Accounting intelligent data analysis platform and method

Abstract

The invention relates to an accounting intelligent data analysis platform and method, and belongs to the technical field of accounting. The platform comprises a background support module, an application service module, an accounting knowledge base, a plurality of MCP servers, an agent main control module and an accounting intelligent assistant client. According to the invention, a natural language interaction interface is provided for accountants, trend analysis and correlation analysis are carried out on accounting business data and report data, dynamic parameters of a report risk monitoring model are supported, static parameter setting of the report risk monitoring model is supported, and the report risk monitoring efficiency is improved. Problems in accounting reports and businesses are found through the development trend of report data items and data item correlation, the business development situation is predicted through the development trend of business data, and decision support capacity is provided for accounting business development.

Description

Technical Field

[0001] This invention belongs to the field of accounting technology, specifically relating to an intelligent data analysis platform and method for accounting. Background Technology

[0002] Financial institutions have many years of experience in building big data platforms, and the analysis of accounting business data has been systematic. However, they do not yet have a systematic support capability for discovering risks in accounting statements.

[0003] In addition, enabling the data analysis function of the big data platform to provide effective support for business development requires a combination of business, technology, and data science capabilities. It is difficult for accounting personnel alone to clearly describe the requirements.

[0004] The current state of accountants' application of big data platform data analysis functions is as follows: 1. In terms of risk detection in financial statements, most accountants still rely on experience or manual processing of analysis reports provided by big data platforms; there is a lack of automated processing for risk detection in financial statements and a lack of dynamic parameter support in risk detection in financial statements.

[0005] 2. Accountants often find it difficult to articulate their needs that would allow big data platforms to leverage their data analytics capabilities.

[0006] 3. The human-computer interface of the big data platform's data analysis function is not user-friendly for accountants.

[0007] Therefore, overcoming the shortcomings of existing technologies is an urgent problem to be solved in the field of accounting technology. Summary of the Invention

[0008] The purpose of this invention is to address the shortcomings of existing technologies and provide an intelligent data analysis platform and method for accounting.

[0009] To achieve the above objectives, the technical solution adopted by the present invention is as follows: An intelligent data analysis platform for accounting includes a back-end support module, an application service module, an accounting knowledge base, multiple MCP servers, an intelligent agent control module, and an intelligent accounting assistant client. in: The backend support modules include a data analysis system, a monitoring system, and a large language model; The application service module includes interconnected accounting systems and accounting reporting systems; The accounting knowledge base is a vector database that stores accounting knowledge and its metadata. Multiple MCP servers, including: The MCP server for the analysis system is used to encapsulate and call the service interfaces of the data analysis system. The accounting system MCP server is used to encapsulate and call the service interfaces of the accounting system. The accounting reporting system MCP server is used to encapsulate and call the service interfaces of the accounting reporting system. The monitoring system's MCP server is used to transmit information between the monitoring system and the intelligent agent's main control module; The knowledge base MCP server is used to respond to queries and retrieve information from the accounting knowledge base; The intelligent agent main control module is used to control the conversation flow. Specifically, it is used to receive the user's interactive request input through natural language, organize prompt words containing callable function information and submit them to the large language model, receive the operation instructions returned by the large language model, and call the corresponding MCP server according to the operation instructions to perform data analysis, risk monitoring or report query operations. The accounting intelligent assistant client provides a user interface for receiving user input, displaying session results, and receiving alarm notifications.

[0010] Furthermore, preferably, the data analysis system is used to perform trend analysis and correlation analysis on data from accounting systems and accounting reporting systems, and provides a trend analysis application programming interface (i.e., AP) that includes a range of predicted values. The monitoring system has a built-in report risk monitoring model; the monitoring system is used to monitor the report data generated by the accounting report system, and will issue an alarm message after identifying a report risk event; The report risk monitoring model in the monitoring system obtains dynamic parameters by calling the trend analysis application interface.

[0011] Furthermore, preferably, the original data on which the accounting reporting system relies to generate reports is managed by the accounting system. The accounting knowledge base stores content structure description information for various reports. The content structure description information includes semantic descriptions of report data items and their corresponding resource names in the accounting reporting system. The content structure description information is JSON format text used to establish the mapping relationship between accounting terms and report data items.

[0012] Furthermore, preferably, the intelligent agent main control module integrates a program execution environment; when the large language model returns automatically generated program code, the program code is executed in the program execution environment, and the program code can call the corresponding MCP server or the functions provided by the intelligent agent main control module.

