Digital financial management system and method based on business and financial integration

Abstract

The invention provides a digital financial management system and method based on business and financial fusion, and the method comprises the steps: formulating a data format standard according to enterprise business and financial data, removing repeated data, filling missing data, and processing abnormal data, and obtaining standardized enterprise business and financial data; constructing a business and financial integration management index system, and combining enterprise business and financial data to obtain various indexes for decision making; training an LSTM model based on the historical values of the indexes, and reasoning and predicting the development trends of the indexes based on the trained LSTM model; the historical values of the indexes and the development trends of the indexes are visually displayed; accuracy and consistency of enterprise business and financial data are ensured, various indexes useful for decision making are provided in combination with the enterprise business and financial data, enterprises are helped to better understand and grasp business conditions and financial conditions, the LSTM model is trained based on historical data, future development trends of the indexes can be predicted, and the enterprise business and financial data can be predicted. And an enterprise can conveniently plan and adjust a strategy in advance.

Description

Technical Field

[0001] The present invention belongs to the technical field of digital financial management, and in particular relates to a digital financial management system and method based on business-finance integration. Background Art

[0002] Digital financial management refers to a management method that uses modern information technology to manage, process and analyze financial data using digital tools and systems in the financial management process. It uses cloud computing, big data analysis, artificial intelligence and other technologies to digitize and automate traditional financial management processes and improve the efficiency, accuracy and real-time nature of financial data processing.

[0003] Digital financial management covers many aspects, including financial data collection and integration, financial report generation, financial analysis and forecasting, risk management, decision support, etc. By applying digital tools and systems, financial data can be automatically collected from different business systems and data sources, financial data can be automatically integrated and cleaned, accurate financial reports can be generated, real-time financial analysis and forecasting can be performed, and risk management and decision-making can be assisted.

[0004] The existing technologies focus on financial data, without considering business data closely related to finance. A digital financial management system and method based on business-finance integration can speed up the information exchange between business and finance. Efficient interaction in information exchange and status update can help achieve a high degree of coupling between finance and business, integrate financial management into the entire process of business before, during and after the event, effectively identify opportunities and goals for business improvement, help solve business pain points and difficulties, and enable production operation optimization. In addition, it can also meet the needs of financial standard nucleic acid, centralized control and other professional applications. Summary of the invention

[0005] In order to solve the problems existing in the prior art, the present invention provides a digital financial management system and method based on business and financial integration to solve the problems raised in the above background.

[0006] In order to achieve the above-mentioned purpose, in a first aspect, the present invention provides a digital financial management system based on business-finance integration, including a data preprocessing module, an integration and interface module, a data analysis module, a time series data analysis module, a decision support module, and a security and privacy module;

[0007] The integration and interface module is used to integrate with other systems and platforms and obtain enterprise business and financial data;

[0008] The data preprocessing module cleans and standardizes the business data and financial data to obtain standardized enterprise business and financial data;

[0009] The data analysis module is based on the data source tables of business and finance, and according to the indicator system of business-finance integration, writes SQL to calculate the indicators of the data source tables stored in the database to obtain the value of each indicator;

[0010] The time series data analysis module is used for time series prediction of historical indicator values, using a multi-round iterative training method and a long short-term memory time series model to predict the trend of a single indicator;

[0011] The decision support module provides data visualization and reporting tools through data reports, icons and visual dashboards to display the historical and predicted values ​​of various indicators.

[0012] Furthermore, the integration and interface module reads the CSV data source output from the enterprise financial management or business system through Spark SQL and stores it in the database.

[0013] Furthermore, the data preprocessing module includes a data acquisition unit, a cleaning unit and a conversion unit. The data acquisition unit collects multi-source heterogeneous data such as enterprise business activity information, industry information, and other system integration interface data. The cleaning unit removes duplicate data, fills in missing data and processes abnormal data. The conversion unit standardizes the original data according to the data format standard to ensure the accuracy of the data.

