CN-121979971-A - Intelligent data analysis platform and method for accounting

CN121979971ACN 121979971 ACN121979971 ACN 121979971ACN-121979971-A

Abstract

The invention relates to an intelligent data analysis platform and method for accounting, 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, wherein a natural language interaction interface is provided for accounting staff, 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, problems in an accounting report and business are found through the development trend of report data items and the correlation of the data items, the business development situation is predicted through the development trend of business data, and decision support capability is provided for the development of accounting business.

Inventors

FENG LIUJUN
PENG JINSONG
LI BO
ZHU KE
BAO RUI

Assignees

北京南天软件有限公司
云南南天电子信息产业股份有限公司

Dates

Publication Date: 20260505
Application Date: 20251212

Claims (10)

1. The intelligent data analysis platform for the accounting is characterized by comprising a background supporting module, an application service module, an accounting knowledge base, a plurality of MCP servers, an agent main control module and an intelligent assistant client for the accounting; Wherein: the background support module comprises a data analysis system, a monitoring system and a large language model; The application service module comprises an accounting system and a meeting-counting report system which are mutually related; The accounting knowledge base is a vector database and stores accounting knowledge and metadata thereof; a plurality of MCP servers, comprising: The analysis system MCP server is used for packaging and calling a service interface of the data analysis system; the MCP server of the accounting system is used for packaging and calling a service interface of the accounting system; the accounting report system MCP server is used for packaging and calling a service interface of the accounting report system; The monitoring system MCP server is used for transmitting information between the monitoring system and the intelligent agent main control module; A knowledge base MCP server for responding to queries and retrieving information from an accounting knowledge base; the intelligent body main control module is used for controlling a session flow, in particular for receiving an interaction request input by a user through natural language, organizing prompt words containing callable function information, submitting the prompt words to the large language model, receiving an operation instruction returned by the large language model, and calling a corresponding MCP server according to the operation instruction to execute data analysis, risk monitoring or report query operation; And the accounting intelligent assistant client provides a user interaction interface for receiving user input, displaying session results and receiving alarm notification.
2. The intelligent data analysis platform of claim 1, wherein the data analysis system is configured to perform trend analysis and correlation analysis on data from the accounting system and the accounting reporting system and provide a trend analysis application program interface that includes a range of predicted values; The monitoring system is used for monitoring report data generated in the accounting report system, and sending alarm information after judging a report risk event; And the report risk monitoring model in the monitoring system acquires dynamic parameters by calling the trend analysis application program interface.
3. The intelligent data analysis platform for accounting according to claim 1, wherein the raw data on which the report is generated by the accounting system is managed by the accounting system; The content structure description information of various reports is stored in the accounting knowledge base, and comprises semantic descriptions of report data items and corresponding resource names in the accounting report system, wherein the content structure description information is a JSON format text for establishing a mapping relation between the accounting terms and the report data items.
4. The accounting intelligent data analysis platform according to claim 1, wherein the agent host module is integrated with a program execution environment, and 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 a function provided by a corresponding MCP server or the agent host module.
5. The intelligent data analysis platform for accounting according to claim 1, wherein the initial version of the report content structure description information is automatically generated by a report format file of the accounting report system in combination with data dictionary information, and can be updated and stored in an accounting knowledge base after adjustment.
6. The accounting intelligent data analysis platform of claim 1, wherein an accounting intelligent assistant client obtains alarm information from the agent host module by a timed polling mode and carries out popup prompt.
7. A data analysis method based on the intelligent data analysis platform for accounting according to any one of claims 1 to 6, comprising a dynamic parameter supporting method in report risk discovery, comprising the steps of: The data analysis system runs a trend analysis model and deploys an API that provides a range of predicted values; the report risk monitoring model in the monitoring system calls the API in operation to acquire the predicted value range of the appointed data item as a dynamic detection parameter; And when the monitored actual report data exceeds the range of the predicted value, triggering an alarm by the report risk monitoring model.
8. A data analysis method based on the accounting intelligent data analysis platform of any one of claims 1 to 6, characterized by comprising a method of generating and submitting data analysis requirements through natural language, comprising the steps of: receiving data analysis demand description input by a user in natural language through an accounting intelligent assistant client; The intelligent agent main control module organizes prompt words and submits the prompt words to the large language model, wherein the prompt words comprise user requirements, callable function information and a data analysis requirement template acquired from the knowledge base MCP server; the large language model generates a data analysis demand document which accords with the template specification according to the prompt word; And the agent master control module submits the generated data analysis demand document to the data analysis system through the analysis system MCP server.
9. The data analysis method of claim 8, further comprising the step of requirement review and iteration prior to submitting the data analysis requirement document to the data analysis system: Returning the generated data analysis demand document to the client for user examination; And correcting the requirement description and regenerating the data analysis requirement document according to the adjustment opinion input by the user in the natural language until the user confirms.
10. A data analysis method based on the intelligent data analysis platform for accounting according to any one of claims 1 to 6, characterized by comprising a method for querying data analysis results through natural language, comprising the steps of: Receiving a data analysis result query request input by a user in natural language through an accounting intelligent assistant client; the agent main control module organizes prompt words and submits the prompt words to the large language model so as to obtain a query instruction for calling the analysis system MCP server; The agent main control module calls an analysis system MCP server according to the query instruction, and obtains corresponding data analysis results from the data analysis system; and returning the acquired data analysis result to the accounting intelligent assistant client for display.