[0013] Furthermore, preferably, the initial version of the report content structure description information is automatically generated by the report format file of the accounting report system combined with data dictionary information, and can be adjusted and updated and stored in the accounting knowledge base.

[0014] Furthermore, preferably, the accounting intelligent assistant client obtains alarm information from the intelligent main control module through periodic polling and displays pop-up prompts.

[0015] This invention also provides a data analysis method based on the aforementioned intelligent data analysis platform for accounting, including a dynamic parameter support method for report risk discovery, the steps of which include: The data analysis system runs a trend analysis model and is deployed as an API to provide a range of predicted values; The report risk monitoring model in the monitoring system calls the API during operation to obtain the predicted value range of a specified data item as a dynamic detection parameter; When the actual report data detected exceeds the predicted value range, the report risk monitoring model triggers an alarm.

[0016] Furthermore, preferably, the data analysis method based on the aforementioned intelligent data analysis platform for accounting includes a method for generating and submitting data analysis requests through natural language, the steps of which include: The accounting intelligent assistant client receives data analysis request descriptions input by users in natural language. The intelligent agent main control module organizes prompt words and submits them to the large language model. The prompt words include user needs, callable function information, and data analysis requirement templates obtained from the knowledge base MCP server. The large language model generates a data analysis requirement document that conforms to the template specification based on the prompt words; The intelligent agent's main control module submits the generated data analysis requirement document to the data analysis system through the analysis system's MCP server.

[0017] Furthermore, preferably, before submitting the data analysis requirements document to the data analysis system, a requirements review and iteration step is also included: The generated data analysis requirements document is returned to the client for user review. Based on the user's feedback input in natural language, the requirement description is revised and the data analysis requirement document is regenerated until the user confirms it.

[0018] Furthermore, preferably, the data analysis method based on the aforementioned intelligent data analysis platform for accounting includes a method for querying data analysis results using natural language, the steps of which include: The accounting intelligent assistant client receives data analysis result query requests from users, inputted in natural language. The intelligent agent main control module organizes prompt words and submits them to the large language model to obtain a query instruction to call the analysis system MCP server; The intelligent agent main control module calls the analysis system MCP server according to the query command to obtain the corresponding data analysis results from the data analysis system; The acquired data analysis results are returned to the accounting intelligent assistant client for display.

[0019] This invention provides dynamic parameter support for monitoring report risks through real-time model services; and provides decision support for accounting operations, including but not limited to improvements and adjustments to accounting systems, accounting reporting systems and their parameter configurations.

[0020] This invention provides accountants with a natural language interactive interface to perform trend analysis and correlation analysis on accounting business data and report data. It supports dynamic parameters and static parameter settings for the report risk monitoring model. By analyzing the development trends and correlations of report data items, it identifies problems in accounting reports and business operations. By analyzing the development trends of business data, it predicts business development trends and provides decision support capabilities for accounting business development.

[0021] This invention addresses issues such as reporting risk detection and assisting accountants in refining data analysis. It leverages trend and correlation analysis based on a big data platform, utilizes the MCP protocol and a large language model as its intelligent core, and is supported by an accounting knowledge base to provide dynamic parameter support for the reporting risk monitoring model in the monitoring system. Based on the accounting knowledge base and the large language model, it assists accountants in refining their data analysis based on trend and correlation analysis techniques. Accountants can generate and submit data analysis requests through natural language interaction; view data analysis reports and various data analysis results; and, based on these results, implement improvements and adjustments to the accounting reporting system, accounting system, and their parameter configurations.

[0022] Compared with the prior art, the beneficial effects of this invention are as follows: 1. In the data analysis system, after building the trend analysis model, the trained model is deployed as an API service. When the monitoring system runs the report risk analysis model, it automatically calls the model's API interface to obtain the predicted value range of monitoring indicators at a specified confidence level in real time. When the indicator value exceeds this range, the report risk analysis model sends an alarm message. By scheduling incremental data training for the trend analysis model in the data analysis system when the accounting reporting system can provide new report data, the model's continuous support capability is ensured. This method achieves dynamic parameter support in report risk detection.

