Operation management data intelligent analysis system and method based on AIGC
Through the intelligent analysis system of business management data based on AIGC, using pre-trained large models and natural language processing technology, the problems of time-consuming, labor-intensive and response speed of traditional data analysis methods are solved, and intelligent and personalized data analysis and Real-time decision-making support improves the analysis efficiency and decision-making ability of corporate business data.
Patent Information
- Application Number
- CN202410979381.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-22
- Publication Date
- 2024-11-19
AI Technical Summary
Traditional data analysis methods are time-consuming and labor-intensive, prone to human errors, and are difficult to meet the business needs of massive heterogeneous data and dynamic changes in enterprise management. The existing AIGC technology is difficult to take into account the feature correlation when processing multi-dimensional and multi-type data. , and the computing resource requirements are huge, and the response speed and accuracy are insufficient.
A smart analysis system for business management data based on AIGC is designed, including data acquisition and processing module, industry knowledge management module, intention analysis module, task dismantling and allocation module, data analysis and insight module and automatic report generation module, using pre-pre-pre-pre-pre-pre-pre-pre-production module Training large models and natural language processing technology to realize intelligent data analysis and report generation through intention analysis, task disassembly and data integration.
Implement intelligent data analysis services, improve the efficiency and accuracy of data analysis, provide personalized economic operation analysis reports, have real-time data analysis capabilities, meet rapidly changing market demands, and improve the system's operating efficiency and decision-making support capabilities. .
Smart Images

Figure CN118982175A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the field of information technology, and specifically relates to an AIGC-based business management data intelligent analysis system and method. Background Art
[0002] In today's wave of digitalization and informatization, business operations and management are facing an increasingly complex and changing market environment.
[0003] In the process of enterprise management, a large amount of business data is generated and stored, covering finance, sales, inventory, human resources and other aspects. However, traditional data analysis methods usually rely on manual data collation and report generation, which is not only time-consuming and labor-intensive, but also prone to human errors, resulting in the accuracy and timeliness of analysis results being affected. Although traditional business intelligence (BI) systems can achieve a certain degree of data visualization and report generation, they still have many limitations in data in-depth analysis and intelligent decision support, especially when dealing with massive heterogeneous data and dynamically changing business environments. It is difficult to provide decision support in a timely and accurate manner. Not only is it inefficient, but it is also difficult to meet rapidly changing business needs.
[0004] Therefore, finding a data analysis system that can achieve intelligent and efficient data analysis has become an important task for business management. With the rapid development of artificial intelligence (AI) technology, especially the rise of generative artificial intelligence (AIGC) technology, new opportunities have been brought to the field of data analysis. Through advanced algorithms such as deep learning and natural language processing (NLP), AIGC technology can automatically generate content, perform intelligent semantic analysis and intent understanding, thereby achieving efficient analysis and interpretation of data. This provides enterprises with a new way of intelligent data exploration and analysis, greatly improving the efficiency and accuracy of data analysis.
[0005] However, the existing AIGC technology still faces some challenges in practical applications. First, the diversity and complexity of data make model training and optimization difficult. When dealing with multi-dimensional and multi-type data, existing models often find it difficult to take into account the correlation between various features. Secondly, the dynamic changes in the business environment of enterprises require data analysis systems to have high flexibility and adaptability, but the response speed and accuracy of existing systems when dealing with emergencies and rapidly changing market demands still need to be improved. In addition, large-scale data processing also requires huge computing resources. How to improve the operating efficiency of the system while ensuring the quality of analysis is an important direction of current research.
[0006] Based on the above background, how to use artificial intelligence technology to effectively improve the analysis efficiency and decision-making support capabilities of corporate business data has become an urgent problem to be solved. Summary of the invention
[0007] In order to overcome the problems existing in the related technologies, an AIGC-based business management data intelligent analysis system and method are provided.
