Enterprise intelligent analysis response method and device based on large model and computer equipment

Abstract

The invention relates to an enterprise intelligent analysis response method and device based on a large model, computer equipment, a computer readable storage medium and a computer program product. The method comprises the steps of receiving data processing demand information input by a user through a natural language in response to an interactive operation of the user on a large model; analyzing the data processing demand information through a large model, and extracting to obtain a target variable and a target parameter; the target variable is used for describing data attribute information, and the target parameter is used for describing data range information and processing mode information; according to the target variable and the target parameter, calling a pre-packaged ant script to obtain a data processing result corresponding to the data processing demand information; and displaying a data processing result through the large model. By adopting the method, the enterprise data analysis efficiency can be improved.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a method, apparatus, computer device, computer-readable storage medium and computer program product for enterprise intelligent analysis and response based on a large model. Background Art

[0002] With the continuous improvement of enterprise informatization, enterprises have accumulated a large amount of business data, which plays an important role in supporting business decisions and improving operational efficiency.

[0003] However, traditional business intelligence (BI) analysis systems require professional technical teams to manually write query and analysis scripts, which means that users must have a certain technical background to use these systems, resulting in poor efficiency in the data analysis process. Summary of the invention

[0004] Based on this, it is necessary to provide a large-model-based enterprise intelligent analysis response method, device, computer equipment, computer-readable storage medium and computer program product that can improve the efficiency of enterprise data analysis in response to the above-mentioned technical problems.

[0005] In a first aspect, the present application provides an enterprise intelligent analysis and response method based on a large model, comprising:

[0006] In response to the user's interactive operation on the large model, receiving data processing requirement information input by the user through natural language;

[0007] The data processing requirement information is parsed through a large model to extract target variables and target parameters; the target variables are used to describe data attribute information, and the target parameters are used to describe data range information and processing method information;

[0008] According to the target variable and the target parameter, a pre-packaged ant script is called to obtain a data processing result corresponding to the data processing requirement information;

[0009] The data processing results are displayed through the large model.

[0010] In one of the embodiments, before responding to the user's interactive operation on the large model, the method further includes:

[0011] Obtain the initial variables and initial parameters set by the user for preset analysis requirements;

[0012] The initial variables and initial parameters are expanded respectively through the large model to generate related variables and related parameters;

[0013] Obtaining a preset parameter pool according to the relevant variables, the relevant parameters, the initial variables and the initial parameters;

[0014] The step of parsing the data processing requirement information through a large model and extracting target variables and target parameters includes:

[0015] According to the preset parameter pool, target variables and target parameters are extracted from the data processing requirement information.

[0016] In one embodiment, the step of calling a prepackaged ant script according to the target variable and the target parameter to obtain a data processing result corresponding to the data processing requirement information includes:

[0017] Determine the corresponding data source according to the target variable, and determine the data range according to the target parameter;

[0018] Acquire the data to be processed within the data range from the data source;

[0019] The data to be processed is processed using a data processing method corresponding to the target parameter to obtain a data processing result corresponding to the data processing requirement information.

[0020] In one of the embodiments, before responding to the user's interactive operation on the large model, the method further includes:

[0021] Analyze the data in the enterprise database to obtain the field information of the enterprise database;

[0022] Determining the corresponding data source according to the target variable includes:

[0023] The data source is determined from the enterprise database according to the target variable and the field information.

[0024] In one embodiment, in response to the user's interactive operation on the large model, receiving the data processing requirement information input by the user through natural language includes:

[0025] In response to the user's interactive operation on the large model, generating guidance content through the large model;

[0026] The guidance content is displayed so that the user can input data processing requirement information including the target variable and the target parameter through natural language.

[0027] In one embodiment, displaying the data processing result through the large model includes:

[0028] Determine a corresponding visualization form according to the data processing requirement information;

[0029] According to the visualization form, converting the data processing result into corresponding visualization display data;

[0030] The large model is used to display visual display data corresponding to the data processing results.

[0031] In a second aspect, the present application also provides an enterprise intelligent analysis and response device based on a large model, comprising:

[0032] An information receiving module, used for receiving data processing requirement information input by a user through natural language in response to the user's interactive operation on the large model;

[0033] An information extraction module is used to parse the data processing requirement information through a large model to extract target variables and target parameters; the target variables are used to describe data attribute information, and the target parameters are used to describe data range information and processing method information;

[0034] A data processing module, used to call a pre-packaged ant script according to the target variable and the target parameter to obtain a data processing result corresponding to the data processing requirement information;

[0035] The result display module is used to display the data processing results through the large model.