Description

Intelligent data analysis platform and method for accounting Technical Field The invention belongs to the technical field of accounting, and particularly relates to an intelligent data analysis platform and method for accounting. Background The large data platform of the financial institution has been built for many years, and analysis of accounting business data has also formed a system, but in the aspect of accounting report risk discovery, no systematic supporting capability is formed at present. In addition, the data analysis function of the big data platform can provide effective supporting capability for service development, and the capability in the aspects of service, technology and data science is required to be combined, so that the requirement is difficult to be clearly described only by accountant. The application of accounting personnel to the data analysis function of a big data platform is as follows: 1. In the aspect of report risk discovery, most accountants still rely on experience or analysis reports provided by a big data platform to be manually processed at present, and the report risk discovery is lack of automatic processing and dynamic parameter support in report risk discovery. 2. Accounting personnel have difficulty in proposing a need to be able to have the data analysis capabilities of a large data platform function. 3. The man-machine interface of big data platform data analysis function is not friendly to accounting personnel. Therefore, how to overcome the defects of the prior art is a problem to be solved in the technical field of the current accounting technology. Disclosure of Invention The invention aims to solve the defects in the prior art and provides an intelligent data analysis platform and method for accounting. In order to achieve the above purpose, the technical scheme adopted by the invention is as follows: an intelligent data analysis platform for accounting 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 intelligent assistant client for accounting; Wherein: the background support module comprises a data analysis system, a monitoring system and a large language model; The application service module comprises an accounting system and a meeting-counting report system which are mutually related; The accounting knowledge base is a vector database and stores accounting knowledge and metadata thereof; a plurality of MCP servers, comprising: The analysis system MCP server is used for packaging and calling a service interface of the data analysis system; the MCP server of the accounting system is used for packaging and calling a service interface of the accounting system; the accounting report system MCP server is used for packaging and calling a service interface of the accounting report system; The monitoring system MCP server is used for transmitting information between the monitoring system and the intelligent agent main control module; A knowledge base MCP server for responding to queries and retrieving information from an accounting knowledge base; the intelligent body main control module is used for controlling a session flow, in particular for receiving an interaction request input by a user through natural language, organizing prompt words containing callable function information, submitting the prompt words to the large language model, receiving an operation instruction returned by the large language model, and calling a corresponding MCP server according to the operation instruction to execute data analysis, risk monitoring or report query operation; And the accounting intelligent assistant client provides a user interaction interface for receiving user input, displaying session results and receiving alarm notification. Further, it is preferable that the data analysis system is configured to perform trend analysis and correlation analysis on data originating from the accounting system and the accounting system, and provide a trend analysis application program interface (i.e., AP) containing a range of predicted values; The monitoring system is used for monitoring report data generated in the accounting report system, and sending alarm information after judging a report risk event; And the report risk monitoring model in the monitoring system acquires dynamic parameters by calling the trend analysis application program interface. Further, it is preferable that the raw data on which the report is generated by the accounting system is managed by the accounting system; The content structure description information of various reports is stored in the accounting knowledge base, and comprises semantic descriptions of report data items and corresponding resource names in the accounting report system, wherein the content structure description information is a JSON format text for establishing a mapping relation between the accounting terms and the report