A financial risk analysis method and system based on a big data analysis model

Abstract

The present application relates to a financial risk analysis method and system based on a big data analysis model. The method includes obtaining enterprise fiscal and tax data and external industry data within a certain time period and performing data preprocessing to obtain preprocessed internal enterprise data and external enterprise data, randomly screening the internal enterprise data and the external enterprise data for correlation analysis, constructing a big data analysis model for enterprise finance and taxation, obtaining fiscal and tax risk data in the internal enterprise data that exceeds the preset fiscal and tax item indicators, inputting the fiscal and tax risk data into the big data analysis model, performing risk decision-making processing on the fiscal and tax risk data, performing classification and iterative processing on the fiscal and tax risk data according to the risk decision result, calculating a fiscal and tax risk decision deviation coefficient according to the iterative result and correcting the risk decision result to obtain financial risk analysis data. The present application has the effect of improving the accuracy of financial risk analysis.

Description

Technical Field

[0001] The present invention relates to the technical field of fiscal and tax analysis, and particularly to a financial risk analysis method and system based on a big data analysis model. Background Art

[0002] At present, with the continuous development of the global economy, financial risk has become an important consideration in enterprise management and investment decision-making. Whether it is an enterprise or an investor, it is necessary to conduct in-depth analysis and evaluation of potential financial risks in order to make correct decisions. Therefore, it is necessary to accurately analyze financial risks.

[0003] Existing financial risk analysis methods usually preset the discriminant criteria for financial indicators of corresponding items according to financial types. When the actual financial data of the corresponding item reaches or exceeds the corresponding discriminant criteria, it indicates that there is a risk in the current financial item. However, in large enterprises, there are many types of finances and there are close connections between financial indicators. Only using qualitative assumption conditions to represent the risks of enterprise finances in a certain aspect is likely to cause the instability of the financial risk verification results, and in the case of ignoring the index correlation, the risk verification conclusions of different financial items may also be contradictory, resulting in the inability to obtain a risk judgment result that conforms to the actual operation situation of the enterprise. Therefore, it is necessary to further optimize the financial risk analysis method. Summary of the Invention

[0004] In order to improve the accuracy of financial risk analysis, the present application provides a financial risk analysis method and system based on a big data analysis model.

[0005] In the first aspect, the above-mentioned invention object of the present application is achieved through the following technical solutions:

[0006] A financial risk analysis method based on a big data analysis model, comprising:

[0007] Obtaining enterprise fiscal and tax data and external industry data within a certain time period for the enterprise and performing data preprocessing to obtain preprocessed enterprise internal data and enterprise external data;

[0008] Randomly screening the enterprise internal data and the enterprise external data respectively, analyzing the correlation relationship between the screened enterprise internal data and enterprise external data, and constructing a big data analysis model for enterprise fiscal and tax;

[0009] Obtaining fiscal and tax risk data in the enterprise internal data that exceeds the preset fiscal and tax item indicators, and inputting the fiscal and tax risk data into the big data analysis model to perform risk decision-making processing on the fiscal and tax risk data;

[0010] Classify and iteratively process the fiscal and tax risk data according to the risk decision result, calculate the deviation coefficient of the fiscal and tax risk decision based on the iterative result, and correct the risk decision result to obtain the financial risk analysis data.

[0011] By adopting the above technical solution, preprocessing the enterprise's fiscal and tax data and external industry data within a selected time period helps filter low-quality or duplicate and invalid data, retain high-value data with commercial value, increase the randomness of data selection by randomly screening data, and combine the correlation between the enterprise's internal data and external data to construct a big data model for the enterprise's fiscal and tax, which helps improve the correlation between the internal and external data of the enterprise. Taking external data as the analysis index, it improves the stability of model demonstration from multiple dimensions. By screening the fiscal and tax risk data that exceeds the preset fiscal and tax project indicators, targeted directional risk analysis is carried out on the fiscal and tax risk data to improve the accuracy of risk decision-making. Through the classification iteration and deviation correction of the fiscal and tax risk data, the fiscal and tax risk analysis result is further corrected, and the fiscal and tax risk analysis result is multi-verified from multiple angles, thereby improving the accuracy of financial risk analysis.

[0012] In a preferred example of the present application, it can be further configured as follows: The classifying and iteratively processing the fiscal and tax risk data according to the risk decision result, calculating the deviation coefficient of the fiscal and tax risk decision based on the iterative result, and correcting the risk decision result to obtain the financial risk analysis data specifically includes:

[0013] According to the risk decision result, classify the fiscal and tax risk data again, and randomly select the node data where the fiscal and tax data changes from the fiscal and tax items of each classified data to obtain the first data set and the second data set;

[0014] According to the number of changes in the fiscal and tax data, perform data iterative processing on the first data set and the second data set respectively, analyze the fiscal and tax risk decision limits of the corresponding data sets, and construct the risk decision conditions of the big data analysis model;

[0015] According to the difference between the risk decision condition and the original risk decision result, calculate the deviation coefficient of the fiscal and tax risk decision, and correct the original risk decision result according to the deviation coefficient of the fiscal and tax risk decision to obtain the financial risk data.

