Financial planning method and system based on data driving and medium

Abstract

The invention discloses a financial planning method and system based on data driving and a medium, and the method comprises the steps: automatically executing a financial process through an automatic execution script, and obtaining the financial information of an enterprise; counting historical financial data, preprocessing the historical financial data, and constructing a training set and a verification set, the historical financial data including reimbursement data, loan and repayment data, and profit data; obtaining financial data vectors, extracting key feature vectors in the financial data vectors, and constructing a neural network model; determining a loss function based on prediction data output by the neural network model, and adjusting model parameters according to the loss function until the neural network model converges; outputting predicted financial data through a convergent neural network model based on the latest financial data; and planning enterprise finance according to the predicted financial data. Through collection, arrangement and analysis of financial data, accurate and comprehensive information can be provided for enterprises, so that decision makers can make decisions based on objective data.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and specifically to a data-driven financial planning method, system and medium. Background Art

[0002] With the rapid development of social economy, the scale of enterprises is constantly expanding, the business of enterprises is becoming more complex, the amount of business data is increasing, and financial data tends to become scattered and diversified, which greatly increases the complexity of data integration and management.

[0003] Traditional corporate financial management methods often have problems such as scattered data, delayed updates, and complex analysis, which lead to inefficient decision-making and increased risks.

[0004] With the vigorous development of the social economy, the scale of enterprises continues to expand, and their business landscape becomes increasingly complex, accompanied by a sharp increase in the amount of business data and a significant increase in the trend of diversification. This change not only enriches the information dimension of the enterprise, but also greatly increases the challenge and complexity of data integration and management. Financial data, as the core lifeblood of enterprise operations, is now more dispersedly stored in various systems and platforms, and its form and structure are also diversified, which undoubtedly puts more stringent requirements on the data processing capabilities, integration efficiency and management level of the enterprise. Therefore, how to efficiently and accurately integrate these complex data resources and build a unified and clear data view has become an indispensable key link in modern enterprise financial management and even the overall operation strategy. Summary of the invention

[0005] In order to solve the above problems, this application proposes a data-driven financial planning method, including:

[0006] Automate financial processes and obtain financial information of enterprises through automated execution of scripts;

[0007] Based on the financial information, historical financial data are collected and pre-processed according to preset financial indicators to obtain standard financial data, and a training set and a validation set are constructed based on the standard financial data, wherein the historical financial data includes reimbursement data, loan repayment data, and profit data;

[0008] Performing vector conversion on the standard financial data to obtain a financial data vector, extracting key feature vectors from the financial data vector according to financial planning requirements, and constructing a neural network model based on the key feature vectors and a preset neural network structure;

[0009] Inputting the training set into the neural network model, determining a loss function of the neural network model based on the prediction data output by the neural network model, adjusting model parameters of the neural network model according to the loss function, and verifying the neural network model through the verification set until the neural network model converges;

[0010] Collecting the latest financial data, and outputting predicted financial data based on the latest financial data through the converged neural network model;

[0011] Based on the forecasted financial data, corporate finances are planned.

[0012] In one implementation of the present application, financial planning needs are obtained, and a needs analysis is performed on the financial planning needs to determine that the financial planning needs include reimbursement management planning, debt management planning, and profit management planning; corresponding key features are determined according to the financial planning needs, and based on the key features, corresponding key feature vectors are extracted from the financial data vector.

[0013] In one implementation of the present application, based on the reimbursement management plan, the key features of the reimbursement data are determined to include reimbursement amount, reimbursement category, reimbursement time, and budget execution; based on the debt management plan, the key features of the loan and repayment data are determined to include repayment amount, repayment time, loan amount, loan term, and interest rate; based on the profit management plan, the key features of the profit data are determined to include cash flow, profit margin, and debt-to-asset ratio.

[0014] In one implementation of the present application, the neural network model is trained based on the reimbursement data, the loan and repayment data, and the profit data, respectively; the historical reimbursement amount in the reimbursement data, the reimbursement type corresponding to the historical reimbursement amount, and the historical approval time are input into the neural network model to obtain the predicted reimbursement amount and predicted approval time output by the neural network model; the historical loan and repayment period, actual loan and repayment date, and loan and repayment amount in the loan and repayment data are input into the neural network model to obtain the borrowing trend and repayment punctuality rate output by the neural network model; the historical cash flow and historical profit in the profit data are input into the neural network model to obtain the predicted profit trend and predicted risk output by the neural network model; the error between the predicted data and the actual data output by the neural network model is determined, and the loss function of the neural network model is determined based on the error.