[0008] According to one aspect of an embodiment of the present disclosure, there is provided an AIGC-based business management data intelligent analysis system, the system comprising: a data collection and processing module, an industry knowledge management module, an intention analysis module, a task decomposition and allocation module, a data analysis and insight module, and an automatic report generation module;
[0009] The data acquisition and processing module is used to acquire data from different sources, and integrate and optimize the acquired data to form a structured data table;
[0010] The industry knowledge management module is used to extract, classify and store knowledge from the data table to form a knowledge base;
[0011] The intention analysis module is used to extract features and semantically represent the input natural language data based on knowledge enhancement and pre-trained large models to obtain user intentions, where the user intentions include data indicators that the user wants to obtain and / or analysis that the user wants to perform;
[0012] The task decomposition and allocation module is used to decompose the user's intention to obtain multiple subtasks, each of which includes a specific data analysis operation;
[0013] The data analysis and insight module is used to convert the natural language instructions corresponding to each subtask into structured query statements and / or programming codes, and perform data analysis based on the structured query statements and / or programming codes in combination with the knowledge base to obtain analysis results;
[0014] The automatic report generation module is used to generate reports based on the analysis results, predefined templates and user needs.
[0015] In a possible implementation, the data acquisition and processing module includes: a data source access submodule, a data cleaning and processing submodule, and a data integration and optimization submodule;
[0016] The data source access submodule is used to access various data sources through preset connectors and adapters;
[0017] The data cleaning and processing submodule is used to perform data cleaning operations on the data collected by the data source access submodule;
[0018] The data integration and optimization submodule is used to integrate and optimize the data cleaned by the data cleaning and processing submodule through the data processing engine to form a structured data table.
[0019] In a possible implementation, the intent analysis module includes: a feature extraction and semantic representation submodule, an intent classification submodule, a dynamic guidance and recursive drill-down submodule;
[0020] The feature extraction and semantic representation submodule uses a pre-trained large model to extract features from the input natural language data and generate semantic representation;
[0021] The intention classification submodule uses a pre-trained classification model to classify the semantic representation generated by the feature extraction and semantic representation submodule to identify the user intention;
[0022] The dynamic guidance and recursive drill-down submodule is used to generate at least one question according to the identified user intention, so as to guide the user to gradually clarify the needs and form a coherent thinking chain and thinking tree.
[0023] In a possible implementation, the task decomposition and allocation module includes: a demand analysis submodule, a task decomposition submodule, and a task allocation submodule;
[0024] The demand analysis submodule is used to analyze the user's overall demand according to the user's intention;
[0025] The task decomposition submodule is used to refine the user's overall needs into multiple stage needs according to the preset correspondence between tasks and needs, and the correspondence between overall needs and stage needs, and determine one or more subtasks corresponding to each stage need, each subtask including a specific data analysis operation;
[0026] The task allocation submodule is used to select a data model or algorithm suitable for each subtask for processing according to the data analysis operation characteristics of the subtask.
[0027] In a possible implementation, the industry knowledge management module includes: a knowledge extraction and structured processing module, a dynamic update and expansion module, and an intelligent retrieval and recommendation module;
[0028] The knowledge extraction and structured processing module is used to extract industry knowledge from multiple industry-related data sources, and perform structured processing and storage to form a knowledge base;
[0029] The dynamic update and expansion module is used to extract incremental data from data sources related to multiple industries and dynamically update and expand the knowledge base;
[0030] The intelligent retrieval and recommendation module is used to perform intelligent retrieval and recommendation on the knowledge base according to specific user needs.
[0031] In a possible implementation, the data analysis and insight module includes: a judgment submodule, a first calling submodule, a second calling submodule and a data analysis execution submodule;
[0032] The judgment submodule is used to judge the indicator type corresponding to each subtask;
[0033] The first calling submodule, when the indicator type corresponding to the subtask is a statistical indicator, uses the Text2SQL tool to convert the natural language input by the user into a structured query statement, and searches the database and obtains data according to the converted structured query statement;
[0034] The second calling submodule, when the indicator type corresponding to the subtask is an analytical indicator, uses the Text2Code tool to convert the natural language description input by the user into programming code, and the programming code is used to perform advanced data analysis tasks such as predictive modeling, pattern recognition and trend analysis;
[0035] The data analysis execution submodule uses the data obtained by the first calling submodule and the programming code generated by the second calling submodule to execute the data analysis task to obtain the analysis result, and writes the analysis result into a preset database.