[0036] 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 when the processor executes the computer program, the following steps are implemented:

[0037] In response to the user's interactive operation on the large model, receiving data processing requirement information input by the user through natural language;

[0038] The data processing requirement information is parsed through a large model to extract target variables and target parameters; the target variables are used to describe data attribute information, and the target parameters are used to describe data range information and processing method information;

[0039] According to the target variable and the target parameter, a pre-packaged ant script is called to obtain a data processing result corresponding to the data processing requirement information;

[0040] The data processing results are displayed through the large model.

[0041] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the following steps are implemented:

[0042] In response to the user's interactive operation on the large model, receiving data processing requirement information input by the user through natural language;

[0043] The data processing requirement information is parsed through a large model to extract target variables and target parameters; the target variables are used to describe data attribute information, and the target parameters are used to describe data range information and processing method information;

[0044] According to the target variable and the target parameter, a pre-packaged ant script is called to obtain a data processing result corresponding to the data processing requirement information;

[0045] The data processing results are displayed through the large model.

[0046] In a fifth aspect, the present application further provides a computer program product, including a computer program, which implements the following steps when executed by a processor:

[0047] In response to the user's interactive operation on the large model, receiving data processing requirement information input by the user through natural language;

[0048] The data processing requirement information is parsed through a large model to extract target variables and target parameters; the target variables are used to describe data attribute information, and the target parameters are used to describe data range information and processing method information;

[0049] According to the target variable and the target parameter, a pre-packaged ant script is called to obtain a data processing result corresponding to the data processing requirement information;

[0050] The data processing results are displayed through the large model.

[0051] The above-mentioned enterprise intelligent analysis response method, device, computer equipment, computer-readable storage medium and computer program product based on a big model firstly respond to the user's interactive operation on the big model, receive the data processing demand information input by the user through natural language, and eliminate the need for professional skills by receiving the user's natural language input, making the data processing process more user-friendly and easy to use, lowering the user's technical threshold and improving the accessibility of data analysis; then, the data processing demand information is parsed through the big model to extract the target variables and target parameters, wherein the target variables are used to describe the data attribute information, and the target parameters are used to describe the data range information and the processing method information, and the intelligent parsing ability of the big model is used to extract the target variables and target parameters from the user's natural language The key elements of data analysis (target variables and target parameters) are automatically extracted from the requirements, so that data query and analysis tasks are more accurate and targeted, and the conversion efficiency from user requirements to analysis operations is improved; then, according to the target variables and target parameters, the pre-packaged ant script is called to obtain the data processing results corresponding to the data processing requirement information. By automatically calling the pre-packaged ant script, the automatic query and processing of data is realized, which greatly reduces the complexity of manual intervention and script writing, and improves the efficiency and accuracy of data processing; finally, the data processing results are displayed through the big model, and the data processing results are presented to the user in the form of charts and reports by using the display ability of the big model, making the results more intuitive and readable. In the above method, through natural language interaction, big model parsing, automated task processing, etc., the complex data analysis process is simplified into a user-friendly interactive form, and professional technicians are no longer required to manually write query scripts, which not only reduces the threshold for use, but also significantly improves the efficiency and accuracy of data analysis. At the same time, the time from data request to result feedback is greatly shortened, supporting enterprises to quickly respond to business needs and improve the scientificity and timeliness of business management decisions. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] 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.

[0053] FIG1 is a schematic diagram of a process of a method for enterprise intelligent analysis and response based on a large model in one embodiment;

[0054] FIG2 is a schematic flow chart of steps for obtaining a preset parameter pool in one embodiment;

[0055] FIG3 is a schematic diagram of the structure of an enterprise intelligent analysis and response system based on a large model in another embodiment;

[0056] FIG4 is a schematic diagram of a flow chart of a method for executing an enterprise intelligent analysis response based on a big model by a big model-based enterprise intelligent analysis response system in one embodiment;

[0057] FIG5 is a structural block diagram of a large model-based enterprise intelligent analysis and response device in one embodiment;

[0058] FIG. 6 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0059] 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.

[0060] In one embodiment, as shown in FIG1 , a method for enterprise intelligent analysis and response based on a large model is provided. This embodiment uses the method applied to a terminal as an example for illustration. It is understandable that the method can also be applied to a server, and can also be applied to a system including a terminal and a server, and is implemented through the interaction between the terminal and the server. Among them, the terminal can be, but is not limited to, various personal computers, laptops, smart phones, and tablet computers. The server can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides cloud computing services. In this embodiment, the method includes the following steps:

[0061] Step S101, in response to the user's interactive operation on the large model, receiving the data processing requirement information input by the user through natural language.

[0062] Among them, the user's interactive operation on the big model refers to the user's natural language interaction with the big model through terminal devices (such as mobile phones, computers, etc.) to make specific data processing requests.