[0015] In one implementation of the present application, the business to which the financial process belongs is determined by automatically executing a script; based on the business, the access data source is determined, and the interface corresponding to the data source is connected to obtain financial information.

[0016] In one implementation of the present application, the latest financial data is input into the convergent neural network model, and based on the latest reimbursement data in the latest financial data, the future short-term reimbursement amount output by the convergent neural network model, the reimbursement type corresponding to the future short-term reimbursement amount, and the future approval time are obtained; based on the latest borrowing and repayment data in the latest financial data, the future short-term borrowing amount and the future short-term repayment rate are obtained; based on the latest profit data in the latest financial data, the future short-term profit trend and future short-term forecast risk are obtained.

[0017] In one implementation of the present application, based on the future approval time, the future short-term reimbursement amount within a preset time period and the reimbursement type corresponding to the future short-term reimbursement amount are counted, and the reimbursement frequency of different reimbursement types within the preset time period is determined respectively; based on the future short-term loan amount, the capital demand and liquidity status of the enterprise are determined, and based on the future short-term repayment rate, the debt repayment ability and credit status of the enterprise are determined; based on the future short-term profit trend, the profitability and operating efficiency of the enterprise are determined, and based on the future short-term predicted risk, the financial risk of the enterprise is determined.

[0018] In one implementation of the present application, based on the reimbursement frequency, the high-frequency reimbursement type within the preset time period is determined, and the reimbursement corresponding to the high-frequency reimbursement type is automatically reviewed to reduce the approval time; based on the capital demand, the liquidity status, the debt repayment ability and the credit status, a capital plan and liquidity management strategy are formulated; based on the profitability and the operational efficiency, an operational strategy is formulated; based on the financial risk, a risk threshold is set, and potential risks are identified through the risk threshold.

[0019] On the other hand, an embodiment of the present invention further provides a large-scale building energy management system, characterized in that the system comprises:

[0020] Automate financial processes and obtain financial information of enterprises through automated execution of scripts;

[0021] Based on the financial information, historical financial data are collected and pre-processed according to preset financial indicators to obtain standard financial data, and a training set and a validation set are constructed based on the standard financial data, wherein the historical financial data includes reimbursement data, loan repayment data, and profit data;

[0022] Performing vector conversion on the standard financial data to obtain a financial data vector, extracting key feature vectors from the financial data vector according to financial planning requirements, and constructing a neural network model based on the key feature vectors and a preset neural network structure;

[0023] Inputting the training set into the neural network model, determining a loss function of the neural network model based on the prediction data output by the neural network model, adjusting model parameters of the neural network model according to the loss function, and verifying the neural network model through the verification set until the neural network model converges;

[0024] Collecting the latest financial data, and outputting predicted financial data based on the latest financial data through the converged neural network model;

[0025] Based on the forecasted financial data, corporate finances are planned.

[0026] On the other hand, the present application also proposes a non-volatile computer storage medium storing computer executable instructions, wherein the computer executable instructions are configured as: a data-driven financial planning method as described in the above example.

[0027] This application proposes a data-driven financial planning method that can bring the following beneficial effects:

[0028] By collecting, organizing and analyzing financial data, we can provide enterprises with accurate and comprehensive information, so that decision makers can make decisions based on objective data and reduce subjective assumptions and decision-making risks. This helps enterprises make more scientific and reasonable judgments when facing a complex and changing market environment.

[0029] Data analysis can reveal the laws and trends behind financial data, helping companies to identify potential market opportunities and risks in a timely manner. This foresight enables companies to plan ahead, seize opportunities, and avoid risks, thereby improving their competitiveness and market adaptability. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0031] Figure 1 A flowchart of a data-driven financial planning method in an embodiment of the present application;

[0032] Figure 2 A schematic diagram of an enterprise-level financial planning framework in an embodiment of the present application;

[0033] Figure 3 This is a schematic diagram of the operation flow of a data-driven financial planning system in an embodiment of the present application;

[0034] Figure 4This is a schematic diagram of a data-driven financial planning system in an embodiment of the present application. DETAILED DESCRIPTION

[0035] In order to make the purpose, technical solution and advantages of the present application clearer, the technical solution of the present application will be clearly and completely described below in combination with the specific embodiments of the present application and the corresponding drawings. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present application.