[0016] In a preferred example of the present application, it can be further configured as follows: The calculating the deviation coefficient of the fiscal and tax risk decision according to the difference between the risk decision condition and the original risk decision result specifically includes:

[0017] Calculate the deviation coefficient of the fiscal and tax risk decision through formula (1), and formula (1) is as follows:

[0018]

[0019] Among them, ρ represents the deviation coefficient of fiscal and tax risk decision-making, n and m respectively represent the number of nodes where fiscal and tax data change and the number of fiscal and tax items with fiscal and tax data changes, and Δp ij represents the data change value of the j-th fiscal and tax node with fiscal and tax data changes in the i-th fiscal and tax item.

[0020] By adopting the above technical solution, the fiscal and tax risk data is secondarily classified according to the risk decision result for data selection, and the node data where the fiscal and tax data changes is randomly selected as the training data set of big data, so as to obtain the first data set and the second data set according to the secondary classification result, which helps to verify the risk decision result by comparison, and takes the number of changes in fiscal and tax data as the number of iterations, iteratively processes the data of the first data set and the second data set respectively, and analyzes the fiscal and tax risk decision limits of the corresponding data sets, so as to construct the risk decision conditions of the big data analysis model, constrain the risk decision of the big data analysis model, and combine the fiscal and tax risk decision deviation coefficient to correct the original risk decision result, further improving the accuracy of financial risk analysis.

[0021] In a preferred example of the present application, it can be further configured as: obtaining the fiscal and tax risk data exceeding the preset fiscal and tax item indicators in the enterprise internal data, and inputting the fiscal and tax risk data into the big data analysis model for risk decision processing of the fiscal and tax risk data, specifically including:

[0022] Constructing a risk association relationship between the fiscal and tax risk data and the enterprise external data, analyzing the fiscal and tax risk distribution of the fiscal and tax risk data according to the risk association relationship to obtain fiscal and tax risk analysis data;

[0023] Calculating the risk partial derivative coefficient of the current fiscal and tax risk according to the fiscal and tax risk analysis data, and optimizing the risk decision conditions of the big data analysis model according to the risk partial derivative coefficient;

[0024] Performing iterative convergence processing on the fiscal and tax risk data according to the optimized risk decision conditions and the number of fiscal and tax items with risks to obtain the risk decision data of the risky fiscal and tax items.

[0025] In a preferred example of the present application, it can be further configured as: calculating the risk partial derivative coefficient of the current fiscal and tax risk according to the fiscal and tax risk analysis data, and optimizing the risk decision conditions of the big data analysis model according to the risk partial derivative coefficient, specifically including:

[0026] Calculating the risk partial derivative coefficient of the current fiscal and tax risk through formula (2), and formula (2) is as follows:

[0027]

[0028] Among them, ω represents the risk partial derivative coefficient of the current fiscal and tax risk, n and m respectively represent the number of nodes where the fiscal and tax data changes and the number of fiscal and tax items with fiscal and tax data changes, and λ ij represents the constraint coefficient of the j-th node of the i-th fiscal and tax item with fiscal and tax data changes, which is related to the importance of the fiscal and tax item to the enterprise value target, and x ij and y ij represent the coordinate values of the j-th node of the i-th fiscal and tax item with fiscal and tax data changes in the coordinate axis with fiscal and tax data as the x-axis and the enterprise value of the corresponding fiscal and tax item as the y-axis;

[0029] Optimize the risk decision conditions of the big data analysis model according to the risk partial derivative coefficient. Among them, the optimization expression of the risk decision conditions is represented by formula (3), and formula (3) is as follows:

[0030]

[0031] Among them, min ω L(ω,x,y) represents the minimum constraint factor of the risk decision conditions, and λ i and λ j respectively represent the constraint coefficients of the i-th fiscal and tax item and the j-th node of the fiscal and tax data change, and x i and y i represent the coordinate values of the i-th fiscal and tax item, and x j and y j represent the coordinate values of the j-th node of the fiscal and tax item with fiscal and tax data changes.

[0032] By adopting the above technical solution, by constructing the risk correlation relationship between the fiscal and tax risk data and the enterprise external data, using the enterprise external data as the correlation reference index, jointly analyzing the distribution of the fiscal and tax risk in the fiscal and tax items, it helps to focus on monitoring the risk fiscal and tax items, and combining with the risk partial derivative coefficient of the current fiscal and tax risk, optimizing the risk decision conditions of the big data analysis model, improving the fit degree between the risk decision conditions and the current fiscal and tax risk, using the number of fiscal and tax items with risks as the number of iteration times, performing data convergence processing on the fiscal and tax risk data, so as to obtain risk decision data that is more matched with the current fiscal and tax risk, and improving the fit degree between the fiscal and tax risk decision and the current fiscal and tax risk.