[0063] Among them, data processing demand information refers to the specific requirements for data analysis expressed by users through natural language when interacting with the big model, including but not limited to analysis objectives, data scope, expected processing methods, etc.

[0064] Exemplarily, the terminal displays an interactive window through an interface, in which the user can input his data processing requirements in natural language. The terminal first sends the natural language request input by the user to the big model through the interface, and the big model parses the input content and identifies the data analysis requirements therein. Based on natural language processing technology, the big model extracts key information from the user's requirements, including the business type and time range to be analyzed, and outputs it in a standardized format for subsequent processing.

[0065] Step S102, parsing data processing requirement information through a large model to extract target variables and target parameters.

[0066] Among them, the target variable is used to describe the data attribute information, and the target parameter is used to describe the data range information and processing method information.

[0067] Among them, the target variable is used to describe the attribute information of the data, which refers to the object or feature that the user wants to analyze, such as "loan balance", "sales", "bank branches", etc. This information is used to clarify the specific content of the data to be queried; the target parameter is used to describe the range information and processing method information of the data, such as the time range (such as "the past 12 months"), statistical method (such as "average" or "maximum"), etc., to help limit the scope of the data and the analysis method.

[0068] For example, after receiving the data processing demand information input by the user, the terminal passes it to the big model for further analysis. The big model uses natural language processing and machine learning technology to analyze the sentences input by the user and extract the key elements in the demand. For example, if the user enters "analyze the changes in the loan balance of a branch in the past 12 months", the big model will identify "loan balance" as the target variable and "past 12 months" as the time range, and then generate the target parameters.

[0069] In the specific implementation process, the big model first performs word segmentation on the data processing requirement information input by the user, and identifies different types of words in the sentence, such as nouns, time words, and geographic words. Then, based on the built-in business semantic model, the big model determines which words belong to the target variable and which belong to the target parameter. The big model converts these parsing results into standardized target variable and target parameter structures to facilitate the subsequent call of the ant script.

[0070] Step S103, calling the pre-packaged ant script according to the target variable and the target parameter to obtain the data processing result corresponding to the data processing requirement information.

[0071] Among them, the pre-packaged Ant script refers to an automated script written and packaged by developers during the system deployment phase to perform specific data processing tasks. Ant scripts usually include access to the database, execution of data processing logic, and result generation, and are the core components for achieving system automation.

[0072] Exemplarily, the terminal first determines the appropriate ant script according to the target variable and target parameter. For example, the target variable is "loan balance", the target parameters are "past 12 months" and "a bank branch", and the terminal will select the ant script that matches these conditions. The Ant script encapsulates the detailed steps for executing specific tasks, such as data extraction, data cleaning, aggregate calculation and report generation, etc. These steps are executed in a specific order to ensure that the data processing results meet user requirements. Next, the terminal executes the specific task by calling the pre-encapsulated Ant script. As the control center of the automated task, the Ant script can execute a series of tasks in the order defined in the script. In this process, the Ant script calls other encapsulated subtasks through the AntCall mechanism. For example, the main Ant script is responsible for the overall process control, and when encountering a scenario where data needs to be extracted from different data sources, AntCall is used to call different subtask scripts to obtain relevant data from the enterprise database. In the process of data processing, the Ant script will gradually call multiple modular processing tasks according to the task requirements. Each processing task is a subtask, such as extracting data from the database, cleaning the data, calculating statistical results, etc. The Ant script is responsible for the scheduling and control of these subtasks. Through the call of AntCall, it ensures the sequential execution and data flow of each task. Therefore, the entire process from the original state of the data to the final processing result can be completed automatically without human intervention. Finally, after the Ant script is executed, the result data corresponding to the data processing requirement information will be obtained. For example, for the requirement of "sales statistics by quarter", the Ant script will generate a result data containing the summary of sales for each quarter, and hand it over to the subsequent module of the system for display to the user. The entire task processing process is highly automated and modularized, which can adapt to different business needs and be flexibly expanded.

[0073] The ant script also includes specific data processing methods for different target parameters, such as performing time series analysis on time range parameters, or generating corresponding statistical charts for specific types of data. The entire ant script is designed to be highly modular, ensuring that relevant operations can be performed efficiently and accurate data results can be generated under different requirements.

[0074] Step S104, displaying the data processing results through a large model.

[0075] For example, after receiving the data processing results returned by the ant script, the terminal further processes them through the big model to ensure that the data results are presented in the best form that meets the user's needs. Specifically, the big model is responsible for selecting the appropriate display method, such as selecting different visual charts such as bar charts, line charts, pie charts, etc. according to the data type and user needs, or explaining the results in the form of a text report.

