AI-based automated financial management system, method, equipment and medium
By introducing AI technology into the financial management system, automatically collecting and processing financial data and conducting intelligent audits, the existing financial management system has solved the problems of high labor costs and cumbersome operations, and efficient and convenient financial management has been achieved.
Patent Information
- Application Number
- CN202411991680.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-31
- Publication Date
- 2025-04-18
AI Technical Summary
The existing financial management system has problems such as high labor costs, cumbersome operations and high investment costs.
It adopts an automated financial management system based on AI, including data acquisition module, data processing module, intelligent audit module and data management module. The system reduces manual operations by automatically collecting enterprise data, performing pre-processing and intelligent auditing.
It realizes automated financial management, reduces labor costs and operational complexity, and improves the efficiency and convenience of financial accounting.
Smart Images
Figure CN119850355A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the field of data management technology, and specifically relates to an AI-based automated financial management system, method, device and medium. Background Art
[0002] Traditional financial software requires too much data to be manually entered by financial personnel, and can only complete basic accounting and reporting work, and it is difficult to provide valuable data for the company's refined management. If the business-finance integration solution is adopted, the company will be required to invest a lot of money and manpower to complete the information construction of other departments. This process is often abandoned halfway due to insufficient funds and insufficient management.
[0003] Therefore, the existing financial management has problems such as high labor costs, complicated operations, and high investment costs. Summary of the invention
[0004] The purpose of the present invention is to provide an AI-based automated financial management system, method, device and medium to solve the problems of high labor cost, complicated operation and high investment cost in existing financial management.
[0005] In order to achieve the above object, the present invention adopts the following technical solutions: In a first aspect, the present invention provides an AI-based automated financial management system, the system comprising: The data collection module is used to automatically collect the original external data of the enterprise based on the data collection tool, identify the business type of the original external data of the enterprise based on the pre-built AI model, and obtain various business documents of the enterprise; The data processing module is used to pre-process various business documents of the enterprise to obtain various business documents after pre-processing, wherein the pre-processing at least includes: redundant processing and marking processing; The intelligent audit module is used to perform intelligent audit processing on various types of pre-processed business documents to obtain audit results. The intelligent audit processing at least includes: AI intelligent form filling, form review, voucher generation and reconciliation; The data management module is used to manage financial and tax data, various business documents and audit results. The financial and tax data at least includes: basic file data, contract data, project data, production line data, asset data, general ledger data, report data, cashier data, capital data, salary data and tax data.
[0006] Preferably, the system further comprises: a data storage module for storing financial data, various business documents and audit results.
[0007] Preferably, the data collection tools include at least: mini-programs, APPs, crawlers and RPA.
[0008] Preferably, the system is constructed using a B / S structure.
[0009] Preferably, the intelligent audit module includes: The graph parsing submodule is used to identify the content of various types of pre-processed business documents based on a preset recognition algorithm to obtain document content data; The document parsing submodule is used to parse the document content data based on the AI semantic parsing model to obtain the parsed document content data; The audit rule submodule is used to match the preset audit rules, audit the parsed document content data based on the audit rules, and obtain the audit results.
[0010] In a second aspect, the present invention provides an AI-based automated financial management method, which is implemented by the above-mentioned AI-based automated financial management system, and the method includes: Obtain the original external data of the enterprise; Based on the pre-built AI model, the business type of the original external data of the enterprise is identified to obtain various business documents of the enterprise; Preprocessing various business documents of the enterprise to obtain various preprocessed business documents, wherein the preprocessing at least includes: redundant processing and marking processing; Perform intelligent audit processing on various pre-processed business documents to obtain audit results.
[0011] Preferably, intelligent audit processing is performed on various types of pre-processed business documents to obtain audit results, including: Based on the preset recognition algorithm, the content of various types of pre-processed business documents is recognized to obtain the document content data; Parse the document content data based on the AI semantic parsing model to obtain the parsed document content data; Match the preset audit rules, audit the parsed document content data based on the audit rules, and obtain the audit results.
[0012] Preferably, the audit result includes: audit passed or audit failed; When the audit result is that the audit fails, the method further includes: re-examining the parsed document content data that has passed the audit for a second time, and generating an audit failure report when the second re-examination fails; When the audit result is audit passed or second re-examination passed, the method further includes: generating a cashier report.
