A kind of audit monitoring method and system based on big data

Abstract

The present application relates to the field of data processing technology, and in particular to an audit monitoring method and system based on big data, the method comprising: collecting the time series within the target time window of each financial data, and using the Granger causality test to determine the causal relationship between each financial data; obtaining the measured trend value of each financial data time series; calculating the predicted trend value of each financial data; in response to the trend deviation of any financial data being greater than a preset threshold, the monitoring result is abnormal, otherwise, the monitoring result is normal, and the trend deviation is the absolute value of the difference between the predicted trend value and the measured trend value. Through the technical solution of the present application, it is possible to integrate the mutual influence between multiple financial data and obtain accurate monitoring results in the process of financial data auditing.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to an audit monitoring method and system based on big data. Background Art

[0002] Financial data audit is a systematic examination of a company's financial statements and related records, designed to assess the accuracy and completeness of the company's financial statements and locate abnormal data in the financial data. The workload of the audit process is huge, and in order to improve the efficiency of the audit work, it is necessary to use artificial intelligence to monitor the audit process.

[0003] At present, a patent application document with application publication number CN118094451A discloses an audit monitoring system based on big data, including: a data acquisition module, used to collect a number of financial data; a benchmark K value acquisition module, used to obtain a number of cluster sets according to the distance changes between the several financial data; obtain the benchmark K value of each financial data according to the number of data in each cluster set; an optimized K value and deviation acquisition module, used to obtain the K value adjustment coefficient of each financial data according to the distance between each financial data and the geometric center of the cluster set to which it belongs, the shape of the cluster set to which each financial data belongs, and the discreteness of the distance between the data in the cluster set to which it belongs; use the K value adjustment coefficient of each financial data to correct the benchmark K value to obtain the optimized K value of each financial data, and obtain the neighborhood data of each financial data according to the optimized K value of each financial data; obtain the deviation of each financial data according to the distribution shape of all neighborhood data of each financial data; an anomaly detection module, used to realize anomaly detection according to the neighborhood data and deviation of each financial data.

[0004] The above method performs anomaly detection based on the distance between several financial data to determine whether each financial data is normal. However, in the financial audit process, various financial data affect each other, such as sales volume and accounts receivable. Changes in sales volume will cause changes in accounts receivable. Ignoring the mutual influence between various financial data will lead to inaccurate monitoring results in the financial data audit process. Summary of the invention

[0005] In order to solve the technical problem of inaccurate monitoring results during the financial data audit process, the present application provides an audit monitoring method and system based on big data, which can integrate the mutual influence between various financial data and obtain accurate monitoring results during the financial data audit process.

[0006] In the first aspect of the present application, an audit monitoring method based on big data is provided, the monitoring method comprising: collecting time series within a target time window of each financial data, and using Granger causality test to determine the causal relationship between each financial data; obtaining the measured trend value of each financial data time series; calculating the predicted trend value of each financial data, and the predicted trend value of each financial data. The predicted trend value of for:

[0007] , is the number of financial data, Representing financial data Financial data The causal relationship, For financial data ; in response to the trend deviation of any financial data being greater than a preset threshold, the monitoring result is abnormal, otherwise, the monitoring result is normal, and the trend deviation is the absolute value of the difference between the predicted trend value and the measured trend value.

[0008] Granger causality test is used to determine the causal relationship between financial data. Financial data The causal relationship can reflect the financial data Prior changes can effectively predict financial data Changes in financial data Changes from previous periods to financial data The influence of changing trends; the measured trend value of each financial data is obtained according to the time series within the target time window of each financial data, and the predicted trend value of each financial data is calculated according to the causal relationship between the financial data; when the trend deviation of any financial data is greater than the preset threshold, the monitoring result is abnormal, otherwise, the monitoring result is normal. By combining the mutual influence among various financial data, accurate monitoring results are obtained.

[0009] Preferably, the financial data is numerical data, and the financial data includes fixed assets, investment assets, accounts receivable, income and administrative expenses.

[0010] Preferably, each financial data corresponds to a target time window, the target time window includes a current collection period and multiple collection periods before the current collection period, and the time series of the financial data includes the values ​​of the financial data of each collection period within the target time window.