[0036] In a possible implementation, the automatic report generation module includes: an access submodule, a visual report generation submodule, and a report automation and optimization submodule;
[0037] The access submodule is used to receive the analysis results output by the data analysis and insight module and generate reports according to predefined templates and user needs;
[0038] The visual report generation submodule is used to convert the obtained analysis results into intuitive charts and graphs through BI tools to generate intelligent reports;
[0039] The report automation and optimization submodule is used to automatically process the report generation process through an intelligent agent, verify the generated reports, and correct report errors.
[0040] According to another aspect of the embodiment of the present disclosure, a method for intelligent analysis of business management data based on AIGC is provided, the method comprising:
[0041] Step 101, based on knowledge enhancement and pre-trained large models, feature extraction and semantic representation are performed on the input natural language data to obtain user intent, where the user intent includes data indicators that the user wants to obtain and / or analysis that the user wants to perform;
[0042] Step 102, decomposing the user intention into multiple subtasks, each of which includes a specific data analysis operation;
[0043] Step 103, converting the natural language instructions corresponding to each subtask into a structured query statement and / or programming code, and performing data analysis according to the structured query statement and / or programming code to obtain analysis results;
[0044] Step 104: Generate a report based on the analysis results, predefined templates and user requirements.
[0045] According to another aspect of an embodiment of the present disclosure, there is provided an AIGC-based business management data intelligent analysis device, the device comprising:
[0046] processor;
[0047] a memory for storing processor-executable instructions;
[0048] Wherein, the processor is configured to execute the above method.
[0049] According to another aspect of an embodiment of the present disclosure, a non-volatile computer-readable storage medium is provided, on which computer program instructions are stored, and the computer program instructions implement the above method when executed by a processor.
[0050] The beneficial effects of the present disclosure are:
[0051] 1) Intelligent data analysis service. This disclosure uses AIGC technology to decompose natural language questions into data analysis tasks through intent analysis algorithms to achieve intelligent data analysis services. Compared with traditional big data processing solutions, this invention is more intelligent. Users do not need to have professional big data processing and big model knowledge and data analysis knowledge. They only need to ask natural language questions to obtain corresponding data analysis results.
[0052] 2) Personalized economic operation analysis report. Based on the large language model Agent and data analysis model, the present disclosure can generate personalized economic operation analysis reports according to the characteristics and needs of user questions. Traditional big data processing solutions often only provide raw statistical analysis results of data, lacking personalized reports and explanations, while the present invention can provide analysis reports that are more in line with the user's actual situation according to the user's questions and needs.
[0053] 3) Real-time data analysis capability. The present invention uses professional Text2Sql and Text2Code tools and BI tools to instantly perform various intelligent calculations from the database and quickly generate data analysis results. Compared with traditional big data processing solutions, the present invention has faster response speed, higher real-time performance and greater intelligence, and can meet users' needs for real-time data analysis.
[0054] The disclosed AIGC-based intelligent analysis system and method for business management data aims to improve the analysis efficiency and decision-making support capabilities of enterprise business data through advanced artificial intelligence technology, provide more accurate and timely business insights for enterprise management, and help enterprises gain advantages in a highly competitive market environment. BRIEF DESCRIPTION OF THE DRAWINGS
[0055] Figure 1 It is a block diagram of an AIGC-based business management data intelligent analysis system shown in the present disclosure.
[0056] Figure 2 It is a flow chart of an intelligent analysis method of business management data based on AIGC shown in the present disclosure. DETAILED DESCRIPTION
[0057] The present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.
[0058] Figure 1 is a block diagram of an AIGC-based business management data intelligent analysis system shown in the present disclosure, such as Figure 1 As shown, the system includes: a data collection and processing module, an industry knowledge management module, an intention analysis module, a task decomposition and allocation module, a data analysis and insight module and an automatic report generation module.
[0059] The data acquisition and processing module is used to acquire data from different sources, and integrate and optimize the acquired data to form a structured data table. The data acquisition and processing module can access multiple data sources, including relational databases, file systems, and API interfaces.
[0060] The industry knowledge management module is used to extract, classify and store the data table to form a knowledge base. The industry knowledge management module collects, organizes, stores and applies knowledge and information of a specific industry by integrating technologies such as natural language processing, machine learning and knowledge graphs.