[0033] In the second aspect, the above object of the present application is achieved by the following technical solution:

[0034] A financial risk analysis system based on a big data analysis model, the system is applied to the above financial risk analysis method based on a big data analysis model, and the system includes:

[0035] A data processing module, configured to obtain enterprise financial and tax data and external industry data of an enterprise within a certain time period and perform data preprocessing to obtain preprocessed internal enterprise data and external enterprise data;

[0036] A model construction module, configured to randomly screen the internal enterprise data and the external enterprise data respectively, analyze the correlation between the screened internal enterprise data and external enterprise data, and construct a big data analysis model for enterprise finance and taxation;

[0037] A risk decision-making module, configured to obtain financial and tax risk data in the internal enterprise data that exceeds a preset financial and tax item index, and input the financial and tax risk data into the big data analysis model to perform risk decision-making processing on the financial and tax risk data;

[0038] A risk analysis module, configured to perform classification and iteration processing on the financial and tax risk data according to the risk decision result, calculate a financial and tax risk decision deviation coefficient according to the iteration result and correct the risk decision result to obtain financial risk analysis data.

[0039] By adopting the above technical solution, preprocessing the enterprise financial and tax data and external industry data within the selected time period helps to filter out low-quality or duplicate and invalid data, retain high-value data with commercial value, increase the randomness of data selection by randomly screening data, and combine the correlation between internal enterprise data and external enterprise data to construct a big data model for enterprise finance and taxation, which helps to improve the correlation between internal and external enterprise data. Taking external data as the analysis index, it improves the stability of model demonstration from multiple dimensions. By screening financial and tax risk data that exceeds the preset financial and tax item index, targeted directional risk analysis is performed on the financial and tax risk data to improve the accuracy of risk decision-making. And through the classification iteration and deviation correction of the financial and tax risk data, the financial and tax risk analysis result is further corrected, and the financial and tax risk analysis result is verified multiple times from multiple angles, thereby improving the accuracy of financial risk analysis.

[0040] In a third aspect, the above object of the present application is achieved through the following technical solution:

[0041] A computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the steps of the above financial risk analysis method based on a big data analysis model are implemented.

[0042] In a fourth aspect, the above object of the present application is achieved through the following technical solution:

[0043] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the above-mentioned financial risk analysis method based on a big data analysis model are implemented.

[0044] In summary, the present application includes at least one of the following beneficial technical effects:

[0045] 1. Preprocess the enterprise's financial and tax data and external industry data within a selected time period, which helps filter low-quality or duplicate and invalid data, retain high-value data with commercial value, increase the randomness of data selection by randomly screening data, and combine the correlation between internal enterprise data and external enterprise data to construct a big data model for enterprise finance and taxation, which helps improve the relevance between internal and external enterprise data. Taking external data as an analysis indicator, the stability of model demonstration is improved from multiple dimensions. By screening financial and tax risk data exceeding the preset financial and tax project indicators, targeted risk analysis of the financial and tax risk data is carried out to improve the accuracy of risk decision-making. Through the classification iteration and deviation correction of the financial and tax risk data, the financial and tax risk analysis results are further corrected, and the financial and tax risk analysis results are verified multiple times from multiple angles, thereby improving the accuracy of financial risk analysis.

[0046] 2. Perform secondary classification on the financial and tax risk data according to the risk decision result for data selection, and randomly select the node data where the financial and tax data changes as the training data set of big data. Thus, the first data set and the second data set are obtained according to the secondary classification result, which helps verify the risk decision result by comparison. Taking the number of changes in the financial and tax data as the number of iterations, data iteration is performed on the first data set and the second data set respectively, and the financial and tax risk decision limits of the corresponding data sets are analyzed, so as to construct the risk decision conditions of the big data analysis model, constrain the risk decision of the big data analysis model, and combine the financial and tax risk decision deviation coefficient to correct the original risk decision result, further improving the accuracy of financial risk analysis.

[0047] 3. By constructing a risk correlation relationship between the financial and tax risk data and external enterprise data, taking the external enterprise data as an associated reference indicator, jointly analyze the distribution of financial and tax risks in financial and tax projects, which helps focus on monitoring risk financial and tax projects. Combining the risk partial derivative coefficient of the current financial and tax risk, optimize the risk decision conditions of the big data analysis model, improve the fit between the risk decision conditions and the current financial and tax risk. Taking the number of financial and tax projects with risks as the number of iterations, perform data convergence processing on the financial and tax risk data, so as to obtain risk decision data that is more matched with the current financial and tax risk, and improve the fit between the financial and tax risk decision and the current financial and tax risk. Description of the Drawings

[0048] Figure 1 It is the implementation flowchart of a financial risk analysis method based on a big data analysis model in this embodiment.