[0011] Due to the different periodicity and change trends of financial data, in order to truly and accurately collect the measured trend values ​​of each financial data, a target time window is set for each financial data to improve the accuracy of the audit monitoring results.

[0012] Preferably, the method for acquiring the target time window includes: acquiring the variation period of each financial data by using an autocorrelation function, and setting the variation period as the target time window of the corresponding financial data.

[0013] Preferably, financial data Financial data Causal relationship Methods for obtaining include: establishing financial data The first autoregressive model, the input of the first autoregressive model is financial data The values ​​of each collection period before the current collection period in the time series are output as the financial data of the current collection period The value of; establish the second autoregressive model, the input of the second autoregressive model is financial data and financial data The values ​​of each collection period before the current collection period in the time series are output as the financial data of the current collection period The absolute value of the difference between the error terms in the first autoregressive model and the second autoregressive model is used as the financial data Financial data Causal relationship .

[0014] The error term in the first autoregressive model only considers financial data The value of the past collection period affects the financial data in the current collection period. The error term in the second autoregressive model takes into account the impact of financial data and financial data The value of the past collection period affects the financial data in the current collection period. Therefore, the absolute value of the difference between the error terms in the first autoregressive model and the second autoregressive model can reflect the impact of considering financial data. Changes from previous periods to financial data The impact of the value of can be achieved by accurately quantifying the causal relationship.

[0015] Preferably, financial data The first autoregressive model is:

[0016] ,

[0017] The second autoregressive model satisfies the relationship:

[0018] ,

[0019] in, It is financial data The autoregression coefficient of For financial data The values ​​of each collection period before the current collection period in the time series, is the error term of the first autoregressive model, It is financial data The autoregression coefficient of For financial data The values ​​of each collection period before the current collection period in the time series, is the error term of the second autoregressive model, Collect financial data for the current period The value of and Financial data and financial data The length of the target time window.

[0020] Preferably, financial data Financial data Causal relationship The methods for obtaining include: If the financial data is judged based on the Granger causality test Financial data If there is a causal relationship, then Set to 1, otherwise, Set to 0.

[0021] Preferably, financial data The measured trend value for:

[0022] ; and is a set of value pairs, which are financial data Number in the time series and number The value of Represents obtaining financial data The time series meets All value pairs of is the mean value function.

[0023] Through financial data Calculate financial data for all value pairs in the time series Measured trend values, financial data A large number of value pairs can be obtained from the time series. By taking the average value, the influence of abnormal points in the time series on the measured trend value can be avoided, and the financial data can be realized. Accurate quantification of the measured trend values.

[0024] Preferably, financial data The measured trend value The methods for obtaining include: using the least squares method to calculate the financial data The time series is fitted to obtain a linear regression model, and the slope of the linear regression model is used as the financial data The measured trend value .

[0025] Preferably, before collecting the time series within the target time window of each financial data, the monitoring method also includes: setting an initial time window for each financial data, calculating the trend deviation of each financial data in the historical audit data with normal monitoring results, and taking the minimum trend deviation as the objective function value; updating the initial time window of each financial data, taking the initial time window when the objective function value takes the minimum value as the target time window of each financial data, and taking the minimum value of the objective function value as the preset threshold.

[0026] In the historical audit data with normal monitoring results, the problem of determining the value of the preset threshold and the problem of setting the target time window of each financial data are transformed into an optimization problem. The optimization goal is to minimize the minimum value of the trend deviation of each financial data. The target time window of each financial data is determined quickly and accurately, and the minimum value of the trend deviation of each financial data is used as the preset threshold to accurately obtain the monitoring results of the financial data audit process.

[0027] In the second aspect of the present application, there is also provided an audit monitoring system based on big data, including a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, an audit monitoring method based on big data as described in the first aspect of the present application is implemented.