[0061] The intent analysis module is used to extract features and semantically represent the input natural language data based on knowledge enhancement and pre-trained large models to obtain user intent, which includes data indicators that the user wants to obtain and / or analysis that the user wants to perform. The intent analysis module uses advanced intent recognition algorithms, combined with knowledge enhancement and pre-trained large models, to extract features and semantically represent the user's natural language queries, and achieve high-precision intent recognition.
[0062] The task decomposition and allocation module is used to decompose the user intention to obtain multiple subtasks, each of which includes a specific data analysis operation.
[0063] The data analysis and insight module is used to convert the natural language instructions corresponding to each subtask into structured query statements and / or programming codes, and perform data analysis based on the structured query statements and / or programming codes in combination with the knowledge base to obtain analysis results.
[0064] The automatic report generation module is used to generate reports according to the analysis results, predefined templates and user needs. The automatic report generation module calls BI tools through the intelligent agent to achieve efficient and accurate automatic report generation.
[0065] This disclosure uses AIGC technology to break down natural language questions into data analysis tasks through intent analysis algorithms to achieve intelligent data analysis services. Compared with traditional big data processing solutions, this invention is more intelligent. Users do not need to have professional knowledge of big data processing, big models, and data analysis. They only need to ask natural language questions to obtain corresponding data analysis results.
[0066] The present invention is based on a large language model Agent and a data analysis model, and can generate a personalized economic operation analysis report based on the characteristics and needs of the user's questions. Traditional big data processing solutions often only provide the original statistical analysis results of the data, lacking personalized reports and explanations, while the present invention can provide an analysis report that is more in line with the user's actual situation based on the user's questions and needs.
[0067] The present invention uses professional Text2Sql and Text2Code tools and BI tools to instantly perform various intelligent calculations from the database and quickly generate data analysis results. Compared with traditional big data processing solutions, the present invention has faster response speed, higher real-time performance and more intelligence, which can meet users' needs for real-time data analysis.
[0068] In a possible implementation, the data collection and processing module includes: a data source access submodule, a data cleaning and processing submodule, and a data integration and optimization submodule.
[0069] The data source access submodule is used to access various data sources through preset connectors and adapters; for example, the data source access submodule can flexibly collect data from relational databases (such as MySQL, PostgreSQL), file systems (such as CSV, Excel) and API interfaces (such as RESTful API).
[0070] The data cleaning and processing submodule is used to perform data cleaning operations on the data collected by the data source access submodule, including removing noise, processing missing values and redundant information, etc. The data cleaning and processing submodule can adopt the ETL (extraction, transformation, loading) process to ensure the quality and consistency of the data.
[0071] The data integration and optimization submodule is used to integrate and optimize the data cleaned by the data cleaning and processing submodule through the data processing engine to form a structured data table, providing a high-quality data foundation for subsequent analysis and processing.
[0072] In a possible implementation, the intent analysis module includes: a feature extraction and semantic representation submodule, an intent classification submodule, and a dynamic guidance and recursive drill-down submodule.
[0073] The feature extraction and semantic representation submodule uses a pre-trained large model to extract features from input natural language data and generate semantic representation.
[0074] The intention classification submodule uses a pre-trained classification model to classify the semantic representation generated by the feature extraction and semantic representation submodule to identify the user intention.
[0075] The dynamic guidance and recursive drill-down submodule is used to generate at least one question according to the identified user intention, so as to guide the user to gradually clarify the needs and form a coherent thinking chain and thinking tree.
[0076] In this way, the feature extraction and semantic representation submodule, the intent classification submodule, the dynamic guidance and recursive drill-down submodule cooperate with each other to further guide the user to give specific needs based on the user intent obtained through preliminary identification, and go deep from the overall information to the specific details to form a coherent thinking chain and thinking tree, thereby more accurately identifying the user intent through progressive interaction.
[0077] In a possible implementation, the task decomposition and allocation module includes: a demand analysis submodule, a task decomposition submodule, and a task allocation submodule.