[0049] Figure 2 It is the implementation flowchart for optimizing decision-making conditions of the financial risk analysis method in this embodiment.

[0050] Figure 3 It is the implementation flowchart of step S40 of the financial risk analysis method in this embodiment.

[0051] Figure 4 It is the structural block diagram of a financial risk analysis system based on a big data analysis model in this embodiment.

[0052] Figure 5 It is the internal structure schematic diagram of a computer device for implementing the financial risk analysis method. Specific implementation manners

[0053] The following further elaborates on this application in conjunction with the accompanying drawings.

[0054] In one embodiment, as Figure 1 shown, this application discloses a financial risk analysis method based on a big data analysis model, which specifically includes the following steps:

[0055] S10: Obtain the enterprise's financial and tax data and external industry data within a certain time period, and perform data preprocessing to obtain the preprocessed enterprise internal data and enterprise external data.

[0056] Specifically, according to the time period selected by the enterprise, such as annual, quarterly, or monthly, etc., extract the enterprise's financial and tax data within the selected time period, and extract the external industry data of the same industry according to the nature of the enterprise. Analyze the low-value data in the extracted enterprise financial and tax data and enterprise external data, and perform filtering processing to retain the financial and tax data with commercial value, so as to obtain the preprocessed enterprise internal data and enterprise external data.

[0057] S20: Randomly screen the enterprise internal data and enterprise external data respectively, analyze the correlation relationship between the screened enterprise internal data and enterprise external data, and construct a big data analysis model for enterprise finance and taxation.

[0058] Specifically, randomly sample data samples from the enterprise internal data and enterprise external data respectively, and combine the relevance of financial and tax items in the enterprise's business transactions to analyze the correlation relationship between the screened enterprise internal data and enterprise external data, including the business transaction correlation between enterprises and the peer correlation of the same financial and tax items in peer enterprises, so as to construct a big data analysis model for enterprise finance and taxation.

[0059] S30: Obtain the financial and tax risk data in the enterprise's internal data that exceeds the preset financial and tax project indicators, and input the financial and tax risk data into the big data analysis model to perform risk decision-making processing on the financial and tax risk data.

[0060] Specifically, set the financial and tax project indicators for each financial and tax project according to the enterprise's revenue indicators, extract the financial and tax risk data in the enterprise's internal data that exceeds the preset financial and tax project indicators, input the financial and tax risk data into the big data analysis model, and perform risk decision-making processing on the financial and tax risk data. As Figure 2 shown, it specifically includes:

[0061] S301: Construct the risk correlation relationship between the financial and tax risk data and the enterprise's external data, analyze the distribution of the financial and tax risks of the financial and tax risk data according to the risk correlation relationship, and obtain the financial and tax risk analysis data.

[0062] Specifically, construct the risk correlation relationship between the financial and tax risk data and the enterprise's external data according to the correlation between the same type of financial and tax projects among peer enterprises and the financial and tax projects in the enterprise's business transactions. And according to the risk correlation relationship, analyze the distribution of the financial and tax risk data in the financial and tax projects, that is, the distribution of the financial and tax projects with financial and tax risks in business transactions and among peer enterprises, as well as the specific distribution positions of the financial and tax risk data in the financial and tax projects, so as to obtain the financial and tax risk analysis data.

[0063] S302: Calculate the risk partial derivative coefficient of the current financial and tax risk according to the financial and tax risk analysis data, and optimize the risk decision-making conditions of the big data analysis model according to the risk partial derivative coefficient.

[0064] Specifically, calculate the risk partial derivative coefficient of the current financial and tax risk through formula (2), and formula (2) is as follows:

[0065]

[0066] Among them, ω represents the risk partial derivative coefficient of the current financial and tax risk, n and m respectively represent the number of nodes where the financial and tax data changes and the number of financial and tax projects with financial and tax data changes, λ ij represents the constraint coefficient of the jth financial and tax project node with financial and tax data changes in the ith financial and tax project, which is related to the importance of the financial and tax project to the enterprise value target, x ij and y ij represent the coordinate values of the jth financial and tax project node with financial and tax data changes in the ith financial and tax project in the coordinate axis with the financial and tax data as the x-axis and the enterprise value of the corresponding financial and tax project as the y-axis.

