxx

[0102] 1.2 Title 2xxx

[0103] 1.3 Title Three xxxx……

[0104] }

[0105] In a specific application, such as Figure 6 As shown in the figure, it is the outline of the financial report generated for Group A, which may include items such as asset and liability overview, current account and current liability analysis.

[0106] 3. Intelligent generation of report content and layout based on outline: The outline generated in the second step is combined with the large model and ERP knowledge base to generate the financial report content, and LLM realizes the layout and layout of the content in PPT / Word.

[0107] 4. Realize the rendering of data to Word / PPT: Use software code to realize the specific rendering function of data to Word / PPT completed in the third step.

[0108] The business data processing method provided in this application guides users to input fields such as the name of the financial analysis model, industry, analysis purpose, and analysis object, and generates a specific, clear, and AI-understandable structured text. Furthermore, the financial analysis model is abstracted into dependent variables, independent variables, and analysis dimensions, and the relationship between each segmentation variable is clearly displayed in a tree structure; the NL2SQL technology is used to fill in the field names of each variable in the financial analysis model. In addition, the financial analysis function is preset, and each field in the financial analysis model is used as a function variable to directly participate in data calculation. In addition, LLM is used to implement the transition from natural language to template to word / ppt, and the problem of more illusions in pure LLM financial analysis is solved through knowledge base training + RAG technology. Finally, the business data processing method provided in this application supports the direct introduction of data imported from the ERP system, interface, and EXCEL and storage as a system data table.

[0109] The business data processing method provided in this application fully taps into the understanding and retrieval capabilities of AI through the setting of the AI ​​model outline, and realizes the extension of the analysis logic from financial variables to business variables. Secondly, by establishing a financial analysis model, it is ensured that the analysis ideas are clear and visible, and can be stably reused. Moreover, it supports data from multiple sources, which can be smoothly used as analysis data for calculation after cleaning and processing. In addition, it supports multi-layer nesting of model variables: the dependent variable of model 1 can be used as the independent variable of model 2, the lower-level model is selected for analysis, and the upper-level model is automatically parsed. Finally, by presetting commonly used financial analysis functions, compared with traditional analysis methods, it is free from setting complex formulas row by row and column by column, and at the same time, code processing is used to make up for the weaknesses of high language expression cost of the relationship between data and poor AI computing power.

[0110] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. 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 executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0111] Based on the same inventive concept, the embodiment of the present application also provides a business data processing device for implementing the business data processing method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in the one or more business data processing device embodiments provided below can refer to the limitations on the business data processing method above, and will not be repeated here.

[0112] In an exemplary embodiment, Figure 7 As shown, a business data processing device 700 is provided, including: an analysis requirement information acquisition module 702, a requirement analysis processing module 704, a data analysis model construction module 706, a data analysis model update module 708 and a data analysis processing module 710, wherein:

[0113] An analysis requirement information acquisition module 702 is used to acquire data analysis requirement information for target business data;

[0114] The demand analysis processing module 704 is used to perform demand analysis on the data analysis demand information to obtain analysis model description information for the target business data;

[0115] A data analysis model building module 706 is used to build a data analysis model for the target business data based on the analysis model description information and a preset analysis model template corresponding to the target business data;

[0116] A data analysis model updating module 708 is used to update the target business data into the data analysis model to obtain an updated data analysis model;

[0117] The data analysis processing module 710 is used to perform data analysis based on the updated data analysis model in accordance with a preset data analysis method to obtain data analysis results for the target business data.

[0118] The analysis requirement information acquisition module 702 is also used to respond to the configuration operation triggered in the data analysis requirement interaction interface, and obtain the data analysis requirement information configured by the configuration operation for the target business data; the requirement analysis processing module 704 is also used to generate structured text based on the data analysis requirement information; input the structured text into the pre-trained requirement analysis model to obtain the analysis model description information for the target business data output by the requirement analysis model.

[0119] The data analysis model construction module 706 is also used to obtain a preset analysis model template corresponding to the target business data; convert the analysis model description information into a structured query statement; update the structured query statement into the analysis model template to obtain a data analysis model for the target business data.

[0120] The data analysis model update module 708 is also used to obtain the target business data from the business system; map and match the fields of the target business data with the fields in the data analysis model to determine matching field pairs; and add the target business data to the data analysis model according to the matching field pairs to obtain an updated data analysis model.