[0023] 2. Accounting personnel, using natural language and a client-side interface, can view various exploratory analysis results provided by the data analysis system, including the mean, median, distribution, line charts, heatmaps, scatter plots, etc., for various data types. Based on their business experience and guided by these exploratory analysis results, they can propose trend analysis and correlation analysis requirements from a business perspective. Supported by the accounting knowledge base and large language model, this automatically generates data analysis requirement documents, which are then submitted to the data analysis system for implementation by technical personnel and data scientists. Artificial intelligence establishes a knowledge mapping relationship between accounting and technical personnel, reducing the breadth of knowledge required for implementing big data analysis and lowering communication costs.

[0024] 3. Accounting personnel can view various data analysis results through natural language, improving work efficiency. Attached Figure Description

[0025] Figure 1 This is a schematic diagram of the architecture of the intelligent data analysis platform for accounting in this invention. Detailed Implementation

[0026] The present invention will now be described in further detail with reference to the embodiments.

[0027] Those skilled in the art will understand that the following embodiments are for illustrative purposes only and should not be construed as limiting the scope of the invention. Where specific techniques or conditions are not specified in the embodiments, they are performed in accordance with the techniques or conditions described in the literature in the field or according to the product instructions. Materials or equipment whose manufacturers are not specified are all conventional products that can be obtained by purchase.

[0028] Example 1 like Figure 1 As shown, an intelligent data analysis platform for accounting includes a back-end support module, an application service module, an accounting knowledge base, multiple MCP servers, an intelligent agent control module, and an intelligent accounting assistant client. in: The backend support modules include a data analysis system, a monitoring system, and a large language model; The application service module includes interconnected accounting systems and accounting reporting systems; The accounting knowledge base is a vector database that stores accounting knowledge and its metadata. Multiple MCP servers, including: The MCP server for the analysis system is used to encapsulate and call the service interfaces of the data analysis system. The accounting system MCP server is used to encapsulate and call the service interfaces of the accounting system. The accounting reporting system MCP server is used to encapsulate and call the service interfaces of the accounting reporting system. The monitoring system's MCP server is used to transmit information between the monitoring system and the intelligent agent's main control module; The knowledge base MCP server is used to respond to queries and retrieve information from the accounting knowledge base; The intelligent agent main control module is used to control the conversation flow. Specifically, it is used to receive the user's interactive request input through natural language, organize prompt words containing callable function information and submit them to the large language model, receive the operation instructions returned by the large language model, and call the corresponding MCP server according to the operation instructions to perform data analysis, risk monitoring or report query operations. The accounting intelligent assistant client provides a user interface for receiving user input, displaying session results, and receiving alarm notifications.

[0029] Example 2 like Figure 1 As shown, an intelligent data analysis platform for accounting includes a back-end support module, an application service module, an accounting knowledge base, multiple MCP servers, an intelligent agent control module, and an intelligent accounting assistant client. in: The backend support modules include a data analysis system, a monitoring system, and a large language model; The application service module includes interconnected accounting systems and accounting reporting systems; The accounting knowledge base is a vector database that stores accounting knowledge and its metadata. Multiple MCP servers, including: The MCP server for the analysis system is used to encapsulate and call the service interfaces of the data analysis system. The accounting system MCP server is used to encapsulate and call the service interfaces of the accounting system. The accounting reporting system MCP server is used to encapsulate and call the service interfaces of the accounting reporting system. The monitoring system's MCP server is used to transmit information between the monitoring system and the intelligent agent's main control module; The knowledge base MCP server is used to respond to queries and retrieve information from the accounting knowledge base; The intelligent agent main control module is used to control the conversation flow. Specifically, it is used to receive the user's interactive request input through natural language, organize prompt words containing callable function information and submit them to the large language model, receive the operation instructions returned by the large language model, and call the corresponding MCP server according to the operation instructions to perform data analysis, risk monitoring or report query operations. The accounting intelligent assistant client provides a user interface for receiving user input, displaying session results, and receiving alarm notifications.

[0030] The data analysis system is used to perform trend analysis and correlation analysis on data from accounting systems and accounting reporting systems, and provides a trend analysis application interface that includes a range of predicted values; The monitoring system has a built-in report risk monitoring model; the monitoring system is used to monitor the report data generated by the accounting report system, and will issue an alarm message after identifying a report risk event; The report risk monitoring model in the monitoring system obtains dynamic parameters by calling the trend analysis application interface.

[0031] The raw data upon which the accounting reporting system relies to generate reports is managed by the accounting system itself. The accounting knowledge base stores content structure description information for various reports. The content structure description information includes semantic descriptions of report data items and their corresponding resource names in the accounting reporting system. The content structure description information is JSON format text used to establish the mapping relationship between accounting terms and report data items.