[0028] The technical solution of this application has the following beneficial technical effects:

[0029] Granger causality test is used to determine the causal relationship between financial data. Financial data The causal relationship can reflect the financial data Prior changes can effectively predict financial data Changes in financial data Changes from previous periods to financial data the influence of the changing trend of the financial data; the measured trend value of each financial data is obtained according to the time series within the target time window of each financial data, and the predicted trend value of each financial data is calculated according to the causal relationship between the financial data; if the predicted trend value of a financial data is slightly different from the measured trend value, it means that the measured trend value satisfies the causal relationship between the financial data, and the measured trend value of the financial data is in a normal state; on the contrary, if the predicted trend value of a financial data is significantly different from the measured trend value, it means that the measured trend value does not satisfy the causal relationship between the financial data, and the measured trend value of the financial data is in an abnormal state, and abnormal fluctuations have occurred; therefore, when the trend deviation of any financial data is greater than the preset threshold, the monitoring result is abnormal, otherwise, the monitoring result is normal, and accurate monitoring results are obtained by combining the mutual influence between various financial data. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] By reading the detailed description below with reference to the accompanying drawings, the above and other purposes, features and advantages of the exemplary embodiments of the present application will become easy to understand. In the accompanying drawings, several embodiments of the present application are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0031] Figure 1 is a flow chart of an audit monitoring method based on big data according to an embodiment of the present application;

[0032] Figure 2 It is a structural block diagram of an audit monitoring system based on big data according to an embodiment of the present application. DETAILED DESCRIPTION

[0033] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present application.

[0034] It should be understood that when the terms "first", "second", etc. are used in the claims, specification and drawings of the present application, they are only used to distinguish different objects, rather than to describe a specific order. The terms "include" and "comprise" used in the specification and claims of the present application indicate the presence of the described features, wholes, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or their collections.

[0035] According to the first aspect of the present application, the present application provides an audit monitoring method based on big data. Figure 1is a flow chart of an audit monitoring method based on big data according to an embodiment of the present application. Figure 1 As shown, the big data-based audit monitoring method 100 includes steps S101 to S104, which are described in detail below.

[0036] S101, collecting the time series within the target time window of each financial data, and using Granger causality test to determine the causal relationship between the financial data.

[0037] In one embodiment, the financial data is numerical data, and each financial data includes fixed assets, investment assets, accounts receivable, income and management expenses, etc. The time series within the target time window of each financial data is collected. Each financial data corresponds to a target time window. The target time window includes the current collection period and multiple collection periods before the current collection period. The time series of the financial data includes the values ​​of the financial data of each collection period within the target time window. For example, the financial data The length of the corresponding target time window is , the current collection cycle and the total number of Financial data for each collection period The value of the financial data Time series within the target time window; the collection period can be 1 day or 1 week, which is not limited in this application.

[0038] Specifically, the target time window corresponding to each financial data is set to 10, that is, the length of each financial data time series is the same; the autocorrelation function can also be used to obtain the change period of each financial data, and the change period is set as the target time window of the corresponding financial data, that is, the length of each financial data time series is different.

[0039] In one embodiment, after obtaining the time series of each financial data, the Granger causality test is used to determine the causal relationship between the financial data. It should be noted that the Granger causality test is a method for analyzing the causal relationship of time series data, which can obtain the causal relationship between financial data. Financial data Causal relationship , Able to reflect financial data Time series for financial data The degree of influence of future values.

[0040] Specifically, financial data Financial data Causal relationship Methods for obtaining include: establishing financial data The first autoregressive model, the input of the first autoregressive model is financial data The values ​​of each collection period before the current collection period in the time series are output as the financial data of the current collection period The value of; establish the second autoregressive model, the input of the second autoregressive model is financial data and financial data The values ​​of each collection period before the current collection period in the time series are output as the financial data of the current collection period The absolute value of the difference between the error terms in the first autoregressive model and the second autoregressive model is used as the financial data Financial data Causal relationship .

[0041] Among them, the autoregressive model is a statistical model widely used to analyze time series data, which can describe the relationship between a variable and its past values. The first autoregressive model is:

[0042] , It is financial data The autoregression coefficient of For financial data The values ​​of each collection period before the current collection period in the time series, is the error term of the first autoregressive model, Collect financial data for the current period The value of For financial data The length of the target time window.

[0043] The second autoregressive model satisfies the relationship:

[0044] ,in, It is financial data The autoregression coefficient of It is financial data The autoregression coefficient of For financial data The values ​​of each collection period before the current collection period in the time series, For financial data The values ​​of each collection period before the current collection period in the time series, is the error term of the second autoregressive model, Collect financial data for the current period The value of and Financial data and financial data The length of the target time window.