[0067] Optimize the risk decision-making conditions of the big data analysis model according to the risk partial derivative coefficient. Among them, the optimization expression of the risk decision-making conditions is represented by formula (3), and formula (3) is as follows:

[0068]

[0069] Among them, min ω L(ω,x,y) represents the minimum constraint factor of the risk decision-making conditions, and λ i and λ j represent the constraint coefficients of the i-th fiscal and tax item and the j-th fiscal and tax data change node respectively, x i and y i represent the coordinate values of the i-th fiscal and tax item, and x j and y j represent the coordinate values of the j-th fiscal and tax item node where there is a change in fiscal and tax data.

[0070] S303: Perform iterative convergence processing on the fiscal and tax risk data according to the optimized risk decision-making conditions and the number of fiscal and tax items with risks to obtain the risk decision-making data of the risky fiscal and tax items.

[0071] Specifically, perform risk decision-making processing on the fiscal and tax risk data according to the optimized risk decision-making conditions, and use the number of fiscal and tax items with risks as the number of data iterations to perform iterative convergence processing on the fiscal and tax risk data until the optimized risk decision-making conditions are reached, thereby obtaining the risk decision-making data of the fiscal and tax risk items.

[0072] S40: Perform classification iteration processing on the fiscal and tax risk data according to the risk decision-making results, calculate the deviation coefficient of the fiscal and tax risk decision-making according to the iteration results, and correct the risk decision-making results to obtain the financial risk analysis data.

[0073] Specifically, as Figure 3 shown, step S40 includes:

[0074] S401: According to the risk decision-making results, perform secondary classification on the fiscal and tax risk data, and randomly select the node data where the fiscal and tax data changes from the fiscal and tax items of each classification data to obtain the first data set and the second data set.

[0075] Specifically, according to the risk decision-making results, perform secondary classification on the fiscal and tax risk data, classify the fiscal and tax data with fiscal and tax risks into one category, classify the normal fiscal and tax data without fiscal and tax risks into one category, and randomly select the node data where the fiscal and tax data changes from the fiscal and tax items of each classification data, such as the fiscal and tax nodes with data fluctuations, thereby obtaining the first data set and the second data set.

[0076] S402: According to the number of changes in the fiscal and tax data, perform data iteration processing on the first data set and the second data set respectively, analyze the fiscal and tax risk decision limits of the corresponding data sets, and construct the risk decision conditions of the big data analysis model.

[0077] Specifically, take the number of changes in the fiscal and tax data as the number of data iterations, perform data iteration processing on the first data set and the second data set respectively, use the data iteration results as the fiscal and tax risk decision limits of the corresponding data sets, and construct the risk decision conditions of the big data analysis model.

[0078] S403: Calculate the fiscal and tax risk decision deviation coefficient according to the difference between the risk decision conditions and the original risk decision results, and correct the original risk decision results according to the fiscal and tax risk decision deviation coefficient to obtain the financial risk data.

[0079] Specifically, calculate the fiscal and tax risk decision deviation coefficient through formula (1), and formula (1) is as follows:

[0080]

[0081] Among them, ρ represents the fiscal and tax risk decision deviation coefficient, n and m respectively represent the number of nodes where the fiscal and tax data changes and the number of fiscal and tax items with fiscal and tax data changes, and Δp ij represents the data change value of the j-th fiscal and tax node with fiscal and tax data changes in the i-th fiscal and tax item.

[0082] Correct the original risk decision results according to the fiscal and tax decision deviation coefficient. For example, calculate the fiscal and tax decision deviation value corresponding to the fiscal and tax decision deviation coefficient, and take the difference between the original risk decision value and the fiscal and tax decision deviation value as the correction result, so as to obtain the fiscal and tax risk data.

[0083] It should be understood that the magnitudes of the sequence numbers of the steps in the above embodiments do not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.

[0084] In one embodiment, a financial risk analysis system based on a big data analysis model is provided. The financial risk analysis system based on the big data analysis model corresponds one-to-one with the financial risk analysis method based on the big data analysis model in the above embodiment. As Figure 4 shown, the financial risk analysis system based on the big data analysis model includes a data processing module, a model construction module, a risk decision module, and a risk analysis module. The detailed descriptions of each functional module are as follows:

[0085] The data processing module is used to obtain the enterprise's fiscal and tax data and external industry data within a certain time period and perform data preprocessing to obtain the preprocessed enterprise internal data and enterprise external data.

[0086] A model construction module, which is used to randomly screen the internal data and external data of an enterprise respectively, analyze the correlation between the screened internal data and external data of the enterprise, and construct a big data analysis model for enterprise finance and taxation.

[0087] A risk decision-making module, which is used to obtain the finance and taxation risk data exceeding the preset finance and taxation project indicators in the internal data of the enterprise, and input the finance and taxation risk data into the big data analysis model to perform risk decision-making processing on the finance and taxation risk data.