[0076] During the implementation process, the big model first determines the appropriate display method based on the user's needs and the characteristics of the data results. For time series data, the big model may choose a line chart for display; for classified statistical data, it may choose a pie chart or a bar chart. Then, the terminal calls the visualization component in the system to generate the corresponding chart or report according to the display method selected by the big model. The big model is also responsible for extracting the key points of the data results and explaining the meaning of these data in the form of natural language for users to understand.

[0077] The data display is presented to the user through the terminal's user interface. The user can see the chart of the data analysis results and the corresponding descriptive text on the interface, which helps the user to more intuitively understand the meaning and trend behind the data. The terminal can also provide interactive functions, such as users can zoom in and out of the chart, select different time periods for detailed analysis, etc., to further enhance the user's understanding and grasp of the data results.

[0078] In the above-mentioned enterprise intelligent analysis response method based on big model, first, in response to the user's interactive operation on the big model, the data processing demand information input by the user through natural language is received. By receiving the user's natural language input, the need for professional skills is eliminated, making the data processing process more user-friendly and easy to use, lowering the user's technical threshold and improving the accessibility of data analysis; then, the data processing demand information is parsed through the big model to extract the target variables and target parameters, wherein the target variables are used to describe the data attribute information, and the target parameters are used to describe the data range information and the processing method information. The intelligent parsing ability of the big model is used to automatically extract the data analysis from the user's natural language requirements. Key elements (target variables and target parameters) are used to make data query and analysis tasks more accurate and targeted, and improve the conversion efficiency from user needs to analysis operations; then, according to the target variables and target parameters, the pre-packaged ant script is called to obtain the data processing results corresponding to the data processing demand information. By automatically calling the pre-packaged ant script, the automatic query and processing of data is realized, which greatly reduces the complexity of manual intervention and script writing, and improves the efficiency and accuracy of data processing; finally, the data processing results are displayed through the big model, and the data processing results are presented to the user in the form of charts and reports by using the display ability of the big model, making the results more intuitive and readable. In the above method, through natural language interaction, big model parsing, automated task processing, etc., the complex data analysis process is simplified into a user-friendly interactive form, and professional technicians are no longer required to manually write query scripts, which not only reduces the threshold for use, but also significantly improves the efficiency and accuracy of data analysis. At the same time, the time from data request to result feedback is greatly shortened, supporting enterprises to quickly respond to business needs and improve the scientificity and timeliness of business management decisions.

[0079] In an exemplary embodiment, as shown in FIG2 , before the above step S101 responds to the user's interactive operation on the large model, the following is further included:

[0080] Step S201, obtaining initial variables and initial parameters set by the user according to preset analysis requirements;

[0081] Step S202, expanding the initial variables and initial parameters respectively through the large model to generate related variables and related parameters;

[0082] Step S203, obtaining a preset parameter pool according to the relevant variables, relevant parameters, initial variables and initial parameters.

[0083] Furthermore, in an exemplary embodiment, the above step S102 parses the data processing requirement information through a large model to extract the target variables and target parameters, and also includes: extracting the target variables and target parameters from the data processing requirement information according to a preset parameter pool.

[0084] Exemplarily, first, the terminal obtains the initial variables and initial parameters from the business requirements list provided by the user, including the objects that the user may need to analyze (such as "loan balance", "customer distribution", etc.) and analysis dimensions (such as "past year"). The initial settings are provided by the user in advance to ensure that the big model can be automatically processed based on clear business logic. Then, the big model expands the variables and parameters of the initial settings by analyzing the company's past data and the general experience of the industry. For example, for the initial variable "customer type", the big model may be expanded to "VIP customers", "new customers", "churn risk customers", etc., and the relevant parameters can also be further supplemented to different business scenarios, so that the parameter pool covers more types of business needs and increases the flexibility and adaptability of the analysis. Then, the terminal integrates the initial variables, initial parameters and expanded variables and parameters to form a unified preset parameter pool. The parameter pool includes dimensions such as different business entities, time dimensions, geographical scopes, and processing methods. For example, the parameter pool may include multiple variables and parameters such as "product type", "regional market", and "quarterly time period". These contents will be used as condition matching in subsequent data analysis tasks to help quickly locate data sources and determine analysis methods.