[0032] The intelligent agent main control module integrates a program execution environment; when the large language model returns automatically generated program code, the program code is executed in the program execution environment, and the program code can call the corresponding MCP server or the functions provided by the intelligent agent main control module.

[0033] The initial version of the report content structure description information is automatically generated by the report format file of the accounting report system combined with data dictionary information, and can be adjusted and updated and stored in the accounting knowledge base.

[0034] The accounting intelligent assistant client obtains alarm information from the main control module of the intelligent entity through periodic polling and displays pop-up prompts.

[0035] Example 3 A data analysis method based on the accounting intelligent data analysis platform described in Embodiment 1 or Embodiment 2, including a dynamic parameter support method for report risk discovery, comprising the following steps: The data analysis system runs a trend analysis model and is deployed as an API to provide a range of predicted values; The report risk monitoring model in the monitoring system calls the API during operation to obtain the predicted value range of a specified data item as a dynamic detection parameter; When the actual report data detected exceeds the predicted value range, the report risk monitoring model triggers an alarm.

[0036] Example 4

[0037] A data analysis method based on the accounting intelligent data analysis platform described in Embodiment 1 or Embodiment 2 includes a method for generating and submitting data analysis requests through natural language, the steps of which include: The accounting intelligent assistant client receives data analysis request descriptions input by users in natural language. The intelligent agent main control module organizes prompt words and submits them to the large language model. The prompt words include user needs, callable function information, and data analysis requirement templates obtained from the knowledge base MCP server. The large language model generates a data analysis requirement document that conforms to the template specification based on the prompt words; The intelligent agent's main control module submits the generated data analysis requirement document to the data analysis system through the analysis system's MCP server.

[0038] Before submitting the data analysis requirements document to the data analysis system, a requirements review and iteration process is also included: The generated data analysis requirements document is returned to the client for user review. Based on the user's feedback input in natural language, the requirement description is revised and the data analysis requirement document is regenerated until the user confirms it.

[0039] Example 5 A data analysis method based on the intelligent data analysis platform for accounting described in Embodiment 1 or Embodiment 2 includes a method for querying data analysis results using natural language, comprising the following steps: The accounting intelligent assistant client receives data analysis result query requests from users, inputted in natural language. The intelligent agent main control module organizes prompt words and submits them to the large language model to obtain a query instruction to call the analysis system MCP server; The intelligent agent main control module calls the analysis system MCP server according to the query command to obtain the corresponding data analysis results from the data analysis system; The acquired data analysis results are returned to the accounting intelligent assistant client for display.

[0040] Example 6

[0041] like Figure 1 As shown, an intelligent data analysis platform for accounting can be divided into six main modules, as detailed below. The first category is backend support modules, including data analysis systems, monitoring systems, and large language models; The second category is application service modules, including accounting reporting systems and accounting systems; The third category is accounting knowledge base; The fourth category is MCP servers that interact with various systems; The fifth category is the intelligent agent main control module; The sixth category is accounting and accounting intelligent assistant clients.

[0042] The descriptions of each system are as follows.

[0043] 1. Backend support modules 1.1 Data Analysis System A data analysis system refers to a data analysis system based on a big data platform used to support accounting operations. It primarily runs trend analysis models and correlation analysis applications. The data sources supporting these two types of analysis applications mainly include data from accounting systems and accounting statement systems. Trend analysis of data from accounting systems mainly predicts the development trends of various aspects of accounting operations for decision support and problem analysis. Trend analysis of data from accounting statement systems mainly analyzes the changing trends of specific data items in accounting statements. Its analysis model runs continuously and provides a real-time predicted value range with a specified confidence level through an API interface. This range is invoked in the report risk monitoring model of the monitoring system, essentially providing dynamic parameter support for the report risk monitoring model. Correlation analysis is mainly used to analyze the correlation coefficients or correlation matrices between data items in the report. Logical relationships between report data items are usually deterministic, largely based on experience, and checked through report validation formulas. Correlation relationships between data items, which are probabilistic, are reviewed through the report risk monitoring model. Other conventional analyses are also provided in the data analysis system and will not be elaborated upon in this invention.

[0044] 1.2 Monitoring System The monitoring system here specifically refers to an application monitoring system used to monitor the report data generated by the accounting reporting system. The monitoring system provides both event-driven and timed information acquisition capabilities for accounting reports, supports various report risk monitoring models, and will issue alarm information upon identifying a report risk event.