[0088] A risk analysis module, which is used to perform classified iterative processing on the finance and taxation risk data according to the risk decision-making result, calculate the deviation coefficient of the finance and taxation risk decision-making according to the iterative result, and correct the risk decision-making result to obtain the financial risk analysis data.

[0089] Preferably, the risk analysis module specifically includes:

[0090] A data division sub-module, which is used to perform secondary classification on the finance and taxation risk data according to the risk decision-making result, and randomly select the node data where the finance and taxation data changes from the finance and taxation items of each classified data to obtain a first data set and a second data set.

[0091] A condition construction sub-module, which is used to perform data iterative processing on the first data set and the second data set respectively according to the number of changes of the finance and taxation data, analyze the risk decision-making limit values of the corresponding data sets for finance and taxation, and construct the risk decision-making conditions of the big data analysis model.

[0092] A data correction sub-module, which is used to calculate the deviation coefficient of the finance and taxation risk decision-making according to the difference between the risk decision-making conditions and the original risk decision-making result, and correct the original risk decision-making result according to the deviation coefficient of the finance and taxation risk decision-making to obtain the financial risk data.

[0093] Preferably, the data correction sub-module specifically includes:

[0094] Calculate the deviation coefficient of the finance and taxation risk decision-making through formula (1), and formula (1) is as follows:

[0095]

[0096] Among them, ρ represents the deviation coefficient of the finance and taxation risk decision-making, n and m respectively represent the number of node quantities where the finance and taxation data changes and the number of finance and taxation projects with changes in the finance and taxation data, and Δp ij represents the data change value of the j-th finance and taxation node with changes in the finance and taxation data of the i-th finance and taxation project.

[0097] Preferably, the risk decision-making module specifically includes:

[0098] A risk analysis sub-module, which is used to construct a risk association relationship between fiscal and tax risk data and enterprise external data, analyze the fiscal and tax risk distribution of fiscal and tax risk data according to the risk association relationship, and obtain fiscal and tax risk analysis data.

[0099] A condition optimization sub-module, which is used to calculate the risk partial derivative coefficient of the current fiscal and tax risk according to the fiscal and tax risk analysis data, and optimize the risk decision conditions of the big data analysis model according to the risk partial derivative coefficient.

[0100] A data iteration sub-module, which is used to perform iterative convergence processing on the fiscal and tax risk data according to the optimized risk decision conditions and the number of fiscal and tax items with risks, and obtain the risk decision data of the risky fiscal and tax items.

[0101] Preferably, the condition optimization sub-module specifically includes:

[0102] Calculate the risk partial derivative coefficient of the current fiscal and tax risk through formula (2), and formula (2) is as follows:

[0103]

[0104] Among them, ω represents the risk partial derivative coefficient of the current fiscal and tax risk, n and m respectively represent the number of nodes where the fiscal and tax data changes and the number of fiscal and tax items with fiscal and tax data changes, and λ ij represents the constraint coefficient of the j-th fiscal and tax item node with fiscal and tax data changes of the i-th fiscal and tax item, which is related to the importance of the fiscal and tax item to the enterprise value target, and x ij 、y ij represent the coordinate values of the j-th fiscal and tax item node with fiscal and tax data changes of the i-th fiscal and tax item in the coordinate axis with fiscal and tax data as the x-axis and the enterprise value of the corresponding fiscal and tax item as the y-axis.

[0105] Optimize the risk decision conditions of the big data analysis model according to the risk partial derivative coefficient. Among them, the optimization expression of the risk decision conditions is represented by formula (3), and formula (3) is as follows:

[0106]

[0107] Among them, min ω L(ω,x,y) represents the minimum constraint factor of the risk decision conditions, and λ i 、λ j respectively represent the constraint coefficients of the i-th fiscal and tax item and the j-th fiscal and tax data change node, x i 、y i represent the coordinate values of the i-th fiscal and tax item, and x j 、y j represent the coordinate values of the j-th fiscal and tax item node with fiscal and tax data changes.

[0108] For the specific limitations of the financial risk analysis system based on the big data analysis model, reference can be made to the limitations of the financial risk analysis method based on the big data analysis model in the foregoing text, which will not be elaborated here. Each module in the above-mentioned financial risk analysis system based on the big data analysis model can be implemented in whole or in part by software, hardware, and their combination. Each of the above modules can be embedded in or independent of the processor in the computer device in the form of hardware, or stored in the memory of the computer device in the form of software, so as to facilitate the processor to call and execute the operations corresponding to each of the above modules.

[0109] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as Figure 5 shown. The computer device includes a processor, a memory, a network interface, and a database connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store the financial risk analysis data of the big data analysis model. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, it implements a financial risk analysis method based on the big data analysis model.

[0110] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, it implements the steps of a financial risk analysis method based on the big data analysis model.