[0078] The demand analysis submodule is used to analyze the user's overall demand according to the user's intention;
[0079] The task decomposition submodule is used to refine the user's overall needs into multiple stage needs according to the preset correspondence between tasks and needs, and the correspondence between overall needs and stage needs, and determine one or more subtasks corresponding to each stage need, each subtask including a specific data analysis operation;
[0080] The task allocation submodule is used to select a data model or algorithm suitable for each subtask for processing according to the data analysis operation characteristics of the subtask.
[0081] In this way, the present disclosure decomposes the overall requirements layer by layer through the relationship between preset requirements and tasks, obtains sub-tasks that can perform actual operations, and configures data analysis tools that meet the operational characteristics for each sub-task, thereby ensuring that the overall requirements can be completed efficiently and accurately.
[0082] In a possible implementation, the industry knowledge management module includes: a knowledge extraction and structured processing module, a dynamic update and expansion module, and an intelligent retrieval and recommendation module.
[0083] The knowledge extraction and structured processing module is used to extract industry knowledge from multiple industry-related data sources, and perform structured processing and storage to form a knowledge base.
[0084] The dynamic update and expansion module is used to extract incremental data from data sources related to multiple industries and dynamically update and expand the knowledge base.
[0085] The intelligent retrieval and recommendation module is used to intelligently retrieve and recommend the knowledge base according to specific user needs, supporting the company's applications in strategic decision-making, market analysis, competitive intelligence and risk management.
[0086] In a possible implementation, the data analysis and insight module includes: a judgment submodule, a first calling submodule, a second calling submodule and a data analysis execution submodule.
[0087] The judgment submodule is used to judge the indicator type corresponding to each subtask.
[0088] The first calling submodule, when the indicator type corresponding to the subtask is a statistical indicator, uses the Text2SQL tool to convert the natural language input by the user into a structured query statement, and searches the database and obtains data based on the converted structured query statement.
[0089] The second calling submodule, when the indicator type corresponding to the subtask is an analytical indicator, uses the Text2Code tool to convert the natural language description input by the user into programming code, and the programming code is used to perform advanced data analysis tasks such as predictive modeling, pattern recognition and trend analysis.
[0090] The data analysis execution submodule uses the data obtained by the first calling submodule and the programming code generated by the second calling submodule to execute the data analysis task to obtain the analysis result, and writes the analysis result into a preset database.
[0091] In a possible implementation, the automatic report generation module includes: an access submodule, a visualization report generation submodule, and a report automation and optimization submodule.
[0092] The access submodule is used to receive the analysis results output by the data analysis and insight module, and generate reports based on predefined templates and user needs.
[0093] The visual report generation submodule is used to convert the obtained analysis results into intuitive charts and graphs through BI tools to generate intelligent reports.
[0094] The report automation and optimization submodule is used to automatically process the report generation process through an intelligent agent, verify the generated reports, and correct report errors.
[0095] Figure 2 is a flow chart of an AIGC-based intelligent analysis method for business management data, as shown in the present disclosure. Figure 2 As shown, the method includes:
[0096] Step 101, based on knowledge enhancement and pre-trained large models, feature extraction and semantic representation are performed on the input natural language data to obtain user intent, where the user intent includes data indicators that the user wants to obtain and / or analysis that the user wants to perform;
[0097] Step 102, decomposing the user intention into multiple subtasks, each of which includes a specific data analysis operation;
[0098] Step 103, converting the natural language instructions corresponding to each subtask into a structured query statement and / or programming code, and performing data analysis according to the structured query statement and / or programming code to obtain analysis results;
[0099] Step 104: Generate a report based on the analysis results, predefined templates and user requirements.
[0100] The description of the above method has been explained in detail in the description of the above system and will not be repeated here.
[0101] The present disclosure may be a system, a method and / or a computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for causing a processor to implement various aspects of the present disclosure.
[0102] A computer-readable storage medium may be a tangible device that can hold and store instructions used by an instruction execution device. A computer-readable storage medium may be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. More specific examples of computer-readable storage media (a non-exhaustive list) include: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disk read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanical encoding device, such as a punch card or a raised structure in a groove on which instructions are stored, and any suitable combination of the foregoing. As used herein, a computer-readable storage medium is not to be interpreted as a transient signal per se, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagating through a waveguide or other transmission medium (e.g., a light pulse through a fiber optic cable), or an electrical signal transmitted through a wire.