[0014] Furthermore, the indicator system of business-finance integration includes the following indicators: electricity sales revenue, return on net assets, asset income-generating ability, cost-profit ratio, operating income growth rate, gross profit growth rate of electricity purchase and sales, return on total assets, debt-to-asset ratio, increase in interest-bearing liabilities, comprehensive interest rate of interest-bearing liabilities, proportion of interest-bearing liabilities, current ratio, quick ratio, electricity sales revenue per 10,000 yuan of assets, electricity sales revenue, return on net assets, asset income-generating ability, cost-profit ratio, operating income growth rate, gross profit growth rate of electricity purchase and sales and return on total assets.

[0015] Furthermore, using the long short-term memory time series model to predict the trend of a single indicator includes: constructing a data set corresponding to a single indicator, setting the parameters of the long short-term memory time series model, training with the parameters of the long short-term memory time series model in the training set, verifying the trained parameters with the validation set, and testing the verified long short-term memory time series model parameters with the test set, until the test passes to obtain an optimized long short-term memory time series model.

[0016] In a second aspect, the present invention further provides a digital financial management method based on business-finance integration, comprising the following steps:

[0017] Processing of business and financial data of enterprises: according to the data format standards set by the enterprise, removing duplicate data, filling in missing data, processing abnormal data, and obtaining standardized business and financial data of enterprises;

[0018] Build a business-finance integration management indicator system, combining the company's business and financial data to obtain various indicators that are useful for decision-making;

[0019] Based on the historical values ​​of various indicators, the LSTM model is trained, and the development trend of various indicators is predicted based on the reasoning of the trained LSTM model;

[0020] Visualize the historical values ​​and development trends of various indicators.

[0021] Furthermore, when filling in missing data, the data is supplemented based on the historical year-on-year and month-on-month data of the power company; for abnormal data, a threshold is set and the data exceeding the threshold is removed.

[0022] Furthermore, the historical indicator values ​​obtained by the data analysis module and the predicted indicator values ​​obtained by the time series data analysis module are visualized through Vue and echarts visualization tools.

[0023] In addition, the present invention can also provide a computer device, including a processor and a memory, the memory is used to store a computer executable program, the processor reads the computer executable program from the memory and executes it, and when the processor executes the computer executable program, it can implement the digital financial management method based on business and financial integration described in the present invention.

[0024] At the same time, a computer-readable storage medium is provided, in which a computer program is stored. When the computer program is executed by a processor, the digital financial management method based on business-finance integration described in the present invention can be implemented.

[0025] Compared with the prior art, the present invention has at least the following beneficial effects:

[0026] The present invention proposes a digital financial management system based on the integration of business and finance. The system acquires enterprise business and financial data through an integration and interface module, processes business data and financial data through a data preprocessing module, calculates the index value of the business-finance integration management system through a data analysis module, obtains the development forecast result of the index value through a time series data analysis module, and finally displays the result to users in the form of data reports, icons and visual dashboards through a decision support module. The digital financial management system based on the integration of business and finance provided by the present invention is novel in creativity, perfect in overall process and has promotion value. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 This is a system architecture diagram of a digital financial management system based on business-finance integration according to the present invention;

[0028] Figure 2This is a method flow chart of a digital financial management method based on business-finance integration of the present invention;

[0029] Figure 3 The method of the present invention is used to partially predict the results. DETAILED DESCRIPTION

[0030] In order to make the technical means, creative features, objectives and effects achieved by the present invention easy to understand, a specific implementation method is summarized below to further illustrate the present invention.

[0031] Embodiment 1:

[0032] Please refer to Figure 1 As shown: a digital financial management system based on business-finance integration, including an integration and interface module, a data preprocessing module, a data analysis module, a time series data analysis module, and a decision support module; the integration and interface module is used to integrate with other systems and platforms, and obtain enterprise business and financial data, the data analysis module calculates the indicator value of the business-finance integration indicator system based on the original data, the data preprocessing module processes the original data such as business data and financial data, the time series data analysis module constructs a deep learning model for indicator prediction based on the time series indicator value, and the decision support module provides data visualization and reporting tools to provide data support for user management decisions.

[0033] The integration and interface module is used to integrate with other systems and platforms. When an enterprise introduces a new financial management or business system, Spark SQL reads the CSV data source output from the financial management or business system and stores it in the database.

