Financial approval data monitoring system and method based on big data

Abstract

The invention discloses a financial examination and approval data monitoring system and method based on big data, and relates to the technical field of power supply data supervision. Historical financial examination and approval data is subjected to project type division to obtain data applied and submitted currently, and the data is matched with the divided historical data to accurately calculate the type risk value of the application data; the financial approval risk is calculated in a targeted manner, and the scientificity and rationality of risk management are ensured. An approval risk sliding window is set, real-time monitoring and dynamic adjustment of financial approval risks are achieved, fluctuation of approval data can be captured in time, abnormity can be recognized, approval requirements can be adjusted, and therefore the flexibility of an approval process and the effectiveness of a risk control strategy are improved; and comprehensive risk calculation is provided in combination with the type risk value and the department association risk value. The automatic decision not only improves the efficiency and consistency of financial examination and approval, but also reduces the workload of manual examination, and ensures the uniformity and fairness of financial examination and approval standards.

Description

Technical Field

[0001] The present invention relates to the technical field of financial approval data monitoring, and in particular to a financial approval data monitoring system and method based on big data. Background Art

[0002] Financial approval plays a vital role in corporate management, which is directly related to the rationality of resource allocation, the compliance of processes and the improvement of overall operational efficiency. By optimizing the financial approval process, enterprises can ensure that the use of various resources complies with internal regulations and reduce unnecessary approval steps, thereby improving overall work efficiency. In addition, financial approval provides important support for the company's strategic planning and decision-making, helping management to allocate and supervise resources more effectively. Intelligent financial approval process optimization technology can not only greatly improve the company's operational efficiency, but also enhance the ability to fully control the use of resources and ensure the compliance and transparency of the process.

[0003] In the existing financial approval process, risk calculation is not comprehensive enough, and it is impossible to conduct a comprehensive risk assessment of various types of approval data, resulting in the inability to timely discover and effectively avoid potential financial risks. At the same time, the existing risk calculation methods are often not targeted, and fail to obtain the overall risk status of the same type of approval based on the risk value of a single approval application, resulting in difficulty in providing accurate risk warnings and prevention and control measures when dealing with complex approval scenarios and types. The inadequacy of this risk calculation method has led to the failure to fully identify high-risk approvals, and the transparency and accuracy of financial management have been affected. Therefore, there is an urgent need for a risk calculation method that can be targeted and comprehensively analyze financial approval data to ensure that risks in the approval process can be timely assessed and effectively controlled. Summary of the invention

[0004] The purpose of the present invention is to provide a financial approval data monitoring system and method based on big data to solve the problems raised in the prior art.

[0005] To achieve the above purpose, the present invention provides the following technical solution: a financial approval data monitoring method based on big data, the financial approval data monitoring method comprising the following steps:

[0006] Step S1, collecting historical financial approval data, dividing the historical financial approval data according to the project type, and generating a historical approval data set according to the divided project type;

[0007] Step S1-1, collect historical financial approval data, obtain financial approval data related to each project, and divide it according to project type;

[0008] Step S1-2: The historical financial approval data divided by project type is stored separately according to each project type, and each storage is integrated to generate a historical approval data set.

[0009] The project types of the financial approval data mainly include eight categories: expense reimbursement, purchase application, capital expenditure, budget adjustment, contract payment, salary approval, project fund application and other expense applications;

[0010] By classifying historical financial approval data by project type and generating historical approval data sets, it helps to improve the structure and efficiency of data management. After storing data by type, it is easier to query, analyze and process data, helping to identify approval patterns and potential risks in various projects. At the same time, this classified storage method can also optimize the analysis speed of the approval process.

[0011] Step S2: read the financial approval data submitted by the application, analyze and extract the related departments and project types, match the historical financial approval data in the historical approval data set according to the project type, and calculate the type risk value of the financial approval data submitted by the application through the matched historical financial approval data;

[0012] Step S2-1, read the financial approval data submitted by the application, and extract the project type and the associated department of the financial approval data submitted by the application after analysis;

[0013] Step S2-2: The project type extracted through analysis is used as the search object, and is matched with the historical approval data set to obtain the financial approval data matching the search object;

[0014] Step S2-3: Setting the entry abnormality threshold Z max , extract the abnormal items, average time for approval processing, number of approvals and number of failed approvals for the same project type from the historical financial approval data, and calculate the risk value of the financial approval data type submitted for application. The abnormal value of the abnormal items is calculated using the following formula:

[0015]

[0016] Where Z represents the outlier value of the financial approval data; A and B are the text vectors in the two approval items;

[0017] The calculation formula of the risk value of the type is as follows:

[0018]

[0019] In the formula, The project type risk value expressed as financial approval data; N e Represented as the number of abnormal entries in the financial approval data; Nt It is represented by the total number of abnormal items in the historical financial approval data; T a Indicates the actual time of processing historical financial approval data; T s Indicates the standard time for processing historical financial approval data; C a Represents the actual number of approvals of historical financial approval data; C e Represents the number of approvals expected based on historical financial approval data; R f It is represented by the number of historical financial approval data that failed to pass the approval; R t It is represented by the total number of approvals in historical financial approval data, including those that have passed and those that have not; α is represented by the weight coefficient of abnormal items; β is represented by the weight coefficient of approval processing time; γ is represented by the weight coefficient of the number of approvals; λ is represented by the weight coefficient of the proportion of failed approvals.