[0085] After receiving the user's natural language input request, the terminal sends the request to the big model for parsing. For example, the user inputs "Please analyze the purchase behavior of VIP customers in the past 12 months." The big model first performs natural language parsing on the user input and identifies key demand elements, including "VIP customers", "purchase behavior", "past 12 months" and other information. Then, the big model matches and verifies these demand elements in combination with the content in the preset parameter pool to ensure that they conform to the business logic of the system. For example, the parameter pool contains "VIP customers" in the customer classification and "past 12 months" in the time range. The big model will match these to the demand information entered by the user, thereby extracting the target variables and target parameters. By parsing in combination with the preset parameter pool, it has higher accuracy and flexibility when facing user needs. Compared with directly parsing the user's natural language input, the matching process using the parameter pool not only improves the accuracy of parsing, but also can automatically complete some analysis information that the user has not clearly stated. For example, if some parameters are missing in the user's needs, the big model can refer to the default options in the parameter pool for completion, thereby effectively reducing the complexity of user operations and ensuring the standardization of analysis results.

[0086] In this embodiment, by obtaining the user's initial variables and parameters before user interaction, expanding them using a large model, and building a preset parameter pool, it is possible to respond to the user's data processing needs more intelligently and quickly in actual use. The existence of the preset parameter pool ensures that each time a user requests analysis, the most appropriate combination of variables and parameters can be quickly matched according to the needs, greatly improving the efficiency and reliability of data analysis, while reducing the user's need for professional knowledge in the analysis process, making the company's business intelligence analysis more automated and convenient.

[0087] In an exemplary embodiment, the above step S103 calls a pre-packaged ant script according to the target variable and the target parameter to obtain a data processing result corresponding to the data processing requirement information, and also includes: determining the corresponding data source according to the target variable, and determining the data range according to the target parameter; obtaining the data to be processed within the data range from the data source; using the data processing method corresponding to the target parameter to process the data to be processed, and obtaining the data processing result corresponding to the data processing requirement information.

[0088] Exemplarily, after receiving the target variable and target parameter, the terminal will determine the appropriate data source according to the target variable. For example, if the target variable is "sales", the sales record table in the corresponding business database will be determined. Then, the terminal further refines the scope of the data query according to the target parameter, such as "past 12 months", "North China", etc. It can clearly identify the specific data location to be queried, ensuring the accuracy and efficiency of data extraction.

[0089] After obtaining the data source and range, the terminal performs the corresponding data extraction operation through the Ant script. The Ant script extracts the qualified raw data from the enterprise database according to the specified data source and data range. For example, the Ant script extracts all sales records in the past 12 months from the "sales record table" through query statements to form a data set to be processed. This step ensures that the extracted data is closely related to user needs and avoids redundancy and deviation in the data extraction process. Subsequently, the Ant script processes the data to be processed according to the data processing method specified by the target parameter. For example, for the target parameter of "quarterly statistics", the Ant script will summarize and analyze the extracted data quarterly, calculate the total sales or growth rate for each quarter, and generate a quarterly statistical result. The processing logic in the Ant script has been modularly encapsulated, and the corresponding processing method can be called according to different target parameters, so as to be reused in different data demand scenarios. Finally, after the Ant script executes all processing operations, it obtains the final data processing result corresponding to the data processing demand information. For example, for the demand of "quarterly sales statistics", the Ant script generates a quarterly sales summary table or chart and passes it to the subsequent modules of the system for display to the user. Through this series of operations, the system realizes the automation of the entire process from data extraction, processing to result generation.

[0090] In this embodiment, the specific data source is determined by the target variable, the data range is refined by the target parameter, and the pre-packaged ant script is used to automatically process the data, so that the data required by the user can be quickly and accurately acquired and processed, and finally the results corresponding to the data processing demand information are obtained. The automation level of data analysis is significantly improved, the complexity of manual intervention and manual query is reduced, and the consistency and efficiency of data processing are ensured. Compared with the traditional analysis process, it can better respond to the natural language needs of users, realize seamless connection from data request to result display, improve the response speed and analysis accuracy of enterprises in data analysis, and enhance the user experience.

[0091] In an exemplary embodiment, before the above step S101 responds to the user's interactive operation on the large model, it also includes: analyzing the data in the enterprise database to obtain the field information of the enterprise database.

[0092] Furthermore, in one of the embodiments, the above-mentioned determining the corresponding data source according to the target variable also includes: determining the data source from the enterprise database according to the target variable and field information.

[0093] Field information refers to the field names, data types, and business meanings of each table in the enterprise database. This field information is used to support subsequent data query and processing, ensuring that the target variable matches the actual data field in the enterprise database to improve the accuracy and reliability of data processing.