[0013] In a third aspect, the present invention provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned AI-based automated financial management method when executing the computer program.
[0014] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-mentioned AI-based automated financial management method.
[0015] Beneficial effects: 1. The data acquisition module of the present invention can automatically collect the original external data of the enterprise, reduce the human cost of the enterprise, and improve the efficiency of the enterprise; 2. The intelligent audit module of the present invention can automatically perform intelligent audit processing on various business documents, such as AI intelligent form filling, form review, voucher generation and reconciliation, etc., which can help enterprises to perform financial accounting work with zero operation, thereby reducing workload and improving the convenience of auditing. BRIEF DESCRIPTION OF THE DRAWINGS
[0016] The accompanying drawings are used to provide a further understanding of the embodiments of the present invention and constitute a part of the specification. Together with the following specific implementations, they are used to explain the embodiments of the present invention, but do not constitute a limitation on the embodiments of the present invention. In the accompanying drawings: Figure 1 is a block diagram of an AI-based automated financial management system provided by one embodiment of the present invention; Figure 2 It is a flowchart of an AI-based automated financial management method provided by one embodiment of the present invention. DETAILED DESCRIPTION
[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the present invention will be briefly introduced below in combination with the drawings and the description of the embodiments or the prior art. Obviously, the following description of the structure of the drawings is only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. It should be noted that the description of these embodiments is used to help understand the present invention, but does not constitute a limitation of the present invention.
[0018] Embodiment 1 Figure 1 is a block diagram of an AI-based automated financial management system provided by one embodiment of the present invention. Figure 1As shown, this embodiment provides an AI-based automated financial management system, which is built on the B / S (Browser / Server) structure. The B / S structure is a network architecture model that divides the tasks of the application into two parts: the browser and the server. In this structure, users access server resources on the network through a browser to achieve data interaction and operation; the system can support pure private deployment, public cloud deployment, and private cloud deployment client maintenance-free, making the distribution of the system relatively centralized, which is conducive to the operation and maintenance of the system and has good scalability and flexibility; the system includes the following software function modules: data acquisition module, data processing module, intelligent audit module, and data management module.
[0019] Among them, the data collection module is used to automatically collect the original external data of the enterprise based on the data collection tool, identify the business type of the original external data of the enterprise based on the pre-built AI model, and obtain various business documents of the enterprise.
[0020] In this embodiment, the original external data includes but is not limited to bank and third-party payment channel transaction records, exchange rates, raw material prices on the procurement platform, and reimbursement forms, payment orders, contracts and other data generated by reimbursement, payment requests, and requests for instructions during the business process. These data can be text data or image data.
[0021] In this embodiment, the AI (Artificial Intelligence) model can be Naive Bayes, Support Vector Machine (SVM), Random Forest, Neural Network, etc. The data collection tools mainly include Mini Programs, APPs (Applications), crawlers and RPAs (Robotic Process Automation). RPA robots use user interfaces to capture data and manipulate applications. Mini Programs and APPs support employees in reimbursement, payment requests, and requests for instructions during business operations. Simple actions such as taking photos and uploading can automatically complete tedious processes such as filling out forms.
[0022] In this embodiment, the business system document data dictionary structure and 5-10 document styles are uploaded in advance as sample data, and the AI model is trained using the sample data to obtain a trained AI model. The trained AI model can automatically identify various types of electronic and paper documents of business systems in the future, completely avoiding the traditional requirement of having to deploy business systems in all directions and undergoing lengthy docking and development work with the financial system after the deployment of the system. The AI model is deployed and operated in an off-network localized manner. If the customer chooses to operate in SaaS mode, the data transmission process will unpack, shuffle, and asymmetrically encrypt the data before transmitting it. The server will decrypt, sort, and merge the data before starting processing, thereby ensuring data security.
[0023] The data processing module is used to pre-process various business documents of the enterprise to obtain various business documents after pre-processing, and the pre-processing at least includes: redundant processing and marking processing; In this embodiment, in order to cater to the rigid process, the traditional financial system artificially adds documents that are not business-related in nature to the system or even offline. For example, a purchase contract requires payment after receiving the goods; in order to complete the entire process within the enterprise, the enterprise will be required to have a [receipt], and then generate a [purchase payment application form] based on the [receipt], and then form an [approval record form] after the approval process according to this form. After the finance department receives the payment application form, it will successively add [payables], [payment orders], [vouchers], etc. Each additional document brings a series of problems and workloads such as redundant data, process control, data checking, and system development.