[0020] The abnormal entries refer to data records or behaviors that do not conform to normal standards or expectations in financial approval or other business processes.

[0021] By matching and analyzing financial approval data with historical data, the approval risk value of the application submission can be accurately assessed. By extracting project types, related departments, and calculating key indicators such as abnormal items, processing time, number of approvals, and failure ratios, the system can dynamically assess the potential risks of each approval. This risk assessment method based on historical data not only improves the accuracy and transparency of the approval process, but also helps identify potential abnormalities, reduce risks that may arise during the approval process, and ensure that approval decisions are more scientific, reasonable, and efficient.

[0022] Step S3: Obtain department data information associated with the financial approval data through the associated departments, calculate the risk value of each department, sum the risk values ​​of each department, and calculate the associated risk value of the financial approval data submitted for the application;

[0023] Step S3-1, obtaining the associated departments according to the department information contained in the financial approval data, and extracting the financial approval rejection rate and department operation compliance rate of each department;

[0024] Step S3-2: Calculate the risk value of each department based on the financial approval rejection rate and department operation compliance rate of each department, sum the associated department risk values, and calculate the associated risk value of the financial approval data submitted for application. The department risk value is calculated using the following formula:

[0025]

[0026] In the formula, It represents the risk value corresponding to department i associated with the financial approval data; D riG is the approval rejection rate corresponding to department i; ri It is represented as the operation compliance rate corresponding to department i; μ is represented as the weight coefficient of approval rejection rate of department i; η is represented as the weight coefficient of operation compliance rate of department i;

[0027] The associated risk value is calculated using the following formula:

[0028]

[0029] In the formula, It represents the total risk value of the departments associated with the financial approval data; n represents the number of departments associated with the financial approval data.

[0030] By calculating the department's operational compliance rate and approval rejection rate, you can effectively monitor and improve the department's work efficiency and quality. The calculation of the compliance rate helps identify and correct operational behaviors that do not comply with regulations, ensuring the standardization and regularization of various operational processes, while the approval rejection rate reveals possible problems or potential risks in the approval process, thereby optimizing the approval process and improving work strategies.

[0031] Step S4: Calculate the comprehensive risk value of the financial approval data submitted for application based on the type risk value and the associated risk value, and collect the comprehensive risk values ​​at each moment to generate a comprehensive risk data set;

[0032] Step S5: set an approval risk sliding window and a comprehensive risk fluctuation threshold for the comprehensive risk data set, obtain the corresponding type of financial approval risk value trend and risk value fluctuation range through the data of different approval risk sliding windows, and control the approval requirements of the corresponding type of financial approval according to the comprehensive risk fluctuation threshold and the risk value fluctuation range of the approval risk sliding window;

[0033] Step S5-1, set an approval risk sliding window for the comprehensive risk data set, use the approval risk sliding window to process the comprehensive risk data set, and the approval risk sliding window moves step by step in the order of time. Each time it slides, the approval risk sliding window moves backward from the current time to the next adjacent time, and moves in sequence. The data in the new approval risk sliding window will cover the data in the previous approval risk sliding window;

[0034] Step S5-2: Setting the comprehensive risk fluctuation threshold B for the comprehensive risk data set max , obtain the comprehensive risk value data in the approval risk sliding window in real time, and calculate the risk value fluctuation degree of the financial approval data according to the standard deviation and variance of the comprehensive risk value in the approval risk sliding window. The risk value fluctuation degree calculation uses the following formula:

[0035]

[0036] In the formula, m represents the size of the approval risk sliding window; t represents the time corresponding to the approval risk sliding window; ω represents the moving average standard deviation within the approval risk sliding window corresponding to time t; x i Represented as the data point at time t within the approval risk sliding window; It is expressed as the average value of the data in the approval risk sliding window;

[0037] When ω ≥ B max When the fluctuation value of the financial approval data submitted for application exceeds the normal fluctuation range, it sends a signal that the approval requirements for the same project type financial approval need to be adjusted;

[0038] When ω<B max When the fluctuation value of the financial approval data submitted for application is within the normal fluctuation range, it sends a signal that the approval requirements for the same project type financial approval do not need to be adjusted, and the financial approval data will continue to be approved with the approval requirements.