[0036] The technical solutions provided by various embodiments of the present application are described in detail below in conjunction with the accompanying drawings.

[0037] like Figure 1 As shown, the embodiment of the present application provides a data-driven financial planning method, including:

[0038] S101: Automate financial processes and obtain the company's financial information by automating script execution.

[0039] Specifically, through automated script writing, business logic rule configuration, approval flow configuration, intelligent verification configuration and API interface calling are realized to automatically execute financial processes, including travel expense reimbursement, general business reimbursement, loan business, repayment business, business application, accounts receivable and payable business, etc.

[0040] Furthermore, by automatically executing scripts, we can determine the business to which the financial process belongs, determine the access data source based on the business, connect to the interface corresponding to the data source, acquire, process and store a large amount of financial information in real time, and generate standard format reports (the format can be PDF, Excel, Word, etc.) in combination with pre-configured table templates to automate the financial process. This not only saves time and resource costs, but also reduces the risk of human errors, ensures the accuracy and consistency of financial reports, and provides a reliable financial management foundation for enterprises.

[0041] S102: Based on the financial information, historical financial data are collected and pre-processed according to preset financial indicators to obtain standard financial data. Based on the standard financial data, a training set and a validation set are constructed. The historical financial data include reimbursement data, loan and repayment data, and profit data.

[0042] Specifically, the historical financial data contained in the statistical financial information is selected based on the preset financial indicators to cover financial data at different time points and business scenarios. The historical financial data includes reimbursement data, loan repayment data, and profit data. The statistical historical financial data is preprocessed, including data cleaning, conversion, and normalization processes, cleaning up noise and outliers in the data, filling in missing values, and standardizing the data format so that the data conforms to a unified data format and standard. Through this standardized processing, the consistency and comparability of the data can be ensured, which facilitates the subsequent data comparison and analysis work of the enterprise, thereby providing accurate financial information and decision support for enterprise managers.

[0043] Furthermore, after preprocessing is completed, standard financial data is obtained, and the standard financial data is divided into a training set and a validation set.

[0044] In the embodiment of the present application, the financial data of various departments of the enterprise can also be automatically integrated, including data from accounting systems, ERP systems, financial software, etc. These data cover the company's income, expenditures, assets, liabilities, cash flow, etc., including reimbursement data (general, travel, business entertainment, etc.), loan application repayment data, financial statements, etc.; various operational data of the enterprise are obtained from the enterprise ERP system; statements and capital flow data obtained from the bank; and data provided by the tax bureau, suppliers, and customers. The data collection method includes directly extracting detailed data such as the reimbursement amount, reimbursement time, expense items, itinerary details, reasons for business trips, etc. from the database of the financial sharing platform, or automatically collecting data from different systems through an API interface.

[0045] S103: Perform vector conversion on the standard financial data to obtain a financial data vector, extract key feature vectors from the financial data vector according to financial planning requirements, and construct a neural network model based on the key feature vectors and a preset neural network structure.

[0046] Specifically, the standard financial data is vectorized to obtain the financial data vector, the financial planning needs are obtained, and the financial planning needs are analyzed to determine that the financial planning needs include reimbursement management planning, debt management planning, and profit management planning. According to different financial planning needs, the corresponding key features are determined, and dimensionality reduction processing is performed to reduce the complexity of the data and improve the efficiency of the model.

[0047] Among them, according to the reimbursement management plan, the key features of reimbursement data include reimbursement amount, reimbursement category, reimbursement time, and budget execution; according to the debt management plan, the key features of loan and repayment data include repayment amount, repayment time, loan amount, loan term, and interest rate; according to the profit management plan, the key features of profit data include cash flow, profit margin, and debt-to-asset ratio.

[0048] Furthermore, based on the key feature vectors, the number of neurons in the input layer, hidden layer and output layer of the neural network structure is set, and the activation function and loss function are defined to construct a neural network model.