[0024] This embodiment eliminates redundancy in various business documents of the enterprise, and only retains the delivery notes sent by suppliers and the payment records of the finance department, which can truly represent the business entity.
[0025] Secondly, if different departments or positions need to express opinions on this business or add data dimensions, they will be concentrated in one approval process. Each link in the approval process can and must mark the current business document, and the system will record the complete marking process; therefore, it ensures the unique authority of business data and meets the requirements of multi-dimensional data statistics. Maintaining the unique authority of data, the above method is conducive to reducing repeated review work in the company's internal management processes, and is also conducive to reducing the workload of system development and maintenance; allowing each necessary position to mark the data can form a data basis for data statistics in various dimensions in the later stage, meeting the management needs of the enterprise.
[0026] Among them, the intelligent audit module is used to perform intelligent audit processing on various types of business documents after preprocessing to obtain audit results. The intelligent audit processing at least includes: AI intelligent form filling, form review, voucher generation and reconciliation; specifically, the intelligent audit module includes: a graph parsing submodule, a document parsing submodule and an audit rule submodule; the graph parsing submodule is used to perform content recognition on various types of business documents after preprocessing based on a preset recognition algorithm to obtain document content data; the document parsing submodule is used to parse the document content data based on the AI semantic parsing model to obtain the parsed document content data; the audit rule submodule is used to match the preset audit rules, and audit the parsed document content data based on the audit rules to obtain the audit results.
[0027] AI smart form filling: Taking expense reimbursement as an example, after the user takes a photo or uploads the invoice file, the system uses preset recognition algorithms such as OCR (Optical Character Recognition) or view big models to automatically identify the invoice content. The invoice type, service content, identity of the person handling the matter and other information are comprehensively judged by the big model, and one of the fixed reimbursement income and expenditure items is selected to automatically fill in the reimbursement form. All other information including amount, tax rate, etc. is automatically filled in to realize AI smart form filling. Among them, the OCR recognition process uses a self-built OCR model, and ensures the recognition accuracy through four steps: framework recognition, structure recognition, content recognition, and coordinate positioning re-analysis.
[0028] Document review: By setting the data dictionary of the documents to be reviewed in advance, the recognition accuracy of OCR or view large model is ensured, and the recognized structured data is pushed to the rule engine. The rule engine has already set the audit rules required by the enterprise for each business scenario. After receiving the specific data, it will be reviewed one by one according to the audit rules. The comparison between structured information will be judged by the traditional rule engine, and the semantic model and AGI (Artificial General Intelligence) (AGI) will be called for judgment if it involves semantic understanding, and finally a complete document review report will be formed.
[0029] Voucher generation: Since AI intelligent form filling has been implemented in the early stage, the document information such as relevant income and expenditure items + departments + personnel + projects + product lines is sufficient to accurately determine the accounting subjects, so vouchers can be automatically generated from business documents; coupled with automatic templates such as profit and loss carryforward and tax accrual, the entire accounting can be automatically generated.
[0030] Reconciliation: Receipts from corporate banks and third-party transaction platforms (WeChat, Alipay, UnionPay) are automatically obtained through RPA and crawlers, and automatically checked with the reimbursement documents through preset schemes such as the other party's account, transaction amount, date range, previous payment scheduling results, summary AI semantic understanding comparison, etc. (Note that it is not a voucher, but a payment business document initiated through the WeChat applet). After successful checking, the relevant payment voucher is automatically generated, and 100% accurate automatic bank reconciliation is completed at the same time as the voucher is generated.
[0031] The intelligent audit module of the present invention can automatically perform intelligent audit processing on various business documents, such as AI intelligent form filling, form review, voucher generation and account reconciliation, etc., which can help enterprises to perform financial accounting work with zero operation, thereby reducing workload and improving the convenience of auditing.
[0032] The data management module is used to manage financial and tax data, various business documents and audit results. The financial and tax data at least includes: basic file data, contract data, project data, production line data, asset data, general ledger data, report data, cashier data, capital data, salary data and tax data.
[0033] As a further optimization of this embodiment, the system also includes: a data storage module for storing financial data, various business documents and audit results; wherein the data storage module can use databases such as PGSQL, MySQL, Redis, Milvus, etc.