[0045] 1.3 Large Language Model Large language models refer to generative large language models such as Deepseek, Qianwen, and Doubao. These models can be provided through vendor-provided services or through local deployment. Local deployment ensures data security but is more expensive.

[0046] 2. Application service modules 2.1 Accounting Reporting System An accounting reporting system refers to the internal production system of an enterprise (especially financial institutions such as banks), or it may be an accounting reporting module, providing functions for generating and managing accounting reports. Its data is extracted by big data platforms to support accounting data analysis. Simultaneously, it provides service call support to the reporting system's MCP server. It mainly includes two types of transaction service functions: one is accounting report query transaction services, used by accounting personnel to query report data; the other is accounting report system parameter (including data validation formula) management transaction services, executed when accounting personnel need to adjust the parameters of the reporting system to address report risks.

[0047] 2.2 Accounting System An accounting system refers to the internal production system of an enterprise (especially financial institutions such as banks). It may be called an accounting middle platform or various other names, and its main function is accounting. Its data is extracted by accounting reporting systems and big data platforms to generate reports and conduct accounting data analysis. Simultaneously, it provides service access support to the accounting system's MCP server. This mainly includes two types of transaction service functions: one is raw accounting data query services, used by accounting personnel to query data related to report risk management; the other is accounting system parameter management services, executed when accounting personnel need to adjust accounting system parameters to standardize the raw business data used to generate reports when handling report risks.

[0048] 3. Accounting Knowledge Base 3. The Accounting Knowledge Base is a vector database storing accounting knowledge. Besides knowledge documents, it also stores corresponding metadata, such as knowledge type, source (filename, large language model, etc.), location, publisher, and publication time. Its content includes policies, standards, and regulations related to finance and accounting, as well as content added by accounting personnel. This content is divided into two main categories: management and knowledge. In addition, it includes accounting rule knowledge, which is loaded from the accounting rule base of the accounting system.

[0049] The accounting knowledge base also stores the content structure of each type of report in the accounting reporting system and the mapping information for the resource names of report data items. Each type of report content structure is described using a JSON-formatted text file, which includes the corresponding report data item resource name (used to reference reports and their data items in the accounting reporting system when automatically generating report risk monitoring models). The two-dimensional table illustrates the report structure through a tree-like sequence describing the columns. Describing the report content structure primarily establishes the correspondence between report content and information in the accounting system, providing semantic support for querying data in accounting information. The additional description of the report data item resource names establishes the correspondence between the textual description of the report content and the references to accounting report data items. The initial version of the report content structure is automatically generated from the report format file in the accounting reporting system, with the corresponding information of the report data items in the data dictionary appended. It is then saved after adjustments by accounting personnel.

[0050] 4. MCP server The MCP server primarily provides three types of functions: tools (function interfaces that can be called), resources (data sources that can be queried), and hints (predefined instruction templates).

[0051] MCP servers include: 4.1 Analysis System MCP Server The analysis system's MCP server supports accountants in submitting data analysis requests, querying data analysis reports, and encapsulating interfaces for three main categories of transactions: other routine analysis.

[0052] 4.2 Reporting System MCP Server The MCP server of the reporting system encapsulates the two types of transaction service interfaces provided by the accounting reporting system. It stores the report resource information and parameter resource information of the accounting reporting system.

[0053] 4.3 Accounting System MCP Server The accounting system's MCP server encapsulates the two types of transaction service interfaces provided by the accounting system. It also implements the function of converting between the JSON data format of the accounting intelligent assistant and the transaction data format of the accounting system.

[0054] 4.4 Monitoring System MCP Server The monitoring system's MCP server implements the transaction interface between the monitoring system and the accounting intelligent assistant, providing three types of functions: first, it transmits report risk event alarm information to the accounting intelligent assistant; second, it supports the accounting intelligent assistant in querying detailed information about report risk events from the monitoring server; and third, it supports the accounting intelligent assistant in transmitting new report risk monitoring model information to the monitoring system. The new report risk monitoring model needs to undergo review, adjustment, testing, verification, and release within the monitoring system. These functions are provided by the monitoring system and will not be discussed further in this invention.

[0055] 4.5 Knowledge Base MCP Server The knowledge base MCP server provides knowledge base query functionality. It supports natural language query requests and metadata-based query requests. Multiple query results can be provided.