[0034] Specifically, Spark SQL is used to read CSV files. Spark SQL supports reading data directly from multiple CSV files and converting them into multiple DataFrames. Each processed DataFrame is directly written to an analytical database (such as Hive, Clickhouse) and stored in a columnar format (such as Parquet, ORC) to speed up data processing.

[0035] The data preprocessing module consists of a data acquisition unit, a cleaning unit, and a conversion unit. The data acquisition unit collects multi-source heterogeneous data such as enterprise business activity data, industry data, and other system integration interface data. The cleaning unit removes duplicate data, fills in missing data, and processes abnormal data. The conversion unit standardizes the original data according to the data format standard to ensure the accuracy of the data.

[0036] Specifically, daily data is divided into multiple database tables according to the source type of the business system. Examples of the original CSV file format of some database tables are shown in Table 1:

[0037] Table 1

[0038] project Row Gross labor product 1 Fixed self-depreciation 2 Taxes payable 3 Enterprise labor costs 4 Labor cost of employees 5 Employee wages 6 Retirement planning 7 Total employee remuneration payable today 8 Total operating income 9 Total operating costs 10 Total investment income 11 Total Profit 12 Total net profit 13 Operating Profit 14 Income tax payable 15 Government Grants 16 Cost and expense table 17 Power purchase cost 18 Operating Profit 19 Gross labor production value verification 20

[0039] Based on the integration and interface module, a single CSV is imported into the database to obtain a single table. By writing SQL, missing data is filled according to the default values ​​of different fields, and abnormal data whose values ​​exceed the normal data definition range is cleaned.

[0040] Specifically, through the INSERT and SELECT operations in SQL, the library tables in the original database are read, and the missing data is accurately filled according to the default values ​​of different fields. At the same time, the SQL WHERE conditional statement is used to check for abnormal situations where the data value exceeds the normal data definition value range, and the abnormal data whose data value exceeds the normal data definition value range is cleaned and stored as a standardized database table.

[0041] Among them, the data analysis module defines a set of business-finance integration indicator system based on the data source tables of business and finance, and writes SQL to calculate the indicators of the data source tables stored in the database to obtain the value of each indicator.

[0042] Specifically, the indicator system and calculation method based on business-finance integration are illustrated in Table 2:

[0043] Table 2

[0044]

[0045]

[0046] Specifically, based on the standardized database table obtained by the data preprocessing module, the relevant data is calculated and processed by writing SQL statements, so as to accurately calculate the above-mentioned indicator values. Subsequently, these calculated indicator values ​​are stored in the data indicator library table for subsequent in-depth data analysis and application.

[0047] Among them, the time series data analysis module is used for time series prediction of historical indicator values. It adopts multiple rounds of iterative training to build a long and short-term memory time series model to predict the performance of a single indicator.

[0048] The following is an example of electricity sales revenue:

[0049] 1) The dataset construction process is as follows:

[0050] Based on the electricity sales revenue index values ​​obtained by the data analysis module, the original data set is partially described in Table 3:

[0051] Table 3

[0052]

[0053] Based on the electricity sales revenue index values ​​obtained by the data analysis module, the original data set is described in Table 4.

[0054] Table 4

[0055]

[0056] In order, 70% of the data is used as a training data set to train the model, 20% of the data is used as a validation data set to prevent overfitting, and 10% of the data is used as a test data set to evaluate the quality of model prediction.

[0057] 2) The training process of the time series prediction model is as follows:

[0058] Table 5: Setting LSTM model hyperparameters:

[0059] Table 5

[0060]

[0061]

[0062] Each 30 epochs is a complete model training process, and the model training is performed three times. The train_loss (MSE) of each trained model in the last epoch training and the vali_loss (MSE) based on the validation data set refer to Table 6:

[0063] Table 6

[0064]

[0065] After each model is trained, the test data set is used to input 50 steps of historical data, predict the next 5 steps of future data, and compare them with the label data. The evaluation indicators (MAE, MSE, RMAE) are selected for model evaluation as shown in Table 7:

[0066] Table 7

[0067]

[0068] For the original data set, in the three models obtained through training, the values ​​of MAE, MSE, and RMSE are all less than 0.1, which proves that the model training method has the characteristics of low prediction error and stable prediction.