[0049] In the embodiment of the present application, the financial planning needs include reimbursement management planning, debt management planning and profit management planning. For example, for reimbursement management planning, it is necessary to pay attention to the various expenses to be reimbursed, including the control of the reimbursement amount, the rationality of the expense category and the implementation of the budget. By formulating a detailed budget plan, the enterprise can predict and plan various reimbursement expenses, ensure that the reimbursement meets the budget requirements, and reduce unnecessary expenses. It is also necessary to include the measurement of the effect of reimbursement and long-term planning. The efficiency and effectiveness of the reimbursement process are evaluated by setting key performance indicators (KPIs), such as approval time, compliance rate and cost savings. Generate reimbursement-related reports regularly, analyze the use of expenses and approval efficiency, and optimize the reimbursement process based on the analysis results. At the same time, long-term planning should take into account future reimbursement needs and financial goals, and formulate corresponding strategies to improve the overall efficiency of financial management and support the long-term development and financial health of the enterprise.

[0050] In addition, in the embodiment of the present application, the financial planning requirements also include risk assessment and financial forecasting. Risk assessment should cover the review of the compliance of reimbursement and identify possible non-compliance risks, such as non-compliant reimbursements or over-budget expenses. Financial forecasts need to be based on historical reimbursement data to predict future expense trends and ensure the accuracy and rationality of the budget. At the same time, formulate comprehensive reimbursement policies and processes, including approval processes, expense review standards and reimbursement policies, to ensure the standardization and transparency of reimbursement operations.

[0051] S104: Input the training set into the neural network model, determine the loss function of the neural network model based on the prediction data output by the neural network model, adjust the model parameters of the neural network model according to the loss function, and verify the neural network model through the verification set until the neural network model converges.

[0052] Specifically, the neural network model is trained by the training set, and the neural network model is trained based on the reimbursement data, loan repayment data, and profit data in the training set, and the data is input into the neural network model. The neural network model is trained according to Formula 1: (l) =W (l) a (l-1) +b (l) ,a (l) =σ(z (l) ) to perform forward propagation, thereby outputting the prediction result, where z (l) is the weighted input of layer l, a (l)is the activation output of layer l, W (l) and b (l) are the weight and bias of the lth layer respectively, and σ is the activation function.

[0053] Furthermore, through formula 2: δ (L) =((W (l+1) ) T δ (l+1) ⊙σ′(z (l) ), Entering back propagation, where δ (l) is the error term of the lth layer, σ′(z (l) ) is the derivative of the activation function, through the formula Determine the loss function, where L is the loss function of a single sample, y (i) is the actual output, is the predicted output, m is the number of samples, and the model parameters are updated according to the gradient of the loss function to minimize the loss.

[0054] The historical reimbursement amount, the reimbursement type corresponding to the historical reimbursement amount, and the historical approval time in the reimbursement data are input into the neural network model to obtain the predicted reimbursement amount and predicted approval time output by the neural network model. The historical loan and repayment period, actual loan and repayment date, and loan and repayment amount in the loan and repayment data are input into the neural network model to obtain the borrowing trend and repayment punctuality rate output by the neural network model. The historical cash flow and historical profit in the profit data are input into the neural network model to obtain the predicted profit trend and predicted risk output by the neural network model.

[0055] Furthermore, after each model parameter update, the model loss function is calculated and the model capability is evaluated through the validation set to check its generalization ability. When the loss function value gradually decreases and tends to be stable, the performance of the model reaches a stable level, indicating that it has converged.

[0056] S105: Collect the latest financial data, and output predicted financial data based on the latest financial data through the converged neural network model.

[0057] Specifically, in the process of automatically executing financial processes, the latest financial data is collected and input into a convergent neural network model, so that the model outputs corresponding prediction data based on the latest reimbursement data, the latest loan and repayment data, and the latest profit data.

[0058] More specifically, based on the latest reimbursement data in the latest financial data, the future short-term reimbursement amount, the reimbursement type corresponding to the future short-term reimbursement amount, and the future approval time output by the converged neural network model are obtained. Based on the latest borrowing and repayment data in the latest financial data, the future short-term borrowing amount and the future short-term repayment rate output by the converged neural network model are obtained. Based on the latest profit data in the latest financial data, the future short-term profit trend and future short-term predicted risk output by the converged neural network model are obtained.