[0056] 5. Intelligent Agent Main Control Module The accounting intelligent assistant agent implements the agent's main control function. It is used for overall control of the session flow. Sessions are initiated by the accounting intelligent assistant client or custom events. Functions such as authentication, information security, process management, and error handling are implemented by the agent's main control module. The accounting intelligent assistant agent provides a Python runtime environment. In most cases, the large language model returns JSON format information calling functions provided by the MCP server or the accounting intelligent assistant agent. When functions not provided by existing program modules or without corresponding function call interfaces are involved, the accounting intelligent assistant agent requests the large language model to automatically generate and return a Python program. This Python program can use resources including existing MCP servers or callable function services provided by the accounting intelligent assistant agent to achieve the target function. This Python program is called by the accounting intelligent assistant agent.

[0057] 6. Intelligent Accounting Assistant Client The accounting intelligent assistant client offers several client formats, including web pages, mobile apps, PC client programs, and application system plugins, serving as the primary channels for customers to use the accounting intelligent assistant. It mainly provides various functions through interactive sessions. These clients continuously check the accounting intelligent assistant's report risk alarm information through periodic polling, and notify the user via pop-up windows upon receiving alarm information.

[0058] Application Examples This invention provides a data analysis method for an intelligent accounting assistant. The main components involved in this invention include... Figure 1 As shown.

[0059] We will use several typical application scenarios as examples to illustrate the specific implementation of the present invention.

[0060] I. Dynamic Parameter Support in Report Risk Discovery In financial statement risk management, the main issue that accountants need to address is errors in the report data. The causes of these errors primarily include errors in the original business data used to generate the reports and errors in the data entry itself. These errors can be identified through two main methods: one is to validate the relationships between report data items using formulas; the other is to analyze the trends of the report data item indicators to find the "normal" value range, and then trace the source of data exceeding the "normal" range to verify the specific problem.

[0061] Trend analysis models based on big data platforms are used to analyze the development trends of data items in reports and define the "normal" numerical range over a given time period. Here, we do not restrict the specific modeling method, whether it's the moving average method or seasonal decomposition method in descriptive trend analysis, or the use of traditional statistical models, machine learning models, or pattern recognition methods. Our solution is to deploy the trained model as an API service after completing the trend analysis model construction.

[0062] When defining a report risk analysis model, for models that use trend analysis methods for risk detection, if the data analysis system deploys an API interface that supports trend analysis of data items used by the report risk analysis model (the description and content of this interface, as resource information, will be passed to the large language model as part of the prompt information when the report risk analysis model program is generated through the large language model), the report risk analysis model program will call the trend analysis model API interface to obtain the "normal" value space in real time and complete the detection of whether the relevant data items in the report conform to the trend; when the indicator value exceeds the range, the report risk analysis model sends an alarm message.

[0063] This method enables dynamic parameter support in report risk discovery.

[0064] II. Accounting personnel use natural language to generate data analysis requests and query analysis results. Note: Accountants' data analysis needs primarily focus on trend analysis and correlation analysis. The main data sources are accounting systems and financial reporting systems, with supplementary data sources including various business systems of financial institutions and some external enterprise data. The acquisition of this data is typically planned and implemented uniformly by a big data platform, and will not be elaborated upon in this invention.

[0065] Accountants typically begin trend and correlation analysis with exploratory analysis. This involves understanding the mean, median, and distribution of various data points from the data source, and examining information such as line charts, heatmaps, and scatter plots to identify areas requiring further modeling or correlation analysis. This process can also be aided by data mining methods. We refer to this type of requirement as exploratory requirement analysis.

[0066] Based on this, accountants, using their experience and judgment, can further propose requirements for trend analysis models, or for calculations of correlation coefficients and correlation matrices, statistical significance tests, interpretation, and analysis. We refer to these requirements as model requirements.

[0067] Verification formulas in financial statements describe definite logical relationships; correlation analysis, on the other hand, merely obtains the correlation between data items at the data level. While highly correlated data items may have definite logical relationships, they are more often data-related relationships with probabilistic characteristics. Therefore, the results of correlation analysis are mainly used as evaluation criteria for financial statement risk monitoring models and require further verification by accounting personnel.

[0068] After accountants propose exploration or modeling requirements, these requirements are transmitted to the data analysis system via the MCP server. The process by which technicians and data scientists implement these requirements, whether through AI or traditional project implementation methods, is beyond the scope of this invention. Once the data analysis results are generated, including visualizations and analysis reports, they are organized within the data analysis system based on the requirements and can be queried through the MCP server.