[0103] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to each computing / processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, optical fiber transmissions, wireless transmissions, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in the computer-readable storage medium in each computing / processing device.
[0104] The computer program instructions for performing the operation of the present disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-related instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages, such as Smalltalk, C++, etc., and conventional procedural programming languages, such as "C" language or similar programming languages. Computer-readable program instructions may be executed completely on a user's computer, partially on a user's computer, as an independent software package, partially on a user's computer, partially on a remote computer, or completely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., using an Internet service provider to connect via the Internet). In some embodiments, an electronic circuit, such as a programmable logic circuit, a field programmable gate array (FPGA), or a programmable logic array (PLA), may be customized by utilizing the state information of the computer-readable program instructions, and the electronic circuit may execute the computer-readable program instructions, thereby realizing various aspects of the present disclosure.
[0105] Various aspects of the present disclosure are described herein with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and computer program products according to the embodiments of the present disclosure. It should be understood that each box in the flowchart and / or block diagram and the combination of each box in the flowchart and / or block diagram can be implemented by computer-readable program instructions.
[0106] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, thereby producing a machine, so that when these instructions are executed by the processor of the computer or other programmable data processing device, a device that implements the functions / actions specified in one or more boxes in the flowchart and / or block diagram is generated. These computer-readable program instructions can also be stored in a computer-readable storage medium, and these instructions cause the computer, programmable data processing device, and / or other equipment to work in a specific manner, so that the computer-readable medium storing the instructions includes a manufactured product, which includes instructions for implementing various aspects of the functions / actions specified in one or more boxes in the flowchart and / or block diagram.
[0107] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device so that a series of operating steps are performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to implement the functions / actions specified in one or more boxes in the flowchart and / or block diagram.
[0108] The flow chart and block diagram in the accompanying drawings show the possible architecture, function and operation of the system, method and computer program product according to multiple embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a part of a module, program segment or instruction, and a part of the module, program segment or instruction includes one or more executable instructions for realizing the specified logical function. In some alternative implementations, the function marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two continuous square boxes can actually be executed substantially in parallel, and they can sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs the specified function or action, or can be implemented with a combination of special hardware and computer instructions.
[0109] The embodiments of the present disclosure have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The selection of terms used herein is intended to best explain the principles of the embodiments, practical applications, or improvements to the technology in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.
Claims
1. An AIGC-based business management data intelligent analysis system, characterized in that: The system includes: a data collection and processing module, an industry knowledge management module, an intention analysis module, a task decomposition and allocation module, a data analysis and insight module, and an automatic report generation module; The data acquisition and processing module is used to acquire data from different sources, and integrate and optimize the acquired data to form a structured data table; The industry knowledge management module is used to extract, classify and store knowledge from the data table to form a knowledge base; The intention analysis module is used to extract features and semantically represent the input natural language data based on knowledge enhancement and pre-trained large models to obtain user intentions, where the user intentions include data indicators that the user wants to obtain and / or analysis that the user wants to perform; The task decomposition and allocation module is used to decompose the user's intention to obtain multiple subtasks, each of which includes a specific data analysis operation; The data analysis and insight module is used to convert the natural language instructions corresponding to each subtask into structured query statements and / or programming codes, and perform data analysis based on the structured query statements and / or programming codes in combination with the knowledge base to obtain analysis results; The automatic report generation module is used to generate reports based on the analysis results, predefined templates and user needs.
2. The system according to claim 1, characterized in that The data collection and processing module includes: a data source access submodule, a data cleaning and processing submodule, and a data integration and optimization submodule; The data source access submodule is used to access various data sources through preset connectors and adapters; The data cleaning and processing submodule is used to perform data cleaning operations on the data collected by the data source access submodule; The data integration and optimization submodule is used to integrate and optimize the data cleaned by the data cleaning and processing submodule through the data processing engine to form a structured data table.
3. The system according to claim 1, characterized in that The intention analysis module includes: a feature extraction and semantic representation submodule, an intention classification submodule, a dynamic guidance and recursive drilling-down submodule; The feature extraction and semantic representation submodule uses a pre-trained large model to extract features from the input natural language data and generate semantic representation; The intention classification submodule uses a pre-trained classification model to classify the semantic representation generated by the feature extraction and semantic representation submodule to identify the user intention; The dynamic guidance and recursive drill-down submodule is used to generate at least one question according to the identified user intention, so as to guide the user to gradually clarify the needs and form a coherent thinking chain and thinking tree.