[0059] Furthermore, based on the future approval time, the future short-term reimbursement amount within the preset time period and the reimbursement type corresponding to the future short-term reimbursement amount are counted, and the reimbursement frequency of different reimbursement types within the preset time period is determined respectively. Based on the reimbursement frequency, a long-term trend analysis is conducted on the reimbursement business of different reimbursement types respectively. Based on the future short-term loan amount, the company's capital demand and liquidity status are determined, based on the future short-term repayment rate, the company's debt repayment ability and credit status are determined, based on the short-term profit trend, the company's profitability and operating efficiency are determined, and based on the future short-term predicted risks, the company's financial risk is determined.

[0060] It should be noted that in the embodiments of the present application, key performance indicators such as sales growth rate, gross profit margin, customer acquisition cost, etc. can also be set, the business operation status can be monitored through data, thresholds and rules can be set, abnormal situations can be monitored in real time, and alarms can be automatically sent through emails, text messages, etc., cluster analysis and association rule technology can be used to identify potential business opportunities and risks, and detect abnormal transactions and risk signals in financial data.

[0061] S106: Planning the enterprise's finances based on the predicted financial data.

[0062] Specifically, based on the frequency of reimbursement, determine the high-frequency reimbursement types within the preset time period, and automatically review the reimbursements corresponding to the high-frequency reimbursement types to reduce approval time. Formulate funding plans and liquidity management strategies based on funding needs, liquidity status, debt repayment ability and credit status. Formulate operating strategies based on profitability and operational efficiency. Set risk thresholds based on financial risks, and identify potential risks through risk thresholds.

[0063] It should be noted that if Figure 2As shown, the business modules and functions of the data-driven financial planning system in the embodiment of the present application include business common, basic data, financial sharing, financial accounting, accounts receivable and payable, tax management, fund management, budget management, etc. Among them, the basic data module provides a core data management platform for the enterprise, including public basic data, financial basic data and business basic data. This module supports enterprises to centrally and standardizedly manage and maintain key information such as customers, suppliers, products, accounts and cost centers. Through effective data management, enterprises can ensure the accuracy, consistency and integrity of data; the business common module provides a series of key basic business functions, including business configuration, electronic images, electronic wallets, electronic archives, intelligent auditing, shared services, payment centers, bank-enterprise services and business collaboration. These functions jointly build an integrated platform to support collaborative work and efficient operations inside and outside the enterprise. From business process configuration to financial document management, to cross-departmental collaboration and banking business integration, this module aims to provide comprehensive business support to help enterprises optimize business processes, efficiently utilize resources and collaborate with teams; the financial sharing module provides enterprises with a comprehensive set of financial management solutions, including online reimbursement services, travel service platforms, financial operation centers, shared operation centers, employee credit management, quality management platforms, shared evaluation management, performance management platforms, shared data services, operation support platforms, service agreement management, service customization platforms and automated operation centers. These functions support enterprises to automate and standardize financial processes, improve financial operation efficiency, reduce costs, and enhance financial transparency and compliance. At the same time, through data sharing and integration, this module also promotes cross-departmental and cross-team collaboration, helping enterprises achieve continuous financial optimization and business innovation.

[0064] It should be noted that if Figure 3As shown, it is a flowchart of the data-driven financial planning system in the embodiment of the present application. First, the group system administrator sets public basic data, including: administrative organization, accounting organization, settlement method, public expense items, etc. In addition, it is necessary to define administrative personnel and set unit permissions for each user. When users make orders, inquire, etc., they apply for operations according to the unit permissions and can only access data within their permissions. Secondly, the system administrators of each branch and subsidiary define private basic data, including: accounting department, accounting personnel, organizational expense items, etc. Secondly, the shared center organization definition is carried out in the financial sharing-operation support platform, including: shared service definition, business group definition, and operator definition. When defining, different construction modes and business group grouping methods are selected according to the development needs of the enterprise group itself and the different stages of enterprise development. It is recommended to do a good job of consultation and demand analysis in the early stage of implementation. Secondly, the shared center business definition is carried out in the financial sharing-operation support platform, including reimbursement type definition, task allocation plan, loan write-off control, reimbursement account management, and entrusted management. Secondly, the system settings are defined in the shared center-operation support platform, including reimbursement parameter settings, standard dimension definitions, document type definitions, form format settings, etc. Secondly, before business processing, employees can maintain my account and other information. When the system is initialized, the balance of those with loan records is initialized. Secondly, during the business processing, the document flow is carried out, mainly including application-preliminary review-approval-review-certification-review. During the processing, the financial staff is dispatched by the task allocation of the operation center, and can automatically dispatch and actively raise bills. At the same time, each role can also perform corresponding queries and statistical analysis according to the authority. Finally, after the review is completed, the payment request is sent to the fund management system to complete the settlement. At the same time, the budget system controls the occurrence of expenses in real time.