[0045] It is understandable that the error term of the first autoregressive model is Able to reflect financial data The deviation between the real-time value of and its past value in the target time window (i.e., the value of other acquisition cycles outside the current acquisition cycle in the target time window); the error term of the second autoregressive model Able to reflect financial data Real-time values ​​and financial data and financial data The deviation of past values ​​within the target time window, then Ability to characterize financial data After the past values ​​in the target time window are used as the input of the autoregressive model, the predicted financial data The impact of real-time values, The larger the value, the more financial data Financial data The greater the impact, the greater the financial data Financial data The closer the causal relationship is, the more financial data can be realized. Financial data Causal relationship Accurate quantification of .

[0046] It should be noted that financial data Financial data Causal relationship , and financial data Financial data Causal relationship different.

[0047] In another embodiment, the financial data Financial data Causal relationship is 0 or 1, and the existing Granger causality test is directly used to determine whether there is a causal relationship between the financial data. Financial data If there is a causal relationship, then Set to 1, otherwise, Set to 0. Similarly, financial data can be obtained Financial data Causal relationship .

[0048] In this way, the causal relationship between the financial data is obtained. The causal relationship mentioned here is Refers to financial data Prior changes can effectively predict financial data changes, that is, financial data Changes in financial data can cause This causal relationship is called "Granger cause".

[0049] S102, obtaining the measured trend value of each financial data time series.

[0050] In one embodiment, each financial data time series is analyzed to obtain the measured trend value of each financial data.

[0051] Specifically, financial data The measured trend value for:

[0052] ; and is a set of value pairs, which are financial data Number in the time series and number The value of Represents obtaining financial data The time series meets All value pairs of is the mean value function.

[0053] Among them, when the measured trend value When greater than 0, it indicates financial data The time series within the target time window shows an upward trend. The larger the value, the more obvious the upward trend. When equal to 0, it indicates financial data The time series within the target time window shows a stable trend; when the measured trend value When it is less than 0, it indicates financial data The time series within the target time window shows a downward trend. The smaller the value, the more obvious the downward trend.

[0054] In other embodiments, linear regression analysis may be used to obtain the measured trend value of each financial data time series. The measured trend value The methods for obtaining include: using the least squares method to calculate the financial data The time series is fitted to obtain a linear regression model, and the slope of the linear regression model is used as the financial data The measured trend value .

[0055] Among them, the linear regression model satisfies: ,in For financial data The value of is the collection cycle, is the slope, To offset the bias, the least square method is used to analyze the financial data. Fit the time series to determine the slope and bias The value of .

[0056] In this way, the measured trend value of each financial data time series is obtained.

[0057] S103, calculating the predicted trend value of each financial data.

[0058] In one embodiment, the causal relationship between the financial data has been determined in step S101. If the value of any financial data changes, other financial data that have a causal relationship with the financial data will also change accordingly.

[0059] For example, financial data Financial data There is a causal relationship, that is, financial data Changes in financial data can cause Changes in financial data If the value of That is, financial data The measured trend value can be obtained from the financial data Prediction of the measured trend value.

[0060] Specifically, financial data The predicted trend value of for: , is the number of financial data, Representing financial data Financial data The causal relationship, For financial data The measured trend value.

[0061] S104, in response to the trend deviation of any financial data being greater than a preset threshold, the monitoring result is abnormal, otherwise, the monitoring result is normal, and the trend deviation is the absolute value of the difference between the predicted trend value and the measured trend value.

[0062] In one embodiment, step S103 obtains the predicted trend value of each financial data, and step S102 can obtain the measured trend value of each financial data; in response to the absolute value of the difference between the predicted trend value and the measured trend value of any financial data being greater than a preset threshold, the monitoring result is abnormal, otherwise, the monitoring result is normal. Among them, the absolute value of the difference between the predicted trend value and the measured trend value of a financial data can reflect the deviation between the measured trend value and the predicted trend value of the financial data, which is recorded as the trend deviation. Ideally, the trend deviation should be less than a preset threshold, which can be set manually. In this embodiment, the value of the preset threshold is 0.1.