[0069] 2.1 Accounting personnel generate and submit data analysis requirements Prerequisite: The current accounting personnel have the authority to submit data analysis requests. Data analysis request templates are stored in the accounting knowledge base.

[0070] Process description: 1. In a new session established by the user in the accounting intelligent assistant client, the user enters specific exploration or model requirements. These requirements are related to the data in the accounting system or accounting report system. The accounting intelligent assistant client will then send the request to the accounting intelligent assistant agent. 2. The accounting intelligent assistant will submit the prompt information, including information on callable functions (information on all functions provided by the MCP server and functions supported by the accounting intelligent assistant) and user input information to the large language model; 3. Based on the prompt word information, the large language model returns the function and parameters to be called next; 4. The accounting intelligent assistant, based on information returned by the large language model, queries the corresponding management, accounting rule, report content structure, and data analysis requirement document templates through the knowledge base MCP server. This information represents the matching relationship between the data implementation in the accounting system and accounting report system and the organization's accounting management requirements, as well as the organization's requirements for the chapter structure and content of accounting data analysis. Through prompt word engineering, the accounting intelligent assistant passes the information retrieved from the accounting knowledge base, combined with user input, to the large language model. 5. The accounting intelligent assistant receives the data analysis requirement document generated by the large language model according to the requirements of the data analysis requirement document template, and returns it to the accounting intelligent assistant client; 6. Accounting personnel can check the data analysis requirements through the accounting intelligent assistant client and make adjustments to the data analysis requirements through natural language.

[0071] 7. Repeat steps 1-6 above. During the process, the accounting intelligent assistant will transmit the latest data analysis requirements and adjustment requirements as user input to the big data model until the user confirms and submits the requirements to the data analysis system. 8. The accounting intelligent assistant submits data analysis requests to the data analysis system through the analysis system's MCP server; 9. The successful submission information returned by the data analysis system is processed by the analysis system's MCP server and the accounting intelligent assistant agent, and then returned to the accounting intelligent assistant client.

[0072] 2.2 Accounting personnel review data analysis results By default, all accounting personnel have permission to view data analysis results. If query access control is required, permission settings must be configured.

[0073] Process description: 1. Users can create a new session in the accounting intelligent assistant client and use natural language to input their request to query the analysis results of x recently generated accounting data. 2. The accounting intelligent assistant will submit the prompt information, including information on callable functions (information on all functions provided by the MCP server and functions supported by the accounting intelligent assistant) and user input information to the large language model; 3. The large language model returns the transaction function and parameter information of the calling analysis system's MCP server; 4. The accounting intelligent assistant calls the analysis system's MCP server to query the analysis results of x recently generated accounting data. 5. The analysis system's MCP server queries the data analysis system, which returns a list of titles for the x most recently generated accounting data analysis results. The analysis system's MCP server then transmits the returned information to the accounting intelligent assistant. 6. The accounting intelligent assistant will transmit the returned information to the accounting intelligent assistant client; 7. Users specify the specific accounting data analysis results they wish to view via natural language. The accounting intelligent assistant client then sends the request to the accounting intelligent assistant agent. 8. The accounting intelligent assistant will submit the prompt information, including information on callable functions (information on all functions provided by the MCP server and functions supported by the accounting intelligent assistant) and user input information to the large language model; 9. The large language model returns the transaction function and parameter information of the calling analysis system's MCP server; 10. The accounting intelligent assistant calls the analysis system's MCP server to obtain specific accounting data analysis results, formats the information, and returns it to the accounting intelligent assistant client. 11. Users can view accounting data analysis results through the accounting intelligent assistant client. The content includes, but is not limited to, text and images, mainly analysis reports or charts and images corresponding to exploration needs.