[0065] like Figure 4 As shown, on the other hand, an embodiment of the present invention further provides a data-driven financial planning system, characterized in that the system includes:

[0066] The automated process module is used to automatically execute the financial process and obtain the financial information of the enterprise by automatically executing the script;

[0067] A data integration processing module, the data integration processing module is used to collect historical financial data based on the financial information, pre-process the historical financial data according to preset financial indicators to obtain standard financial data, and construct a training set and a validation set based on the standard financial data, the historical financial data including reimbursement data, loan repayment data, and profit data;

[0068] A data mining module, the data mining module is used to perform vector conversion on the standard financial data to obtain a financial data vector, extract key feature vectors from the financial data vector according to financial planning requirements, and construct a neural network model based on the key feature vectors and a preset neural network structure;

[0069] An intelligent data analysis module, wherein the intelligent data analysis module is used to input the training set into the neural network model, determine the loss function of the neural network model based on the prediction data output by the neural network model, adjust the model parameters of the neural network model according to the loss function, and verify the neural network model through the verification set until the neural network model converges;

[0070] A predictive analysis module, the predictive analysis module is used to collect the latest financial data, and output predicted financial data based on the latest financial data through the converged neural network model;

[0071] A data planning module is used to plan the enterprise's finances based on the predicted financial data.

[0072] An embodiment of the present application further provides a non-volatile computer storage medium storing computer executable instructions, wherein the computer executable instructions are configured to be: a data-driven financial planning method as described in any of the above embodiments.

[0073] Each embodiment in this application is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device and medium embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiments.

[0074] The devices and media provided in the embodiments of the present application correspond one-to-one to the methods. Therefore, the devices and media also have similar beneficial technical effects as the corresponding methods. Since the beneficial technical effects of the methods have been described in detail above, the beneficial technical effects of the devices and media will not be repeated here.

[0075] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take 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 code.

[0076] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks 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 produce 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.

[0077] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0078] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0079] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0080] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0081] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0082] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0083] The above is only an embodiment of the present application and is 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 in the scope of the claims of the present application.

Claims

1. A data-driven financial planning method, characterized in that: include: Automate financial processes and obtain financial information of enterprises through automated execution of scripts; Based on the financial information, historical financial data are collected and pre-processed according to preset financial indicators to obtain standard financial data, and a training set and a validation set are constructed based on the standard financial data, wherein the historical financial data includes reimbursement data, loan repayment data, and profit data; Performing vector conversion on the standard financial data to obtain a financial data vector, extracting key feature vectors from the financial data vector according to financial planning requirements, and constructing a neural network model based on the key feature vectors and a preset neural network structure; Inputting the training set into the neural network model, determining a loss function of the neural network model based on the prediction data output by the neural network model, adjusting model parameters of the neural network model according to the loss function, and verifying the neural network model through the verification set until the neural network model converges; Collecting the latest financial data, and outputting predicted financial data based on the latest financial data through the converged neural network model; Based on the forecasted financial data, corporate finances are planned.

2. A data-driven financial planning method according to claim 1, characterized in that: The extracting of key feature vectors from the financial data vector according to the financial planning requirements specifically includes: Obtaining financial planning needs, performing demand analysis on the financial planning needs, and determining that the financial planning needs include reimbursement management planning, debt management planning, and profit management planning; According to the financial planning requirements, corresponding key features are determined respectively, and based on the key features, corresponding key feature vectors are extracted from the financial data vectors.

3. A data-driven financial planning method according to claim 2, characterized in that: According to the financial planning requirements, the corresponding key features are determined respectively, including: According to the reimbursement management plan, determining key features of the reimbursement data including reimbursement amount, reimbursement category, reimbursement time, and budget execution; Determine, according to the debt management plan, the key features of the loan repayment data including repayment amount, repayment time, loan amount, loan term, and interest rate; According to the profit management plan, the key features of the profit data are determined to include cash flow, profit margin, and debt-to-asset ratio.