[0063] In another embodiment, the value of the preset threshold affects the monitoring results, and artificially setting the preset threshold cannot ensure the accuracy of the monitoring results; and the length of each financial data target time window directly affects the measured trend value of each financial data, and then affects the absolute value of the difference between the predicted trend value and the measured trend value. Therefore, the value of the preset threshold and the length of each financial data target time window jointly affect the accuracy of the monitoring results.

[0064] Specifically, before collecting the time series within the target time window of each financial data, the monitoring method also includes: setting an initial time window for each financial data, calculating the trend deviation of each financial data in the historical audit data whose monitoring results are normal, and taking the minimum trend deviation as the objective function value; updating the initial time window of each financial data, taking the initial time window when the objective function value takes the minimum value as the target time window of each financial data, and taking the minimum value of the objective function value as the preset threshold.

[0065] Among them, historical audit data includes the historical time series of various financial data, and historical audit data with normal monitoring results can be obtained through manual labeling methods.

[0066] Assuming there are 5 types of financial data, we get There are 5 initial time windows in total. According to the initial time windows, the historical time series of the corresponding financial data are intercepted to obtain the historical time series of each financial data in the corresponding initial time window. According to the method of step S101 to step S104, the trend deviation of each financial data under the 5 initial time windows can be obtained. Further, the simulated annealing algorithm or the hill climbing algorithm is used to update the historical time series of the corresponding financial data. Update to obtain the initial time window when the objective function value takes the minimum value, that is, the target time window of each financial data.

[0067] In this way, in the historical audit data with normal monitoring results, the problem of determining the value of the preset threshold and the problem of setting the target time window of each financial data are transformed into an optimization problem, and the minimum value of the trend deviation of each financial data is taken as the optimization goal. The target time window of each financial data is determined quickly and accurately, and the minimum value of the trend deviation of each financial data is used as the preset threshold to accurately obtain the monitoring results of the financial data audit process.

[0068] The above describes the technical principle and implementation details of the audit monitoring method based on big data of this application through specific embodiments. Granger causality test is used to determine the causal relationship between financial data. Financial data The causal relationship can reflect the financial data Prior changes can effectively predict financial data Changes in financial data Changes from previous periods to financial data the influence of the changing trend of the financial data; the measured trend value of each financial data is obtained according to the time series within the target time window of each financial data, and the predicted trend value of each financial data is calculated according to the causal relationship between the financial data; if the predicted trend value of a financial data is slightly different from the measured trend value, it means that the measured trend value satisfies the causal relationship between the financial data, and the measured trend value of the financial data is in a normal state; on the contrary, if the predicted trend value of a financial data is significantly different from the measured trend value, it means that the measured trend value does not satisfy the causal relationship between the financial data, and the measured trend value of the financial data is in an abnormal state, and abnormal fluctuations have occurred; therefore, when the trend deviation of any financial data is greater than the preset threshold, the monitoring result is abnormal, otherwise, the monitoring result is normal, and accurate monitoring results are obtained by combining the mutual influence between various financial data.

[0069] According to the second aspect of the present application, the present application also provides an audit monitoring system based on big data. Figure 2 is a structural block diagram of an audit monitoring system based on big data according to an embodiment of the present application. Figure 2 As shown, the system 50 includes a processor and a memory, the memory stores computer program instructions, and when the computer program instructions are executed by the processor, an audit monitoring method based on big data according to the first aspect of the present application is implemented. The system also includes other components well known to those skilled in the art, such as a communication bus and a communication interface, whose settings and functions are known in the art, so they are not described here.

[0070] In the present application, the aforementioned memory may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, system or device. For example, a computer-readable storage medium may be any suitable magnetic storage medium or magneto-optical storage medium, such as a resistive random access memory RRAM (Resistive Random Access Memory), a dynamic random access memory DRAM (Dynamic Random Access Memory), a static random access memory SRAM (Static Random-Access Memory), an enhanced dynamic random access memory EDRAM (Enhanced Dynamic Random Access Memory), a high-bandwidth memory HBM (High-Bandwidth Memory), a hybrid memory cube HMC (Hybrid Memory Cube), etc., or any other medium that can be used to store the required information and can be accessed by an application, a module, or both. Any such computer storage medium may be part of a device or accessible or connectable to a device. Any application or module described in this application may be implemented using computer-readable / executable instructions that may be stored or otherwise maintained by such a computer-readable medium.