[0111] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above various methods. Among them, any reference to a memory, storage, database, or other medium used in the various embodiments provided in this application can include non-volatile and / or volatile memories. Non-volatile memories can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memories can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and Rambus dynamic RAM (RDRAM), etc.

[0112] Those skilled in the art can clearly understand that for the convenience and brevity of description, only the above division of each functional unit and module is used as an example for illustration. In actual applications, the above functions can be allocated to different functional units and modules as needed, that is, the internal structure of the system can be divided into different functional units or modules to complete all or part of the functions described above.

[0113] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the various embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A financial risk analysis method based on a big data analysis model, characterized in that: include: Obtain the enterprise's financial and tax data and external industry data within a certain period of time and perform data preprocessing to obtain the preprocessed internal enterprise data and external enterprise data; Randomly screening the internal data of the enterprise and the external data of the enterprise respectively, analyzing the correlation between the screened internal data of the enterprise and the external data of the enterprise, and building a big data analysis model for enterprise finance and taxation; Acquire the financial and tax risk data exceeding the preset financial and tax item indicators in the internal data of the enterprise, input the financial and tax risk data into the big data analysis model, and perform risk decision processing on the financial and tax risk data; According to the risk decision results, the financial and tax risk data are classified and iterated, and the financial and tax risk decision deviation coefficient is calculated according to the iterative results and the risk decision results are corrected to obtain financial risk analysis data; The process of classifying and iterating the financial and tax risk data according to the risk decision results, calculating the financial and tax risk decision deviation coefficient according to the iterative results and correcting the risk decision results to obtain the financial risk analysis data specifically includes: According to the risk decision results, the financial and tax risk data are secondary classified, and the node data where the financial and tax data has changed is randomly selected from the financial and tax items of each classified data according to the classification results to obtain the first data set and the second data set; According to the number of changes in the financial and tax data, respectively, the first data set and the second data set are iterated and the financial and tax risk decision limits of the corresponding data sets are analyzed to construct the risk decision conditions of the big data analysis model; According to the difference between the risk decision conditions and the original risk decision results, the financial and tax risk decision deviation coefficient is calculated, and the original risk decision results are corrected according to the financial and tax risk decision deviation coefficient to obtain financial risk data; The calculation of the financial and tax risk decision deviation coefficient based on the difference between the risk decision condition and the original risk decision result specifically includes: The deviation coefficient of financial and tax risk decision is calculated by formula (1), which is as follows: Among them, ρ represents the deviation coefficient of fiscal and tax risk decision, n and m represent the number of nodes with changed fiscal and tax data and the number of fiscal and tax items with changed fiscal and tax data, respectively, and Δp ij Indicates the data change value of the j-th fiscal and taxation node with fiscal and taxation data change of the i-th fiscal and taxation item; The step of obtaining the financial and tax risk data exceeding the preset financial and tax item indicators in the internal data of the enterprise, inputting the financial and tax risk data into the big data analysis model, and performing risk decision processing on the financial and tax risk data specifically includes: Constructing a risk association relationship between the financial and tax risk data and the enterprise's external data, analyzing the financial and tax risk distribution of the financial and tax risk data according to the risk association relationship, and obtaining financial and tax risk analysis data; Calculate the risk partial derivative coefficient of the current financial and tax risk according to the financial and tax risk analysis data, and optimize the risk decision-making conditions of the big data analysis model according to the risk partial derivative coefficient; According to the optimized risk decision conditions and the number of financial and tax projects with risks, the financial and tax risk data are iteratively converged to obtain risk decision data of risky financial and tax projects; The step of calculating the risk partial derivative coefficient of the current financial and tax risk according to the financial and tax risk analysis data and optimizing the risk decision-making conditions of the big data analysis model according to the risk partial derivative coefficient specifically includes: The risk partial derivative coefficient of the current fiscal and tax risk is calculated by formula (2), which is as follows: Among them, ω represents the risk partial derivative coefficient of the current fiscal and taxation risk, n and m represent the number of nodes with changed fiscal and taxation data and the number of fiscal and taxation items with changed fiscal and taxation data, respectively, and λ ij represents the constraint coefficient of the jth taxation item node with taxation data changes in the i-th taxation item, which is related to the importance of the taxation item to the enterprise value target. ij ,y ij Indicates the coordinate value of the jth fiscal and taxation project node with fiscal and taxation data changes in the i-th fiscal and taxation project, in the coordinate axis with fiscal and taxation data as the x-axis and the enterprise value of the corresponding fiscal and taxation project as the y-axis; The risk decision-making condition of the big data analysis model is optimized according to the risk partial derivative coefficient, wherein the optimization expression of the risk decision-making condition is represented by formula (3), and formula (3) is as follows: Among them, min ω L(ω,x,y) represents the minimum constraint factor of the risk decision condition, λ i , j They represent the constraint coefficients of the i-th fiscal and taxation item and the j-th fiscal and taxation data change node, respectively. i ,y i represents the coordinate value of the i-th fiscal and taxation item, x j ,y j Represents the coordinate value of the jth fiscal and taxation item node where fiscal and taxation data has changed.