4. A data-driven financial planning method according to claim 3, characterized in that: The step of inputting the training set into the neural network model and determining the loss function of the neural network model based on the prediction data output by the neural network model specifically includes: Training the neural network model based on the reimbursement data, the loan and repayment data, and the profit data respectively; Inputting the historical reimbursement amount in the reimbursement data, the reimbursement type corresponding to the historical reimbursement amount, and the historical approval time into the neural network model to obtain the predicted reimbursement amount and predicted approval time output by the neural network model; Inputting the historical loan and repayment periods, actual loan and repayment dates, and loan and repayment amounts in the loan and repayment data into a neural network model to obtain the loan trend and repayment punctuality rate output by the neural network model; Inputting historical cash flow and historical profit in the profit data into the neural network model to obtain the predicted profit trend and predicted risk output by the neural network model; Determine the error between the predicted data and the actual data output by the neural network model, and determine the loss function of the neural network model based on the error.

5. The data-driven financial planning method according to claim 1, characterized in that: The automatic execution of scripts to automatically execute financial processes and obtain the financial information of the enterprise specifically includes: Determine the business to which the financial process belongs by automating the execution of scripts; According to the business, determine the access data source, connect to the interface corresponding to the data source, and obtain financial information.

6. A data-driven financial planning method according to claim 1, characterized in that: The outputting of predicted financial data based on the latest financial data through the converged neural network model specifically includes: Inputting the latest financial data into the convergent neural network model, and obtaining the future short-term reimbursement amount, the reimbursement type corresponding to the future short-term reimbursement amount, and the future approval time output by the convergent neural network model according to the latest reimbursement data in the latest financial data; According to the latest loan and repayment data in the latest financial data, obtaining the future short-term loan amount and the future short-term repayment rate output by the converged neural network model; According to the latest profit data in the latest financial data, the future short-term profit trend and future short-term forecast risk output by the converged neural network model are obtained.

7. A data-driven financial planning method according to claim 6, characterized in that: After outputting predicted financial data based on the latest financial data through the converged neural network model, the method further includes: According to the future approval time, the future short-term reimbursement amount and the reimbursement type corresponding to the future short-term reimbursement amount within a preset time period are counted, and the reimbursement frequency of different reimbursement types within the preset time period is determined respectively; Determine the capital demand and liquidity status of the enterprise based on the future short-term loan amount, and determine the debt repayment ability and credit status of the enterprise based on the future short-term repayment rate; Based on the future short-term profit trend, the profitability and operational efficiency of the enterprise are determined, and based on the future short-term predicted risks, the financial risk of the enterprise is determined.

8. A data-driven financial planning method according to claim 7, characterized in that: The planning of enterprise finance according to the predicted financial data specifically includes: According to the reimbursement frequency, determine the high-frequency reimbursement type within the preset time period, and automatically review the reimbursement corresponding to the high-frequency reimbursement type to reduce the approval time; Formulate funding plans and liquidity management strategies based on the funding needs, liquidity status, debt repayment capacity and credit status; Formulate operating strategies based on said profitability and said operating efficiency; A risk threshold is set according to the financial risk, and potential risks are identified through the risk threshold.

9. A data-driven financial planning system, characterized in that: The system comprises: The automated process module is used to automatically execute the financial process and obtain the financial information of the enterprise by automatically executing the script; A data integration processing module, the data integration processing module is used to collect historical financial data based on the financial information, pre-process the historical financial data according to preset financial indicators to obtain standard financial data, and construct a training set and a validation set based on the standard financial data, the historical financial data including reimbursement data, loan repayment data, and profit data; A data mining module, the data mining module is used to perform vector conversion on the standard financial data to obtain a financial data vector, extract key feature vectors from the financial data vector according to financial planning requirements, and construct a neural network model based on the key feature vectors and a preset neural network structure; An intelligent data analysis module, wherein the intelligent data analysis module is used to input the training set into the neural network model, determine the loss function of the neural network model based on the prediction data output by the neural network model, adjust the model parameters of the neural network model according to the loss function, and verify the neural network model through the verification set until the neural network model converges; A predictive analysis module, the predictive analysis module is used to collect the latest financial data, and output predicted financial data based on the latest financial data through the converged neural network model; A data planning module is used to plan the enterprise's finances based on the predicted financial data.

10. A non-volatile computer storage medium storing computer executable instructions, characterized in that: The computer executable instructions are configured as: a data-driven financial planning method as described in any one of claims 1 to 8.

Images