[0071] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0072] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be construed as limiting the scope of the patent application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent application shall be subject to the attached claims.

Claims

1. An audit monitoring method based on big data, characterized in that: The monitoring method comprises: Collect the time series within the target time window of each financial data, and use Granger causality test to determine the causal relationship between the financial data; Obtain the measured trend value of each financial data time series; Calculate the forecast trend value of each financial data, financial data The predicted trend value of for: , is the number of financial data, Representing financial data Financial data The causal relationship, For financial data The measured trend value of In response to the trend deviation of any financial data being greater than a preset threshold, the monitoring result is abnormal, otherwise, the monitoring result is normal, and the trend deviation is the absolute value of the difference between the predicted trend value and the measured trend value; Before collecting the time series within the target time window of each financial data, the monitoring method also includes: setting an initial time window for each financial data, calculating the trend deviation of each financial data in the historical audit data whose monitoring results are normal, and taking the minimum trend deviation as the objective function value; updating the initial time window of each financial data, taking the initial time window when the objective function value takes the minimum value as the target time window of each financial data, and taking the minimum value of the objective function value as the preset threshold.

2. According to the big data-based audit monitoring method of claim 1, it is characterized in that: The financial data is numerical data, and includes fixed assets, investment assets, accounts receivable, income and administrative expenses.

3. According to the big data-based audit monitoring method of claim 1, it is characterized in that: Each financial data corresponds to a target time window, which includes a current collection period and multiple collection periods before the current collection period. The time series of financial data includes the values ​​of financial data of each collection period within the target time window.

4. The audit monitoring method based on big data according to claim 1 is characterized in that: The method for obtaining the target time window includes: The autocorrelation function is used to obtain the change period of each financial data, and the change period is set as the target time window of the corresponding financial data.

5. The audit monitoring method based on big data according to claim 1 is characterized in that: Financial Data Financial data Causal relationship The methods for obtaining include: Building financial data The first autoregressive model, the input of the first autoregressive model is financial data The values ​​of each collection period before the current collection period in the time series are output as the financial data of the current collection period The value of Establish a second autoregressive model, the input of which is financial data and financial data The values ​​of each collection period before the current collection period in the time series are output as the financial data of the current collection period The value of The absolute value of the difference between the error terms in the first autoregressive model and the second autoregressive model is used as the financial data Financial data Causal relationship .

6. The audit monitoring method based on big data according to claim 5 is characterized in that: Financial Data The first autoregressive model is: ， The second autoregressive model satisfies the relationship: ,in, It is financial data The autoregression coefficient of For financial data The values ​​of each collection period before the current collection period in the time series, is the error term of the first autoregressive model, It is financial data The autoregression coefficient of For financial data The values ​​of each collection period before the current collection period in the time series, is the error term of the second autoregressive model, Collect financial data for the current period The value of and Financial data and financial data The length of the target time window.

7. The audit monitoring method based on big data according to claim 1 is characterized in that: Financial Data Financial data Causal relationship The methods for obtaining include: If we use the Granger causality test to judge financial data Financial data If there is a causal relationship, then Set to 1, otherwise, Set to 0.

8. The audit monitoring method based on big data according to claim 1 is characterized in that: Financial Data The measured trend value for: ; and is a set of value pairs, which are financial data Number in the time series and number The value of Represents obtaining financial data The time series meets All value pairs of is the mean value function.

9. The audit monitoring method based on big data according to claim 1 is characterized in that: Financial Data The measured trend value The methods for obtaining include: Using the least squares method to analyze financial data The time series is fitted to obtain a linear regression model, and the slope of the linear regression model is used as the financial data The measured trend value .

10. An audit monitoring system based on big data, characterized in that: It includes a processor and a memory, the memory stores computer program instructions, and when the computer program instructions are executed by the processor, an audit monitoring method based on big data according to any one of claims 1 to 9 is implemented.

Images