2. A financial risk analysis system based on a big data analysis model, characterized in that: The system is applied to the financial risk analysis method based on the big data analysis model described in claim 1 above, and the system includes: The data processing module is used to obtain the enterprise's financial and tax data and external industry data within a certain period of time and perform data preprocessing to obtain the preprocessed internal enterprise data and external enterprise data; A model building module, used to randomly screen the internal data of the enterprise and the external data of the enterprise respectively, analyze the correlation between the screened internal data of the enterprise and the external data of the enterprise, and build a big data analysis model for enterprise finance and taxation; A risk decision module, used to obtain the financial and tax risk data exceeding the preset financial and tax item indicators in the internal data of the enterprise, and input the financial and tax risk data into the big data analysis model to perform risk decision processing on the financial and tax risk data; The risk analysis module is used to classify and iterate the financial and tax risk data according to the risk decision results, calculate the financial and tax risk decision deviation coefficient according to the iterative results, and correct the risk decision results to obtain financial risk analysis data; The risk analysis module specifically includes: The data partitioning submodule is used to perform secondary classification of the financial and tax risk data according to the risk decision results, and randomly select node data where the financial and tax data has changed from the financial and tax items of each classified data according to the classification results to obtain the first data set and the second data set; A condition construction submodule, for performing data iteration processing on the first data set and the second data set respectively according to the number of changes in the financial and tax data, and analyzing the financial and tax risk decision limits of the corresponding data sets, so as to construct the risk decision conditions of the big data analysis model; A data correction submodule is used to calculate the financial and tax risk decision deviation coefficient according to the difference between the risk decision condition and the original risk decision result, and to correct the original risk decision result according to the financial and tax risk decision deviation coefficient to obtain financial risk data; Wherein, the data correction submodule specifically includes: The deviation coefficient of financial and tax risk decision is calculated by formula (1), which is as follows: Among them, ρ represents the deviation coefficient of fiscal and tax risk decision, n and m represent the number of nodes with changed fiscal and tax data and the number of fiscal and tax items with changed fiscal and tax data, respectively, and Δp ij Indicates the data change value of the j-th fiscal and taxation node with fiscal and taxation data change of the i-th fiscal and taxation item; The risk decision module specifically includes: A risk analysis submodule is used to construct a risk association relationship between the financial and tax risk data and the enterprise's external data, analyze the financial and tax risk distribution of the financial and tax risk data according to the risk association relationship, and obtain financial and tax risk analysis data; A condition optimization submodule, used to calculate the risk partial derivative coefficient of the current financial and tax risk according to the financial and tax risk analysis data, and optimize the risk decision-making conditions of the big data analysis model according to the risk partial derivative coefficient; A data iteration submodule is used to perform iterative convergence processing on the financial and tax risk data according to the optimized risk decision conditions and the number of financial and tax projects with risks, so as to obtain risk decision data of risky financial and tax projects; Wherein, the condition optimization submodule specifically includes: The risk partial derivative coefficient of the current fiscal and tax risk is calculated by formula (2), which is as follows: Among them, ω represents the risk partial derivative coefficient of the current fiscal and taxation risk, n and m represent the number of nodes with changed fiscal and taxation data and the number of fiscal and taxation items with changed fiscal and taxation data, respectively, and λ ij represents the constraint coefficient of the jth taxation item node with taxation data changes in the i-th taxation item, which is related to the importance of the taxation item to the enterprise value target. ij ,y ij Indicates the coordinate value of the jth fiscal and taxation project node with fiscal and taxation data changes in the i-th fiscal and taxation project, in the coordinate axis with fiscal and taxation data as the x-axis and the enterprise value of the corresponding fiscal and taxation project as the y-axis; The risk decision-making condition of the big data analysis model is optimized according to the risk partial derivative coefficient, wherein the optimization expression of the risk decision-making condition is represented by formula (3), and formula (3) is as follows: Among them, min ω L(ω,x,y) represents the minimum constraint factor of the risk decision condition, λ i , j They represent the constraint coefficients of the i-th fiscal and taxation item and the j-th fiscal and taxation data change node, respectively. i ,y i represents the coordinate value of the i-th fiscal and taxation item, x j ,y j Represents the coordinate value of the jth fiscal and taxation item node where fiscal and taxation data has changed.

3. A computer 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 steps of the financial risk analysis method based on the big data analysis model as claimed in claim 1 are implemented.

4. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by the processor, the steps of the financial risk analysis method based on the big data analysis model as claimed in claim 1 are implemented.

Images