[0039] Setting the approval risk sliding window and comprehensive risk fluctuation threshold can monitor and adjust the risk management of financial approval in real time. By calculating and analyzing the fluctuation range, abnormal fluctuations can be identified in time and the approval requirements can be adjusted accordingly, thereby preventing potential high-risk situations, improving the flexibility and accuracy of the approval process, optimizing risk control, and reducing potential financial risks.

[0040] Step S6: Set a comprehensive risk threshold, obtain new data points in the approval risk sliding window, and use the new data points and the comprehensive risk threshold to determine whether the financial approval application meets the approval conditions.

[0041] The newly added data point represents a data point that newly enters the approval risk sliding window at the current moment when the approval risk sliding window moves. The data point is specifically a comprehensive risk value of the data for financial approval submitted for application.

[0042] The comprehensive risk value of the financial approval data is calculated based on the calculated type risk value and associated risk value. The comprehensive risk value calculation uses the following formula:

[0043]

[0044] In the formula, Expressed as a comprehensive risk value of financial approval data; Project type risk value expressed as financial approval data; It is represented by the total risk value of the department associated with the financial approval data; θ is represented by the weight coefficient of the type risk value; It is expressed as a weight coefficient of the associated risk value;

[0045] Obtain the comprehensive risk value of the financial approval data at each moment, store the comprehensive risk values ​​at each moment accordingly to generate a comprehensive risk data set.

[0046] Setting comprehensive risk thresholds Get the new data point D in the approval risk sliding window new ;

[0047] when When the risk value of the financial approval data is greater than the risk value set in the approval condition, it is judged that the financial approval data submitted by the application cannot be approved and is rejected;

[0048] when When the risk value of the financial approval data is less than the risk value set in the approval conditions, the financial approval data submitted by the application is judged to be eligible for approval and is approved.

[0049] A financial approval data monitoring system based on big data, the financial approval data monitoring system includes an approval data acquisition module, a type extraction and matching module, an approval risk calculation module, an approval risk data storage module, an approval risk sliding window module and an approval judgment module;

[0050] The approval data collection module is used to collect financial approval data, including the financial approval data submitted by the application and the historical financial approval data; the type extraction and matching module is used to extract the types in the financial approval data, and to match all the data of the same type of financial approval data in the database according to the types; the approval risk calculation module is used to calculate the risk value of the financial approval data; the approval risk data storage module is used to store the risk value of the financial approval data calculated by the approval risk calculation module; the approval risk sliding window module is used to create and manage the approval risk sliding window; the approval judgment module is used to judge the financial approval data submitted by the application, and to identify and alarm according to the fluctuation of the comprehensive risk value of the financial approval data, and to issue an alarm signal for the financial approval data that meets the conditions;

[0051] The output end of the approval data acquisition module is electrically connected to the input end of the type extraction and matching module, the output end of the type extraction and matching module is electrically connected to the input end of the approval risk calculation module, the output end of the approval risk calculation module is electrically connected to the input end of the approval risk data storage module, the output end of the approval risk data storage module is electrically connected to the input end of the approval risk sliding window module, and the output end of the approval risk sliding window module is electrically connected to the input end of the approval judgment module.

[0052] The approval data acquisition module includes a financial approval acquisition unit and a financial approval data transmission unit; the financial approval acquisition unit is used to acquire data information of financial approval data; the financial approval data transmission unit is used to transmit the acquired financial approval data to other modules;

[0053] The type extraction and matching module includes a feature extraction unit, a type matching unit and a type feedback unit; the feature extraction unit is used to extract keyword features from the financial approval data submitted by the application; the type matching unit is used to match the type of the financial approval data; and the type feedback unit is used to feedback the matching result;

[0054] The approval risk calculation module includes a risk weight allocation unit, a risk type allocation unit and a risk calculation unit; the risk weight allocation unit is used to allocate weights to different types of risks; the risk type allocation unit is used to classify risks according to their characteristics; and the risk calculation unit is used to calculate the risk value of each type;

[0055] The approval risk data storage module includes a risk data storage unit and a risk grading unit; the risk data storage unit is used to store risk data of financial approval data; the risk grading unit is used to grade the risk level according to the risk value of the financial approval data;

[0056] The approval risk sliding window module includes a sliding window management unit, a sliding window definition unit and a sliding window translation unit; the sliding window management unit is used to control the creation and operation of the sliding window management unit; the sliding window definition unit is used to set the window size of the sliding window; the sliding window translation unit is used to control the translation of the sliding window;

[0057] The approval judgment module includes an approval alarm unit and an approval judgment unit; the approval alarm unit is used to issue an alarm signal to the financial approval data that meets the conditions; the approval judgment unit is used to judge whether the financial approval data submitted by the application meets the approval conditions.