[0069] In order to observe the prediction effect of the model more intuitively, based on the MSE and MAE evaluation indicators, we selected the No. 1 model with better average prediction performance for testing. We input the data of the first 50 days and predicted the data of the next 5 days. The following shows the results of one of the experiments, as shown in the attached figure. Figure 3 As shown in the figure, observing the data of the last five days, the blue line is the actual value, and the orange and yellow lines are the predicted values. It can be seen from the figure that the prediction difference is small.

[0070] Among them, the decision support module uses Vue and echarts visualization tools to draw data reports, icons and visualization dashboards to display the history and forecast of each indicator. Users can view the corresponding analysis results by selecting key management decision indicators, providing data support for user management decisions.

[0071] Specifically, we use Vue, a front-end development framework, to build the front-end page, and pass the processed and integrated indicator data from the back-end to the front-end through the front-end and back-end API interfaces. At the same time, we use the ECharts library to visualize the indicator data, such as drawing clear data reports, intuitive icons, and visual dashboards that fully display data dynamics. These visualization results can provide users with intuitive data insights, and then provide strong support for decision-making, helping users make reasonable decisions more quickly and accurately, and improving work efficiency and decision-making quality.

[0072] Embodiment 2:

[0073] Please refer to Figure 2 As shown: A digital financial management method based on business-finance integration includes the following steps:

[0074] Step 1: First, use Spark SQL through the integration and interface module to read the CSV data source output from the enterprise financial management or business system and store it in the database.

[0075] Step 2, then write SQL through the data preprocessing module to clean, process and convert the enterprise's business and financial integration data.

[0076] Step 3, the data analysis module builds a business-finance integration management indicator system, and calculates the corresponding indicator values ​​by writing SQL based on the database tables stored in the database.

[0077] Step 4: The time series data analysis module trains the LSTM model based on historical indicator values ​​and infers and predicts the development trend of indicators.

[0078] Step 5: The decision support module uses the Vue+echarts visualization tool to visualize the historical indicator values ​​obtained by the data analysis module and the predicted indicator values ​​obtained by the time series data analysis module to provide data support for user management decisions.

[0079] The present invention proposes a digital financial management system based on business-finance integration. The system obtains enterprise business and financial data through an integration and interface module, processes business data and financial data through a data preprocessing module, calculates the business-finance integration management system index value through a data analysis module, obtains the development forecast result of the index value through a time series data analysis module, and finally displays the result to the user in the form of data reports, icons and visual dashboards through a decision support module. The digital financial management system based on business-finance integration provided by the present invention is novel in creativity, perfect in overall process, and has promotion value.

[0080] The present invention can also provide a computer device, including a processor and a memory, the memory is used to store a computer executable program, the processor reads the computer executable program from the memory and executes it, and when the processor executes the computer executable program, it can implement the digital financial management method based on business and financial integration described in the present invention.

[0081] On the other hand, the present invention provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, the digital financial management method based on business-finance integration described in the present invention can be implemented.

[0082] The computer device may be a laptop computer, a desktop computer or a workstation.

[0083] The processor may be a central processing unit (CPU), a digital signal processor (DSP), an application specific integrated circuit (ASIC), or an off-the-shelf field programmable gate array (FPGA).

[0084] The memory described in the present invention may be an internal storage unit of a notebook computer, a desktop computer or a workstation, such as a memory or a hard disk; or an external storage unit, such as a mobile hard disk or a flash memory card.

[0085] Computer-readable storage media may include computer storage media and communication media. Computer storage media include volatile and non-volatile, removable and non-removable media implemented by any method or technology for storing information such as computer-readable instructions, data structures, program modules or other data. Computer-readable storage media may include: read-only memory (ROM), random access memory (RAM), solid-state drive (SSD) or optical disk, etc. Among them, random access memory may include resistance random access memory (ReRAM) and dynamic random access memory (DRAM).