[0094] Exemplarily, the terminal automatically analyzes the enterprise database to identify the field information in each data table, such as "customer ID", "sales", "date", etc. The field information will be combined with the big model to provide an accurate basis for the target variables and target parameters in the subsequent data processing process. By analyzing the structure and field information of the database in advance, it is possible to understand the available data sources and the business meaning of each field, so as to quickly locate the specific data location when parsing user needs. After receiving the target variable parsed by the big model, the terminal first matches the target variable with the field information previously obtained. For example, if the target variable is "sales", the terminal will look up the field information to determine the specific data table (such as "sales record table") and the corresponding field (such as "monthly_sales") where the variable "sales" is located. Through such a field information matching process, the most appropriate data source can be determined to ensure that the subsequent data extraction process is accurate.

[0095] In this embodiment, by analyzing the enterprise database before user interaction to obtain the field information of each table, the specific data source can be quickly located based on the target variable and field information. This not only improves the accuracy of data query, but also enhances the response speed when facing different data requirements, ensuring that the data processing process is more efficient and intelligent.

[0096] In an exemplary embodiment, the above step S101 receives data processing requirement information input by the user through natural language in response to the user's interactive operation on the big model, and also includes: generating guidance content through the big model in response to the user's interactive operation on the big model; displaying the guidance content so that the user can input data processing requirement information including target variables and target parameters through natural language.

[0097] Among them, guidance content refers to the information prompts or interactive guidance generated by the big model to help users accurately describe their data analysis needs. These guidance contents are designed to guide users to clearly express their needs and ensure that the input information contains enough details to support subsequent data processing steps.

[0098] For example, when a user starts to interact with the terminal, the terminal generates guidance content through the big model, such as "Please specify the data object you want to analyze, such as 'sales' or 'customer distribution'", or "Please specify the time range for analysis, such as 'past year' or 'most recent quarter'". These guidance contents will be presented in the terminal's user interface in a natural language to help users better understand and input their data processing requirements.

[0099] After the guidance content is displayed, users can enter data requirements according to the guidance prompts, such as "Please analyze the loan balance of a branch in the past 12 months." The big model extracts target variables and target parameters based on the user's input content to ensure that the user's needs are accurately understood and converted into executable analysis requests.

[0100] In this embodiment, by generating and displaying guidance content before the user inputs data requirements, the user is helped to more clearly input the data processing requirements including the target variables and target parameters. This makes the user input process simpler, reduces the requirements for the user's professional skills, and ensures the quality and integrity of the input information. This significantly reduces the risk of unclear user expression or incomplete information, improves the accuracy of understanding user requirements and data processing efficiency, and ultimately enhances the user experience and ease of use of data analysis.

[0101] In an exemplary embodiment, the above step 104 displays the data processing results through a large model, and also includes: determining the corresponding visualization form according to the data processing requirement information; converting the data processing results into corresponding visualization display data according to the visualization form; and displaying the visualization display data corresponding to the data processing results through the large model.

[0102] Among them, visualization refers to the chart or graphic method used to display the data processing results, such as line charts, bar charts, pie charts, etc. The choice of visualization is closely related to the user's data processing needs. Through a reasonable visualization form, the data analysis results can be made more intuitive and easy to understand.

[0103] For example, after receiving the target variables and target parameters generated by the large model and completing data processing, the terminal will determine the appropriate visualization form based on the data processing demand information. For example, if the user's data processing needs involve analysis of time trends (such as "sales changes in the past 12 months"), the system will choose a line chart as the visualization form; if the needs involve classification comparisons (such as "market share of different product lines"), a pie chart or bar chart may be selected for display.

[0104] After determining the visualization form, the terminal will convert it into appropriate visualization data based on the corresponding data processing results. The visualization data contains information such as specific data points, axis labels, color settings, etc. used to generate charts, so that users can easily understand visualization content generated by the terminal. For example, for "Sales changes in the past 12 months", the terminal will calculate and organize the sales data points of each month into a form suitable for line chart display.

[0105] Finally, the terminal displays the converted visual display data through the big model, presenting the generated line chart, bar chart or other charts to the user in the form of a graphical interface, with corresponding explanations. For example, the big model can generate a description of the chart through natural language, such as "The chart shows the gradual growth of sales over the past 12 months, with sales reaching a peak in December." This combination of charts and language descriptions enables users to understand the data analysis results more clearly.

[0106] In this embodiment, by determining the visualization form according to the user's data processing needs and converting the data processing results into corresponding visualization display data, the intuitiveness and comprehensibility of data display are effectively enhanced. Compared with pure numerical results, complex data can be converted into more understandable graphical information through visualization, so that users can more easily obtain the trends and insights behind the data when making business decisions. In this process, the large model not only generates and displays charts, but also provides users with explanations through natural language, making the data display process more comprehensive and friendly, and significantly improving the user experience and the comprehensibility of data results.