[0058] Compared with the prior art, the present invention has the following beneficial effects:

[0059] 1. The present invention divides the historical financial approval data into project types, and uses the financial approval data submitted by the current application to match the divided historical financial approval data, so as to accurately calculate the type risk value of the applied financial approval data. This risk assessment method based on historical data not only improves the accuracy of the approval process, but also can perform targeted risk calculations according to the type when dealing with different types of financial approvals, thus ensuring the scientificity and rationality of risk management.

[0060] 2. The present invention realizes real-time monitoring and dynamic adjustment of financial approval risks by setting an approval risk sliding window. The sliding window mechanism can timely capture the fluctuation of approval data, and the fluctuation threshold helps the system identify abnormal fluctuations and adjust the approval requirements accordingly. This dynamic adjustment mechanism improves the flexibility of the approval process and can effectively respond to changes in risks, thereby optimizing the approval process and risk control strategy.

[0061] 3. The present invention can automatically determine whether the financial approval application submitted meets the approval conditions by calculating the comprehensive risk value. This comprehensive assessment method combines the type risk value and the department-related risk value, provides a comprehensive risk perspective, and improves the efficiency and consistency of approval through an automatic decision support system. This automated decision not only reduces the workload of manual review, but also ensures the uniformity and fairness of approval standards. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 A flowchart of a financial approval data monitoring method based on big data according to the present invention;

[0063] Figure 2 It is a structural schematic diagram of a financial approval data monitoring system based on big data of the present invention. DETAILED DESCRIPTION

[0064] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0065] Embodiment 1: Figure 1 As shown, the present invention provides a technical solution, a financial approval data monitoring method based on big data, and the financial approval data monitoring method comprises the following steps:

[0066] Step S1, collecting historical financial approval data, dividing the historical financial approval data according to the project type, and generating a historical approval data set according to the divided project type;

[0067] Step S1-1, collect historical financial approval data, obtain financial approval data related to each project, and divide it according to project type;

[0068] Step S1-2: The historical financial approval data divided by project type is stored separately according to each project type, and each storage is integrated to generate a historical approval data set.

[0069] Step S2: read the financial approval data submitted by the application, analyze and extract the related departments and project types, match the historical financial approval data in the historical approval data set according to the project type, and calculate the type risk value of the financial approval data submitted by the application through the matched historical financial approval data;

[0070] Step S2-1, read the financial approval data submitted by the application, and extract the project type and the associated department of the financial approval data submitted by the application after analysis;

[0071] Step S2-2: The project type extracted through analysis is used as the search object, and is matched with the historical approval data set to obtain the financial approval data matching the search object;

[0072] Step S2-3: Setting the entry abnormality threshold Z max , extract the abnormal items, average time for approval processing, number of approvals and number of failed approvals for the same project type from the historical financial approval data, and calculate the risk value of the financial approval data type submitted for application. The abnormal value of the abnormal items is calculated using the following formula:

[0073]

[0074] Where Z represents the outlier value of the financial approval data; A and B are the text vectors in the two approval items;

[0075] For example, assuming that the text vectors of item A and item B are A[0.2, 0.3, 0.5] and B[0.1, 0.4, 0.4] respectively, we can calculate using the formula:

[0076]

[0077] The calculation formula of the risk value of the type is as follows:

[0078]

[0079] In the formula, The project type risk value expressed as financial approval data; N e Represented as the number of abnormal entries in the financial approval data; N t It is represented by the total number of abnormal items in the historical financial approval data; T a Indicates the actual time of processing historical financial approval data; T s Indicates the standard time for processing historical financial approval data; C a Represents the actual number of approvals of historical financial approval data; C e Represents the number of approvals expected based on historical financial approval data; R f It is represented by the number of historical financial approval data that failed to pass the approval; Rt It is represented by the total number of approvals in historical financial approval data, including those that have passed and those that have not; α is represented by the weight coefficient of abnormal items; β is represented by the weight coefficient of approval processing time; γ is represented by the weight coefficient of the number of approvals; λ is represented by the weight coefficient of the proportion of failed approvals.

[0080] For example, the financial approval data obtained for application submission belongs to the expense reimbursement type, and the specific data is as follows:

[0081] Unusual entry: Existing N e =0, historical abnormal entries N t =7 items, weight coefficient α=0.4;

[0082] Approval processing time: historical actual processing time T a = 120 hours, the historical expected processing time is T s = 100 hours, weight coefficient β = 0.3;

[0083] Number of approvals: Historical actual number of approvals C a = 30 times, historical expected approval times C e =25 times, weight coefficient γ=0.2;

[0084] Approval failure rate: The number of historical financial approval data that failed R f =10, the total number of historical financial approvals R t =100, weight coefficient λ=0.1;

[0085] According to the formula, we can get:

[0086]

[0087] The risk value of the type of financial approval data submitted for application can be calculated is 0.31.