[0034] The present invention can help enterprises to perform financial accounting work without any operation, reduce the labor cost of enterprises, and improve the efficiency of enterprises; Financial data comes with its own product lines and project structures, which can be used to guide daily operations. Enterprises with weak information technology foundations and weak financial internal control awareness can quickly achieve good results. It seamlessly connects online and offline businesses, and there are no blind spots in financial data.
[0035] Embodiment 2 Figure 2 : is a flow chart of an AI-based automated financial management method provided by one embodiment of the present invention. Figure 2 As shown, this embodiment provides an AI-based automated financial management method, which is implemented by the AI-based automated financial management system in Embodiment 1, and includes: Step S10: Obtain the original external data of the enterprise. This embodiment mainly uses data collection tools such as mini-programs, APPs (Applications), crawlers, and RPA (Robotic Process Automation) to collect the original external data of the enterprise. The original external data includes but is not limited to bank and third-party payment channel transaction records, exchange rates, raw material prices on the procurement platform, and reimbursement forms, payment forms, contracts, and other data generated by reimbursement, payment requests, and instructions during the business process. These data can be text data or image data.
[0036] Step S20: Identify the business type of the original external data of the enterprise based on the pre-built AI model to obtain various business documents of the enterprise.
[0037] In this embodiment, the AI (Artificial Intelligence) model can be Naive Bayes, Support Vector Machine (SVM), Random Forest, Neural Network, etc. This embodiment requires uploading the business system document data dictionary structure and 5-10 document styles as sample data in advance, and using the sample data to train the AI model to obtain a trained AI model. The trained AI model can automatically identify various types of electronic and paper documents of business systems in the future.
[0038] Step S30: Preprocess various types of business documents of the enterprise to obtain various types of preprocessed business documents, wherein the preprocessing at least includes: redundancy elimination processing and labeling processing; redundancy elimination processing is conducive to reducing repeated review work of internal management processes of the enterprise, and is also conducive to reducing the workload of system development and maintenance; labeling processing mainly allows each necessary position to label the data, which can form a data basis for data statistics in various dimensions in the later stage to meet the management needs of the enterprise.
[0039] Step S40: Perform intelligent audit processing on various types of pre-processed business documents to obtain audit results.
[0040] In this embodiment, intelligent audit processing is performed on various types of pre-processed business documents to obtain audit results, including: Step S401: performing content recognition on various types of pre-processed business documents based on a preset recognition algorithm to obtain document content data; wherein the preset recognition algorithm may be OCR (Optical Character Recognition) or a view macro model; For example, taking expense reimbursement as an example, after the user takes a photo or uploads the invoice file, the system uses preset recognition algorithms such as OCR (Optical Character Recognition) or view large models to automatically identify the invoice content, including: invoice type, service content, identity of the person in charge, and other information.
[0041] Step S402: Parse the document content data based on the AI semantic parsing model to obtain parsed document content data.
[0042] Step S403: matching preset audit rules, auditing the parsed document content data based on the audit rules, and obtaining audit results.
[0043] The identified structured data is pushed to the rule engine, which has pre-set the audit rules required by the enterprise for each business scenario. After receiving the specific data, it will be reviewed one by one according to the audit rules. The comparison between structured information will be judged by the traditional rule engine, and the semantic model and AGI (Artificial General Intelligence) (AGI) (Artificial General Intelligence) will be called for judgment if it involves semantic understanding, and finally a complete audit report will be formed.
[0044] In this embodiment, the audit results include: audit passed or audit failed; When the audit result is that the audit fails, the method further includes: re-examining the parsed document content data that has passed the audit for a second time, and generating an audit failure report when the second re-examination fails; When the audit result is audit passed or second re-examination passed, the method further includes: generating a cashier report.
[0045] In this embodiment, the cashier report contains the specific cashier payment amount, and then a voucher is generated based on the cashier report; in the process of generating the voucher, the receipts from the corporate bank and third-party transaction platforms (WeChat, Alipay, UnionPay) are automatically obtained through RPA and crawlers, and are automatically checked with the billing document through preset schemes such as the other party's account, transaction amount, date range, previous payment scheduling results, and summary AI semantic understanding comparison (note that it is not a voucher, but a payment business document initiated through the WeChat applet). After the check is successful, the relevant payment voucher is automatically generated, and 100% accurate automatic bank reconciliation is completed at the same time as the voucher is generated.