[0107] In an exemplary embodiment, as shown in FIG3 , the present application provides an enterprise intelligent analysis response system based on a big model. The system first needs to combine the enterprise's data warehouse to analyze all tables in the data warehouse; then combine the enterprise's business situation to analyze and summarize the enterprise's BI (business intelligence) needs, and define all parameters required for all data analysis to form a parameter pool; then sort out all chart components required for data display technology in data analysis; then encapsulate all the BI requirements of the entire enterprise through Ant scripts (a tool for performing any automated tasks) to call Task components (independent units or modules for performing specific operations and implementing specific functions); let the big model call these Task components to implement the intelligent BI method.

[0108] As shown in FIG4 , it is a schematic diagram of a process of executing a method for intelligent analysis and response of an enterprise based on a big model by an intelligent analysis and response system of an enterprise based on a big model in one embodiment, including:

[0109] Step S401: The user starts asking questions.

[0110] Step S402: Prompt the user in combination with the variable pool.

[0111] Step S403, providing the definition information of the question and script task components to the large model.

[0112] Step S404: the large model interprets the parameters and calls the script.

[0113] Step S405, the script executes the script task in combination with the parameters given by the large model.

[0114] Step S406: the script obtains data from the data warehouse.

[0115] Step S407, the generated result is returned to the large model.

[0116] Step S408: The large model is organized into answer words to give answers.

[0117] For example, the system sorts out the tables and fields in the data warehouse table, which improves the accuracy of some subsequent SQL generation. The system analyzes the BI needs of the enterprise and presets some SQL templates to ensure the accuracy of data acquisition. The system will define a large variable pool and variable value definitions based on the current business status and BI needs of the enterprise. Then, through the prompt project, the user is allowed to directly provide the variable value, and the system matches according to the variable value, and then through the ant script call, the accurate result can be obtained.

[0118] For example, when the system analyzes the BI needs of a certain bank, it can obtain the bank's key variables: product type variables: deposits, loans, wealth management, etc.; entity type: XX institution, XX branch, outlet; BI result variables: point-in-time balance, daily average balance, and growth amount. When the user asks a question, the system can guide the user: "You can ask this: What is the point-in-time balance of deposits of institution A? Give the point-in-time balance of loans of institution B at the end of the last 12 months, and use a chart to show its changing trend." At this time, the big model will automatically execute the corresponding script task (Ant Task) through parameter matching and point-in-time balance, and then generate a line chart with the result data of 12 months, and give the data to the big model, generate a reply script, and send the picture and reply to the user in a unified manner.

[0119] In this embodiment, through the combination of big model and ant script, business personnel can directly ask questions to the big model to get the data analysis results they want; through the intelligent technology of the big model, various data charts can be generated to facilitate business personnel to better obtain data value. Through the prompt project to guide users to ask questions, the desired variable values ​​can be obtained from the user's questions, which improves the accuracy of data analysis results. Allow users to realize data analysis needs through natural language, and improve user experience.

[0120] 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.

[0121] Based on the same inventive concept, the embodiment of the present application also provides a large model-based enterprise intelligent analysis response device for implementing the large model-based enterprise intelligent analysis response 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 one or more large model-based enterprise intelligent analysis response device embodiments provided below can refer to the limitations of the large model-based enterprise intelligent analysis response method above, and will not be repeated here.

[0122] In an exemplary embodiment, as shown in FIG5 , a large model-based enterprise intelligent analysis and response device is provided, comprising: an information receiving module 501, an information extraction module 502, a data processing module 503 and a result display module 504, wherein:

[0123] The information receiving module 501 is used to receive data processing requirement information input by the user through natural language in response to the user's interactive operation on the large model;

[0124] The information extraction module 502 is used to analyze the data processing requirement information through the large model and extract the target variables and target parameters; the target variables are used to describe the data attribute information, and the target parameters are used to describe the data range information and the processing method information;

[0125] The data processing module 503 is used to call the pre-packaged ant script according to the target variables and target parameters to obtain the data processing results corresponding to the data processing requirement information;

[0126] The result display module 504 is used to display the data processing results through a large model.

[0127] In one of the embodiments, the above-mentioned enterprise intelligent analysis response device based on the big model also includes a parameter pool preset module, which is used to obtain the initial variables and initial parameters set by the user according to the preset analysis requirements; the initial variables and initial parameters are expanded respectively through the big model to generate related variables and related parameters; and a preset parameter pool is obtained according to the related variables, related parameters, initial variables and initial parameters.

[0128] In one embodiment, the information extraction module 502 is further used to extract target variables and target parameters from the data processing requirement information according to a preset parameter pool.