[0088] Step S3: Obtain department data information associated with the financial approval data through the associated departments, calculate the risk value of each department, sum the risk values ​​of each department, and calculate the associated risk value of the financial approval data submitted for the application;

[0089] Step S3-1, obtaining the associated departments according to the department information contained in the financial approval data, and extracting the financial approval rejection rate and department operation compliance rate of each department;

[0090] Step S3-2: Calculate the risk value of each department based on the financial approval rejection rate and department operation compliance rate of each department, sum the associated department risk values, and calculate the associated risk value of the financial approval data submitted for application. The department risk value is calculated using the following formula:

[0091]

[0092] In the formula, It represents the risk value corresponding to department i associated with the financial approval data; D ri G is the approval rejection rate corresponding to department i; ri It is represented as the operation compliance rate corresponding to department i; μ is represented as the weight coefficient of approval rejection rate of department i; η is represented as the weight coefficient of operation compliance rate of department i;

[0093] The associated risk value is calculated using the following formula:

[0094]

[0095] In the formula, It represents the total risk value of the departments associated with the financial approval data; n represents the number of departments associated with the financial approval data.

[0096] For example, the financial approval data submitted for application is associated with Department A, Department B, and Department C;

[0097] Department A: Financial approval rejection rate D ri =8%, department operation compliance rate G ri =92%, the weight coefficient of the approval rejection rate of Department A μ = 0.3, and the weight coefficient of the operational compliance rate of Department A η = 0.7;

[0098] Department B: Financial approval rejection rate D ri =12%, department operation compliance rate G ri =88%, the weight coefficient of approval rejection rate of department B μ = 0.3, the weight coefficient of operation compliance rate of department B η = 0.7;

[0099] Department C: Financial approval rejection rate D ri =5%, department operation compliance rate G ri =95%, the weight coefficient of approval rejection rate of department C μ = 0.3, the weight coefficient of operation compliance rate of department C η = 0.7;

[0100] According to the formula, we can get:

[0101]

[0102]

[0103]

[0104] The associated risk value is calculated as follows:

[0105]

[0106] Step S4: Calculate the comprehensive risk value of the financial approval data submitted for application based on the type risk value and the associated risk value, and collect the comprehensive risk values ​​at each moment to generate a comprehensive risk data set;

[0107] Step S5: set an approval risk sliding window and a comprehensive risk fluctuation threshold for the comprehensive risk data set, obtain the corresponding type of financial approval risk value trend and risk value fluctuation range through the data of different approval risk sliding windows, and control the approval requirements of the corresponding type of financial approval according to the comprehensive risk fluctuation threshold and the risk value fluctuation range of the approval risk sliding window;

[0108] Step S5-1, set an approval risk sliding window for the comprehensive risk data set, use the approval risk sliding window to process the comprehensive risk data set, and the approval risk sliding window moves step by step in the order of time. Each time it slides, the approval risk sliding window moves backward from the current time to the next adjacent time, and moves in sequence. The data in the new approval risk sliding window will cover the data in the previous approval risk sliding window;

[0109] Step S5-2: Setting the comprehensive risk fluctuation threshold B for the comprehensive risk data set max , obtain the comprehensive risk value data in the approval risk sliding window in real time, and calculate the risk value fluctuation degree of the financial approval data according to the standard deviation and variance of the comprehensive risk value in the approval risk sliding window. The risk value fluctuation degree calculation uses the following formula:

[0110]

[0111] In the formula, m represents the size of the approval risk sliding window; t represents the time corresponding to the approval risk sliding window; ω represents the moving average standard deviation within the approval risk sliding window corresponding to time t; x i Represented as the data point at time t within the approval risk sliding window; It is expressed as the average value of the data in the approval risk sliding window;

[0112] When ω ≥ B max When the fluctuation value of the financial approval data submitted for application exceeds the normal fluctuation range, it sends a signal that the approval requirements for the same project type financial approval need to be adjusted;

[0113] When ω<B max When the fluctuation value of the financial approval data submitted for application is within the normal fluctuation range, it sends a signal that the approval requirements for the same project type financial approval do not need to be adjusted, and the financial approval data will continue to be approved with the approval requirements.

[0114] For example, set the approval risk sliding window size m = 5, and the data points are: xt-4 =17,x t-3 =14,x t-2 =15,x t-1 =19,x t =16, set the comprehensive risk fluctuation threshold B of the comprehensive risk data set max =2;

[0115] Calculate the average value of the comprehensive risk value within the approval risk sliding window according to the formula:

[0116]

[0117] That is, 1.71<2, ω<B max The fluctuation value of the financial approval data submitted for the application is within the normal fluctuation range, which sends a signal that the approval requirements for the same project type financial approval do not need to be adjusted, and the approval requirements for the financial approval data continue to be maintained.

[0118] Step S6: Set a comprehensive risk threshold, obtain new data points in the approval risk sliding window, and use the new data points and the comprehensive risk threshold to determine whether the financial approval application meets the approval conditions.