[0086] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A digital financial management system based on business and financial integration, characterized in that: It includes data preprocessing module, integration and interface module, data analysis module, time series data analysis module, decision support module, and security and privacy module; The integration and interface module is used to integrate with other systems and platforms and obtain enterprise business and financial data; The data preprocessing module cleans and standardizes the business data and financial data to obtain standardized enterprise business and financial data; The data analysis module is based on the data source tables of business and finance, and according to the indicator system of business-finance integration, writes SQL to calculate the indicators of the data source tables stored in the database to obtain the value of each indicator; The time series data analysis module is used for time series prediction of historical indicator values, using a multi-round iterative training method and a long short-term memory time series model to predict the trend of a single indicator; The decision support module provides data visualization and reporting tools through data reports, icons and visual dashboards to display the historical and predicted values ​​of various indicators.

2. A digital financial management system based on business-finance integration according to claim 1, characterized in that: The integration and interface module reads the CSV data source output from the enterprise financial management or business system through Spark SQL and stores it in the database.

3. According to claim 1, a digital financial management system based on business-finance integration is characterized in that: The data preprocessing module includes a data acquisition unit, a cleaning unit and a conversion unit. The data acquisition unit collects multi-source heterogeneous data such as enterprise business activity information, industry information and other system integration interface data. The cleaning unit removes duplicate data, fills in missing data and processes abnormal data. The conversion unit standardizes the original data according to the data format standard to ensure the accuracy of the data.

4. According to claim 1, a digital financial management system based on business-finance integration is characterized in that: The indicator system of business-finance integration includes the following indicators: electricity sales revenue, return on net assets, asset income-generating ability, cost-profit ratio, operating income growth rate, gross profit growth rate of electricity purchase and sales, return on total assets, asset-liability ratio, increase in interest-bearing liabilities, comprehensive interest rate of interest-bearing liabilities, proportion of interest-bearing liabilities, current ratio, quick ratio, electricity sales revenue per 10,000 yuan of assets, electricity sales revenue, return on net assets, asset income-generating ability, cost-profit ratio, operating income growth rate, gross profit growth rate of electricity purchase and sales and return on total assets.

5. According to claim 1, a digital financial management system based on business-finance integration is characterized in that: Using the long short-term memory time series model to predict the trend of a single indicator includes: constructing a data set corresponding to a single indicator, setting the parameters of the long short-term memory time series model, training the parameters of the long short-term memory time series model using the training set, verifying the trained parameters using the validation set, and testing the verified long short-term memory time series model parameters using the test set, until the test passes to obtain the optimized long short-term memory time series model.

6. A digital financial management method based on business-finance integration, characterized in that: The following steps are involved: Processing of business and financial data of enterprises: according to the data format standards set by the enterprise, removing duplicate data, filling in missing data, processing abnormal data, and obtaining standardized business and financial data of enterprises; Build a business-finance integration management indicator system, combining the company's business and financial data to obtain various indicators that are useful for decision-making; Based on the historical values ​​of various indicators, the LSTM model is trained, and the development trend of various indicators is predicted based on the reasoning of the trained LSTM model; Visualize the historical values ​​and development trends of various indicators.

7. The digital financial management method based on business-finance integration according to claim 6 is characterized in that: When filling in missing data, the data is supplemented based on the historical year-on-year and month-on-month data of the power company; for abnormal data, a threshold is set and the data exceeding the threshold is removed.

8. The digital financial management method based on business-finance integration according to claim 6 is characterized in that: The historical indicator values ​​obtained by the data analysis module and the predicted indicator values ​​obtained by the time series data analysis module are visualized through Vue and echarts visualization tools.

9. A computer device, characterized in that: It includes a processor and a memory, the memory is used to store a computer executable program, the processor reads the computer executable program from the memory and executes it, and when the processor executes the computer executable program, it can implement the digital financial management method based on business-finance integration as described in any one of claims 6-8.

10. A computer-readable storage medium, characterized in that: A computer program is stored in a computer-readable storage medium. When the computer program is executed by a processor, it can implement a digital financial management method based on business-finance integration as described in any one of claims 6 to 8.

Images