[0129] In one embodiment, the above-mentioned data processing module 503 is also used to determine the corresponding data source according to the target variable, and determine the data range according to the target parameter; obtain the data to be processed within the data range from the data source; use the data processing method corresponding to the target parameter to process the data to be processed, and obtain the data processing result corresponding to the data processing requirement information.

[0130] In one of the embodiments, the above-mentioned enterprise intelligent analysis response device based on a large model also includes a field analysis module, which is used to analyze the data in the enterprise database to obtain the field information of the enterprise database.

[0131] In one embodiment, the data processing module 503 is further used to determine the data source from the enterprise database according to the target variable and field information.

[0132] In one of the embodiments, the information receiving module 501 is also used to generate guidance content through the big model in response to the user's interactive operation on the big model; and display the guidance content so that the user can input data processing requirement information including target variables and target parameters through natural language.

[0133] In one of the embodiments, the above-mentioned result display module 504 is also used to determine the corresponding visualization form according to the data processing requirement information; convert the data processing results into corresponding visualization display data according to the visualization form; and display the visualization display data corresponding to the data processing results through a large model.

[0134] Each module in the above-mentioned enterprise intelligent analysis and response device based on a large model can be implemented in whole or in part by software, hardware, or a combination thereof. Each of the above-mentioned modules 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 of the above modules.

[0135] In an exemplary embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be shown in FIG6. 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 via a system bus, and the communication interface, the display unit, and the input device are connected to the system bus via 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 may 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 method for intelligent analysis and response of an enterprise based on a large model is implemented. The display unit of the computer device is used to form a visually visible picture, which may 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.

[0136] Those skilled in the art will understand that the structure shown in FIG. 6 is merely a block diagram of a partial 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 shown in the figure, or combine certain components, or have a different arrangement of components.

[0137] 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.

[0138] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0139] 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.

[0140] 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.

[0141] 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.

[0142] 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.

[0143] 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 method for enterprise intelligent analysis and response based on a large model, characterized in that: The method comprises: In response to the user's interactive operation on the large model, receiving data processing requirement information input by the user through natural language; The data processing requirement information is parsed through a large model to extract target variables and target parameters; the target variables are used to describe data attribute information, and the target parameters are used to describe data range information and processing method information; According to the target variable and the target parameter, a pre-packaged ant script is called to obtain a data processing result corresponding to the data processing requirement information; The data processing results are displayed through the large model.

2. The method according to claim 1, characterized in that Before responding to the user's interactive operation on the large model, it also includes: Obtain the initial variables and initial parameters set by the user for preset analysis requirements; The initial variables and initial parameters are expanded respectively through the large model to generate related variables and related parameters; Obtaining a preset parameter pool according to the relevant variables, the relevant parameters, the initial variables and the initial parameters; The step of parsing the data processing requirement information through a large model and extracting target variables and target parameters includes: According to the preset parameter pool, target variables and target parameters are extracted from the data processing requirement information.

3. The method according to claim 1, characterized in that The step of calling a pre-packaged ant script according to the target variable and the target parameter to obtain a data processing result corresponding to the data processing requirement information includes: Determine the corresponding data source according to the target variable, and determine the data range according to the target parameter; Acquire the data to be processed within the data range from the data source; The data to be processed is processed using a data processing method corresponding to the target parameter to obtain a data processing result corresponding to the data processing requirement information.

4. The method according to claim 3, characterized in that Before responding to the user's interactive operation on the large model, it also includes: Analyze the data in the enterprise database to obtain the field information of the enterprise database; Determining the corresponding data source according to the target variable includes: The data source is determined from the enterprise database according to the target variable and the field information.

5. The method according to claim 1, characterized in that The step of receiving data processing requirement information input by the user through natural language in response to the user's interactive operation on the large model includes: In response to the user's interactive operation on the large model, generating guidance content through the large model; The guidance content is displayed so that the user can input data processing requirement information including the target variable and the target parameter through natural language.

6. The method according to any one of claims 1 to 5, characterized in that: The method of displaying the data processing result through the large model includes: Determine a corresponding visualization form according to the data processing requirement information; According to the visualization form, converting the data processing result into corresponding visualization display data; The large model is used to display visual display data corresponding to the data processing results.

7. An enterprise intelligent analysis and response device based on a large model, characterized in that: The device comprises: An information receiving module, used to receive data processing requirement information input by a user through natural language in response to the user's interactive operation on the large model; An information extraction module is used to parse the data processing requirement information through a large model to extract target variables and target parameters; the target variables are used to describe data attribute information, and the target parameters are used to describe data range information and processing method information; A data processing module, used to call a pre-packaged ant script according to the target variable and the target parameter to obtain a data processing result corresponding to the data processing requirement information; The result display module is used to display the data processing results through the large model.

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