[0121] The business data processing device 700 also includes an analysis report generation module, which is used to, when a report generation request for the data analysis result is obtained, perform a report requirement analysis based on the report generation request and the knowledge base data of the business system to which the target business data belongs, and obtain report requirement information; and generate a data analysis report corresponding to the data analysis result based on the report requirement information, the knowledge base data and the data analysis result through a large language model.

[0122] The analysis report generation module is also used to generate a data analysis report template according to the report requirement information, the knowledge base data and the data analysis results through a large language model; generate analysis report data according to the data analysis report template, the knowledge base data and the data analysis results through the large language model; and obtain a data analysis report corresponding to the data analysis results based on the analysis report data.

[0123] Each module in the above-mentioned business data processing device can be implemented in whole or in part by software, hardware or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute the operations corresponding to each module above.

[0124] In an exemplary embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as shown in FIG. Figure 8As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store various data involved in the business data processing method. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a business data processing method is implemented.

[0125] In an exemplary embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as shown in FIG. Fig. 9 As shown. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a mobile cellular network, near field communication (NFC) or other technologies. When the computer program is executed by the processor, a business data processing method is implemented. The display unit of the computer device is used to form a visually visible picture, which can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device shell, or an external keyboard, touchpad or mouse.

[0126] Those skilled in the art will understand that Figure 8 and Fig. 9The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0127] In one embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above method embodiments when executing the computer program.

[0128] In one embodiment, a computer-readable storage medium is provided, storing a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0129] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0130] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0131] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0132] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above 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 application.

[0133] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A business data processing method, characterized in that: The method comprises: Obtain data analysis requirements information for target business data; Performing demand analysis on the data analysis demand information to obtain analysis model description information on the target business data; Building a data analysis model for the target business data based on the analysis model description information and a preset analysis model template corresponding to the target business data; Updating the target business data into the data analysis model to obtain an updated data analysis model; Data analysis is performed based on the updated data analysis model in accordance with a preset data analysis method to obtain data analysis results for the target business data.

2. The method according to claim 1, characterized in that The obtaining of data analysis requirement information for target business data includes: In response to a configuration operation triggered on a data analysis requirement interaction interface, obtaining data analysis requirement information configured by the configuration operation for target business data; The performing demand analysis on the data analysis demand information to obtain analysis model description information for the target business data includes: Generate structured text based on the data analysis requirement information; The structured text is input into a pre-trained demand analysis model to obtain analysis model description information output by the demand analysis model for the target business data.

3. The method according to claim 1, characterized in that The step of constructing a data analysis model for the target business data based on the analysis model description information and a preset analysis model template corresponding to the target business data includes: Obtaining a preset analysis model template corresponding to the target business data; Converting the analysis model description information into a structured query statement; The structured query statement is updated into the analysis model template to obtain a data analysis model for the target business data.

4. The method according to claim 1, characterized in that: The step of updating the target business data into the data analysis model to obtain an updated data analysis model includes: Acquire the target business data from the business system; Mapping and matching the fields of the target business data with the fields in the data analysis model to determine matching field pairs; According to the matching field pairs, the target business data is added to the data analysis model to obtain an updated data analysis model.

5. The method according to any one of claims 1 to 4, characterized in that: The method further comprises: When a report generation request for the data analysis result is obtained, a report requirement analysis is performed based on the report generation request and the knowledge base data of the business system to which the target business data belongs, to obtain report requirement information; A data analysis report corresponding to the data analysis result is generated according to the report requirement information, the knowledge base data and the data analysis result through a large language model.

6. The method according to claim 5, characterized in that The generating, by means of the large language model, a data analysis report corresponding to the data analysis result according to the report requirement information, the knowledge base data and the data analysis result, comprises: Generate a data analysis report template according to the report requirement information, the knowledge base data and the data analysis results through a large language model; Generate analysis report data according to the data analysis report template, the knowledge base data and the data analysis results through the large language model; A data analysis report corresponding to the data analysis result is obtained according to the analysis report data.

7. A business data processing device, characterized in that: The device comprises: An analysis requirement information acquisition module is used to acquire data analysis requirement information for target business data; A demand analysis processing module, used to perform demand analysis on the data analysis demand information and obtain analysis model description information for the target business data; A data analysis model building module, used to build a data analysis model for the target business data based on the analysis model description information and a preset analysis model template corresponding to the target business data; A data analysis model updating module, used to update the target business data into the data analysis model to obtain an updated data analysis model; The data analysis processing module is used to perform data analysis based on the updated data analysis model in accordance with a preset data analysis method to obtain data analysis results for the target business data.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images