[0074] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of this invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent data analysis platform for accounting, characterized in that, It includes a backend support module, an application service module, an accounting knowledge base, multiple MCP servers, an intelligent agent control module, and an accounting intelligent assistant client; in: The backend support modules include a data analysis system, a monitoring system, and a large language model; The application service module includes interconnected accounting systems and accounting reporting systems; The accounting knowledge base is a vector database that stores accounting knowledge and its metadata. Multiple MCP servers, including: The MCP server for the analysis system is used to encapsulate and call the service interfaces of the data analysis system. The accounting system MCP server is used to encapsulate and call the service interfaces of the accounting system. The accounting reporting system MCP server is used to encapsulate and call the service interfaces of the accounting reporting system. The monitoring system's MCP server is used to transmit information between the monitoring system and the intelligent agent's main control module; The knowledge base MCP server is used to respond to queries and retrieve information from the accounting knowledge base; The intelligent agent main control module is used to control the conversation flow. Specifically, it is used to receive the user's interactive request input through natural language, organize prompt words containing callable function information and submit them to the large language model, receive the operation instructions returned by the large language model, and call the corresponding MCP server according to the operation instructions to perform data analysis, risk monitoring or report query operations. The accounting intelligent assistant client provides a user interface for receiving user input, displaying session results, and receiving alarm notifications.

2. The intelligent data analysis platform for accounting as described in claim 1, characterized in that, The data analysis system is used to perform trend analysis and correlation analysis on data from accounting systems and accounting reporting systems, and provides a trend analysis application interface that includes a range of predicted values; The monitoring system has a built-in report risk monitoring model; The monitoring system is used to monitor the report data generated by the accounting reporting system, and will issue an alarm message after identifying a risk event in the report. The report risk monitoring model in the monitoring system obtains dynamic parameters by calling the trend analysis application interface.

3. The intelligent data analysis platform for accounting as described in claim 1, characterized in that, The raw data upon which the accounting reporting system relies to generate reports is managed by the accounting system itself. The accounting knowledge base stores content structure description information for various reports. The content structure description information includes semantic descriptions of report data items and their corresponding resource names in the accounting reporting system. The content structure description information is JSON format text used to establish the mapping relationship between accounting terms and report data items.

4. The intelligent data analysis platform for accounting as described in claim 1, characterized in that, The intelligent agent main control module integrates a program execution environment; when the large language model returns automatically generated program code, the program code is executed in the program execution environment, and the program code can call the corresponding MCP server or the functions provided by the intelligent agent main control module.

5. The intelligent data analysis platform for accounting as described in claim 1, characterized in that, The initial version of the report content structure description information is automatically generated by the report format file of the accounting report system combined with data dictionary information, and can be adjusted and updated and stored in the accounting knowledge base.

6. The intelligent data analysis platform for accounting as described in claim 1, characterized in that, The accounting intelligent assistant client obtains alarm information from the main control module of the intelligent entity through periodic polling and displays pop-up prompts.

7. A data analysis method based on the intelligent data analysis platform for accounting as described in any one of claims 1 to 6, characterized in that, This includes dynamic parameter support methods for risk discovery in reports, the steps of which include: The data analysis system runs a trend analysis model and is deployed as an API to provide a range of predicted values; The report risk monitoring model in the monitoring system calls the API during operation to obtain the predicted value range of a specified data item as a dynamic detection parameter; When the actual report data detected exceeds the predicted value range, the report risk monitoring model triggers an alarm.

8. A data analysis method based on the intelligent data analysis platform for accounting as described in any one of claims 1 to 6, characterized in that, This includes methods for generating and submitting data analysis requests using natural language, the steps of which include: The accounting intelligent assistant client receives data analysis request descriptions input by users in natural language. The intelligent agent main control module organizes prompt words and submits them to the large language model. The prompt words include user needs, callable function information, and data analysis requirement templates obtained from the knowledge base MCP server. The large language model generates a data analysis requirement document that conforms to the template specification based on the prompt words; The intelligent agent's main control module submits the generated data analysis requirement document to the data analysis system through the analysis system's MCP server.

9. The data analysis method according to claim 8, characterized in that, Before submitting the data analysis requirements document to the data analysis system, a requirements review and iteration process is also included: The generated data analysis requirements document is returned to the client for user review. Based on the user's feedback input in natural language, the requirement description is revised and the data analysis requirement document is regenerated until the user confirms it.

10. A data analysis method based on the intelligent data analysis platform for accounting as described in any one of claims 1 to 6, characterized in that, This includes methods for querying data analysis results using natural language, the steps of which include: The accounting intelligent assistant client receives data analysis result query requests from users, inputted in natural language. The intelligent agent main control module organizes prompt words and submits them to the large language model to obtain a query instruction to call the analysis system MCP server; The intelligent agent main control module calls the analysis system MCP server according to the query command to obtain the corresponding data analysis results from the data analysis system; The acquired data analysis results are returned to the accounting intelligent assistant client for display.

Images