4. The system according to claim 1, characterized in that The task disassembly and allocation module includes: a demand analysis submodule, a task disassembly submodule, and a task allocation submodule; The demand analysis submodule is used to analyze the user's overall demand according to the user's intention; The task decomposition submodule is used to refine the user's overall needs into multiple stage needs according to the preset correspondence between tasks and needs, and the correspondence between overall needs and stage needs, and determine one or more subtasks corresponding to each stage need, each subtask including a specific data analysis operation; The task allocation submodule is used to select a data model or algorithm suitable for each subtask for processing according to the data analysis operation characteristics of the subtask.
5. The system according to claim 1, characterized in that The industry knowledge management module includes: knowledge extraction and structured processing module, dynamic update and expansion module, intelligent retrieval and recommendation module; The knowledge extraction and structured processing module is used to extract industry knowledge from multiple industry-related data sources, and perform structured processing and storage to form a knowledge base; The dynamic update and expansion module is used to extract incremental data from data sources related to multiple industries and dynamically update and expand the knowledge base; The intelligent retrieval and recommendation module is used to perform intelligent retrieval and recommendation on the knowledge base according to specific user needs.
6. The system according to claim 1, characterized in that The data analysis and insight module includes: a judgment submodule, a first calling submodule, a second calling submodule and a data analysis execution submodule; The judgment submodule is used to judge the indicator type corresponding to each subtask; The first calling submodule, when the indicator type corresponding to the subtask is a statistical indicator, uses the Text2SQL tool to convert the natural language input by the user into a structured query statement, and searches the database and obtains data according to the converted structured query statement; The second calling submodule, when the indicator type corresponding to the subtask is an analytical indicator, uses the Text2Code tool to convert the natural language description input by the user into programming code, and the programming code is used to perform advanced data analysis tasks such as predictive modeling, pattern recognition and trend analysis; The data analysis execution submodule uses the data obtained by the first calling submodule and the programming code generated by the second calling submodule to execute the data analysis task to obtain the analysis result, and writes the analysis result into a preset database.
7. The system according to claim 1, characterized in that The automatic report generation module includes: an access submodule, a visual report generation submodule, and a report automation and optimization submodule; The access submodule is used to receive the analysis results output by the data analysis and insight module and generate reports according to predefined templates and user needs; The visual report generation submodule is used to convert the obtained analysis results into intuitive charts and graphs through BI tools to generate intelligent reports; The report automation and optimization submodule is used to automatically process the report generation process through an intelligent agent, verify the generated reports, and correct report errors.
8. An intelligent analysis method for business management data based on AIGC, characterized in that: The method comprises: Step 101, based on knowledge enhancement and pre-trained large models, feature extraction and semantic representation are performed on the input natural language data to obtain user intent, where the user intent includes data indicators that the user wants to obtain and / or analysis that the user wants to perform; Step 102, decomposing the user intention into multiple subtasks, each of which includes a specific data analysis operation; Step 103, converting the natural language instructions corresponding to each subtask into a structured query statement and / or programming code, and performing data analysis according to the structured query statement and / or programming code to obtain analysis results; Step 104: Generate a report based on the analysis results, predefined templates and user requirements.
9. An AIGC-based intelligent analysis device for business management data, characterized in that: The device comprises: processor; a memory for storing processor-executable instructions; Wherein, the processor is configured to execute the method of claim 8.
10. A non-volatile computer-readable storage medium having computer program instructions stored thereon, characterized in that: When the computer program instructions are executed by a processor, the method of claim 8 is implemented.
Citation Information
Cited By
Enterprise intelligent analysis response method and device based on large model and computer equipment
CN119417174A
Multi-agent ecological environment big data analysis system based on large model
CN119722419A
Multi-agent automatic analysis method, system, equipment and device based on complex report engine
CN119940541A
Multi-agent automatic analysis method, system, device and apparatus
CN119940541B
Tax data processing analysis method and device based on large model
CN120030112A