[0046] The data acquisition module of the present invention can automatically collect the original external data of the enterprise, reduce the human cost of the enterprise and improve the efficiency of the enterprise; and through the intelligent audit module, various business documents can be automatically audited and processed intelligently, such as AI intelligent form filling, form review, voucher generation and reconciliation, etc., which can help the enterprise's financial accounting work with zero operation, thereby reducing workload and improving the convenience of auditing.
[0047] Embodiment 3 This embodiment also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the AI-based automated financial management method in Embodiment 2 when executing the computer program.
[0048] This embodiment also provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the AI-based automated financial management method in Embodiment 2 is implemented.
[0049] The data acquisition module of the present invention can automatically collect the original external data of the enterprise, reduce the human cost of the enterprise and improve the efficiency of the enterprise; and through the intelligent audit module, various business documents can be automatically audited and processed intelligently, such as AI intelligent form filling, form review, voucher generation and reconciliation, etc., which can help the enterprise's financial accounting work with zero operation, thereby reducing workload and improving the convenience of auditing.
[0050] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.
[0051] The present application is described 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 application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0052] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.
Claims
1. An AI-based automated financial management system, characterized in that: The system comprises: The data collection module is used to automatically collect the original external data of the enterprise based on the data collection tool, identify the business type of the original external data of the enterprise based on the pre-built AI model, and obtain various business documents of the enterprise; The data processing module is used to pre-process various business documents of the enterprise to obtain various business documents after pre-processing, wherein the pre-processing at least includes: redundant processing and marking processing; The intelligent audit module is used to perform intelligent audit processing on various types of pre-processed business documents to obtain audit results. The intelligent audit processing at least includes: AI intelligent form filling, form review, voucher generation and reconciliation; The data management module is used to manage financial and tax data, various business documents and audit results. The financial and tax data at least includes: basic file data, contract data, project data, production line data, asset data, general ledger data, report data, cashier data, capital data, salary data and tax data.
2. The AI-based automated financial management system according to claim 1, characterized in that: The system also includes: a data storage module for storing financial data, various business documents and audit results.
3. The AI-based automated financial management system according to claim 1, characterized in that: The data collection tools include at least: mini-programs, APPs, crawlers and RPA.
4. The AI-based automated financial management system according to claim 1, characterized in that: The system is built using a B / S structure.
5. The AI-based automated financial management system according to claim 1, characterized in that: The intelligent audit module includes: The graph parsing submodule is used to identify the content of various types of pre-processed business documents based on a preset recognition algorithm to obtain document content data; The document parsing submodule is used to parse the document content data based on the AI semantic parsing model to obtain the parsed document content data; The audit rule submodule is used to match the preset audit rules, audit the parsed document content data based on the audit rules, and obtain the audit results.
6. An AI-based automated financial management method, the method being implemented by the AI-based automated financial management system according to any one of claims 1 to 5, characterized in that: The method comprises: Obtain the original external data of the enterprise; Based on the pre-built AI model, the business type of the original external data of the enterprise is identified to obtain various business documents of the enterprise; Preprocessing various business documents of the enterprise to obtain various preprocessed business documents, wherein the preprocessing at least includes: redundant processing and marking processing; Perform intelligent audit processing on various pre-processed business documents to obtain audit results.
7. The AI-based automated financial management method according to claim 6, characterized in that: Perform intelligent audit processing on various business documents after pre-processing to obtain audit results, including: Based on the preset recognition algorithm, the content of various types of pre-processed business documents is recognized to obtain the document content data; Parse the document content data based on the AI semantic parsing model to obtain the parsed document content data; Match the preset audit rules, audit the parsed document content data based on the audit rules, and obtain the audit results.
8. The AI-based automated financial management method according to claim 7, characterized in that: The audit results include: audit passed or audit failed; When the audit result is that the audit fails, the method further includes: re-examining the parsed document content data that has passed the audit for a second time, and generating an audit failure report when the second re-examination fails; When the audit result is audit passed or second re-examination passed, the method further includes: generating a cashier report.
9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the AI-based automated financial management method described in any one of claims 6 to 8 is implemented.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the AI-based automated financial management method described in any one of claims 6 to 8 is implemented.