[0119] The comprehensive risk value of the financial approval data is calculated based on the calculated type risk value and associated risk value. The comprehensive risk value calculation uses the following formula:

[0120]

[0121] In the formula, Expressed as a comprehensive risk value of financial approval data; Project type risk value expressed as financial approval data; It is represented by the total risk value of the department associated with the financial approval data; θ is represented by the weight coefficient of the type risk value; It is expressed as a weight coefficient of the associated risk value;

[0122] Obtain the comprehensive risk value of the financial approval data at each moment, store the comprehensive risk values ​​at each moment accordingly to generate a comprehensive risk data set.

[0123] Setting comprehensive risk thresholds Get the new data point D in the approval risk sliding window new ;

[0124] when When the risk value of the financial approval data is greater than the risk value set in the approval condition, it is judged that the financial approval data submitted by the application cannot be approved and is rejected;

[0125] when When the risk value of the financial approval data is less than the risk value set in the approval conditions, the financial approval data submitted by the application is judged to be eligible for approval and is approved.

[0126] For example, according to the above calculation, the type risk value The weight coefficient of the type risk value is θ = 0.3, and the associated risk value Weight coefficient of associated risk value Setting comprehensive risk thresholds is 20;

[0127] The comprehensive risk value of the financial approval data submitted for application is calculated according to the formula as follows:

[0128]

[0129] New data point D in the approval risk sliding window new is the comprehensive risk value of the new financial approval data, that is

[0130] The risk value of the financial approval data is less than the risk value set in the approval conditions. The financial approval data submitted by the application is judged to be eligible for approval and is approved.

[0131] Embodiment 2, as Figure 2 As shown, the present invention provides a financial approval data monitoring system based on big data, which includes an approval data acquisition module, a type extraction and matching module, an approval risk calculation module, an approval risk data storage module, an approval risk sliding window module and an approval judgment module;

[0132] The approval data collection module is used to collect financial approval data, including the financial approval data submitted by the application and the historical financial approval data; the type extraction and matching module is used to extract the types in the financial approval data, and to match all the data of the same type of financial approval data in the database according to the types; the approval risk calculation module is used to calculate the risk value of the financial approval data; the approval risk data storage module is used to store the risk value of the financial approval data calculated by the approval risk calculation module; the approval risk sliding window module is used to create and manage the approval risk sliding window; the approval judgment module is used to judge the financial approval data submitted by the application, and to identify and alarm according to the fluctuation of the comprehensive risk value of the financial approval data, and to issue an alarm signal for the financial approval data that meets the conditions;

[0133] The output end of the approval data acquisition module is electrically connected to the input end of the type extraction and matching module, the output end of the type extraction and matching module is electrically connected to the input end of the approval risk calculation module, the output end of the approval risk calculation module is electrically connected to the input end of the approval risk data storage module, the output end of the approval risk data storage module is electrically connected to the input end of the approval risk sliding window module, and the output end of the approval risk sliding window module is electrically connected to the input end of the approval judgment module.

[0134] The approval data acquisition module includes a financial approval acquisition unit and a financial approval data transmission unit; the financial approval acquisition unit is used to acquire data information of financial approval data; the financial approval data transmission unit is used to transmit the acquired financial approval data to other modules;

[0135] The type extraction and matching module includes a feature extraction unit, a type matching unit and a type feedback unit; the feature extraction unit is used to extract keyword features from the financial approval data submitted by the application; the type matching unit is used to match the type of the financial approval data; and the type feedback unit is used to feedback the matching result;

[0136] The approval risk calculation module includes a risk weight allocation unit, a risk type allocation unit and a risk calculation unit; the risk weight allocation unit is used to allocate weights to different types of risks; the risk type allocation unit is used to classify risks according to their characteristics; and the risk calculation unit is used to calculate the risk value of each type;

[0137] The approval risk data storage module includes a risk data storage unit and a risk grading unit; the risk data storage unit is used to store risk data of financial approval data; the risk grading unit is used to grade the risk level according to the risk value of the financial approval data;

[0138] The approval risk sliding window module includes a sliding window management unit, a sliding window definition unit and a sliding window translation unit; the sliding window management unit is used to control the creation and operation of the sliding window management unit; the sliding window definition unit is used to set the window size of the sliding window; the sliding window translation unit is used to control the translation of the sliding window;

[0139] The approval judgment module includes an approval alarm unit and an approval judgment unit; the approval alarm unit is used to issue an alarm signal to the financial approval data that meets the conditions; the approval judgment unit is used to judge whether the financial approval data submitted by the application meets the approval conditions.

[0140] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. A financial approval data monitoring method based on big data, characterized by: The financial approval data monitoring method comprises the following steps: Step S1, collecting historical financial approval data, dividing the historical financial approval data according to the project type, and generating a historical approval data set according to the divided project type; Step S2: read the financial approval data submitted by the application, analyze and extract the related departments and project types, match the historical financial approval data in the historical approval data set according to the project type, and calculate the type risk value of the financial approval data submitted by the application through the matched historical financial approval data; Step S3: Obtain department data information associated with the financial approval data through the associated departments, calculate the risk value of each department, sum the risk values ​​of each department, and calculate the associated risk value of the financial approval data submitted for the application; Step S4: Calculate the comprehensive risk value of the financial approval data submitted for application based on the type risk value and the associated risk value, and collect the comprehensive risk values ​​at each moment to generate a comprehensive risk data set; Step S5: set an approval risk sliding window and a comprehensive risk fluctuation threshold for the comprehensive risk data set, obtain the corresponding type of financial approval risk value trend and risk value fluctuation range through the data of different approval risk sliding windows, and control the approval requirements of the corresponding type of financial approval according to the comprehensive risk fluctuation threshold and the risk value fluctuation range of the approval risk sliding window; Step S6: Set a comprehensive risk threshold, obtain new data points in the approval risk sliding window, and use the new data points and the comprehensive risk threshold to determine whether the financial approval application meets the approval conditions.

2. The method for monitoring financial approval data based on big data according to claim 1, characterized in that: The specific steps of step S1 are as follows: Step S1-1, collect historical financial approval data, obtain financial approval data related to each project, and divide it according to project type; Step S1-2: The historical financial approval data divided by project type is stored separately according to each project type, and each storage is integrated to generate a historical approval data set.

3. The method for monitoring financial approval data based on big data according to claim 2 is characterized in that: The specific steps of step S2 are as follows: Step S2-1, read the financial approval data submitted by the application, and extract the project type and the associated department of the financial approval data submitted by the application after analysis; Step S2-2: The project type extracted through analysis is used as the search object, and is matched with the historical approval data set to obtain the financial approval data matching the search object; Step S2-3: Setting the entry abnormality threshold Z max , extract the abnormal items, average time for approval processing, number of approvals and number of failed approvals for the same project type from the historical financial approval data, and calculate the risk value of the financial approval data type submitted for application. The abnormal value of the abnormal items is calculated using the following formula: Where Z represents the outlier value of the financial approval data; A and B are the text vectors in the two approval items; The calculation formula of the risk value of the type is as follows: In the formula, The project type risk value expressed as financial approval data; N e Represented as the number of abnormal entries in the financial approval data; N t It is expressed as the total number of abnormal items in the historical financial approval data; T a Indicates the actual time for processing historical financial approval data; T s Indicates the standard time for processing historical financial approval data; C a Represents the actual number of approvals of historical financial approval data; C e Represents the number of approvals expected based on historical financial approval data; R f It is represented by the number of historical financial approval data that failed to pass the approval; R t It is represented by the total number of approvals in historical financial approval data, including those that have passed and those that have not; α is represented by the weight coefficient of abnormal items; β is represented by the weight coefficient of approval processing time; γ is represented by the weight coefficient of the number of approvals; λ is represented by the weight coefficient of the proportion of failed approvals.

4. The method for monitoring financial approval data based on big data according to claim 3 is characterized by: The specific steps of step S3 are as follows: Step S3-1, obtaining the associated departments according to the department information contained in the financial approval data, and extracting the financial approval rejection rate and department operation compliance rate of each department; Step S3-2: Calculate the risk value of each department based on the financial approval rejection rate and department operation compliance rate of each department, sum the associated department risk values, and calculate the associated risk value of the financial approval data submitted for application. The department risk value is calculated using the following formula: In the formula, It represents the risk value corresponding to department i associated with the financial approval data; D ri G is the approval rejection rate corresponding to department i; ri It is represented as the operation compliance rate corresponding to department i; μ is represented as the weight coefficient of approval rejection rate of department i; η is represented as the weight coefficient of operation compliance rate of department i; The associated risk value is calculated using the following formula: In the formula, It represents the total risk value of the departments associated with the financial approval data; n represents the number of departments associated with the financial approval data.

5. The method for monitoring financial approval data based on big data according to claim 4 is characterized in that: In step S4, the comprehensive risk value of the financial approval data is calculated according to the calculated type risk value and associated risk value. The comprehensive risk value calculation uses the following formula: In the formula, Expressed as a comprehensive risk value of financial approval data; Project type risk value expressed as financial approval data; Indicates the total risk value of the department associated with the financial approval data; θ represents the weight coefficient of the type risk value; It is expressed as a weight coefficient of the associated risk value; Obtain the comprehensive risk value of the financial approval data at each moment, store the comprehensive risk values ​​at each moment accordingly to generate a comprehensive risk data set.

6. The method for monitoring financial approval data based on big data according to claim 5 is characterized by: The specific steps of step S5 are as follows: Step S5-1, set an approval risk sliding window for the comprehensive risk data set, use the approval risk sliding window to process the comprehensive risk data set, and the approval risk sliding window moves step by step in the order of time. Each time it slides, the approval risk sliding window moves backward from the current time to the next adjacent time, and moves in sequence. The data in the new approval risk sliding window will cover the data in the previous approval risk sliding window; Step S5-2: Setting the comprehensive risk fluctuation threshold B for the comprehensive risk data set max , obtain the comprehensive risk value data in the approval risk sliding window in real time, and calculate the risk value fluctuation degree of the financial approval data according to the standard deviation and variance of the comprehensive risk value in the approval risk sliding window. The risk value fluctuation degree calculation uses the following formula: In the formula, m represents the size of the approval risk sliding window; t represents the time corresponding to the approval risk sliding window; ω represents the moving average standard deviation within the approval risk sliding window corresponding to time t; x i Represented as the data point at time t within the approval risk sliding window; It is expressed as the average value of the data in the approval risk sliding window; When ω ≥ B max When the fluctuation value of the financial approval data submitted for application exceeds the normal fluctuation range, it sends a signal that the approval requirements for the same project type financial approval need to be adjusted; When ω<B max When the fluctuation value of the financial approval data submitted for application is within the normal fluctuation range, it sends a signal that the approval requirements for the same project type financial approval do not need to be adjusted, and the financial approval data will continue to be approved with the approval requirements.

7. A financial approval data monitoring method based on big data according to claim 6, characterized in that: In step S6, a comprehensive risk threshold is set. Get the new data point D in the approval risk sliding window new ; when When the risk value of the financial approval data is greater than the risk value set in the approval condition, it is judged that the financial approval data submitted by the application cannot be approved and is rejected; when When the risk value of the financial approval data is less than the risk value set in the approval conditions, the financial approval data submitted by the application is judged to be eligible for approval and is approved.

8. A financial approval data monitoring system based on big data, which is applied to a financial approval data monitoring method based on big data as claimed in any one of claims 1 to 7, characterized in that: The financial approval data monitoring system includes an approval data acquisition module, a type extraction and matching module, an approval risk calculation module, an approval risk data storage module, an approval risk sliding window module and an approval judgment module; The approval data collection module is used to collect financial approval data, including the financial approval data submitted for application and historical financial approval data; the type extraction and matching module is used to extract the types in the financial approval data, and to match all data of the same type of financial approval data in the database according to the types; the approval risk calculation module is used to calculate the risk value of the financial approval data; The approval risk data storage module is used to store the financial approval data risk value calculated by the approval risk calculation module; The approval risk sliding window module is used to create and manage the approval risk sliding window; The approval judgment module is used to judge the financial approval data submitted by the application, identify and alarm according to the fluctuation of the comprehensive risk value of the financial approval data, and issue an alarm signal for the financial approval data that meets the conditions; The output end of the approval data acquisition module is electrically connected to the input end of the type extraction and matching module, the output end of the type extraction and matching module is electrically connected to the input end of the approval risk calculation module, the output end of the approval risk calculation module is electrically connected to the input end of the approval risk data storage module, the output end of the approval risk data storage module is electrically connected to the input end of the approval risk sliding window module, and the output end of the approval risk sliding window module is electrically connected to the input end of the approval judgment module.

9. A financial approval data monitoring system based on big data according to claim 8, characterized in that: The financial approval data monitoring system includes: The approval data acquisition module includes a financial approval acquisition unit and a financial approval data transmission unit; the financial approval acquisition unit is used to acquire data information of financial approval data; the financial approval data transmission unit is used to transmit the acquired financial approval data to other modules; The type extraction and matching module includes a feature extraction unit, a type matching unit and a type feedback unit; the feature extraction unit is used to extract keyword features from the financial approval data submitted by the application; the type matching unit is used to match the type of the financial approval data; and the type feedback unit is used to feedback the matching result; The approval risk calculation module includes a risk weight allocation unit, a risk type allocation unit and a risk calculation unit; the risk weight allocation unit is used to allocate weights to different types of risks; the risk type allocation unit is used to classify risks according to their characteristics; and the risk calculation unit is used to calculate the risk value of each type.

10. A financial approval data monitoring system based on big data according to claim 8, characterized in that: The financial approval data monitoring system also includes: The approval risk data storage module includes a risk data storage unit and a risk classification unit; the risk data storage unit is used to store risk data of financial approval data; The risk grading unit is used to grade the risk level according to the risk value of the financial approval data; The approval risk sliding window module includes a sliding window management unit, a sliding window definition unit and a sliding window translation unit; The sliding window management unit is used to control the creation and operation of the sliding window management unit; the sliding window definition unit is used to set the window size of the sliding window; the sliding window translation unit is used to control the translation of the sliding window; The approval judgment module includes an approval alarm unit and an approval judgment unit; the approval alarm unit is used to issue an alarm signal to the financial approval data that meets the conditions; the approval judgment unit is used to judge whether the financial approval data submitted by the application meets the approval conditions.

Images