Data verification method and device, electronic equipment, storage medium and program product

Abstract

The invention provides a data verification method and device, electronic equipment, a storage medium and a program product. The method comprises the steps of determining a first correlation coefficient between a first field and a second field based on a field value of the first field and a field value of the second field in first business data; under the condition that the first correlation coefficient is larger than or equal to a first threshold value, the field value of the first field and the field value of the second field are combined, and a field value combination is obtained; determining second service data corresponding to the field value combination from the first service data; and performing exception verification on the field value combination based on the first quantity of the second business data to obtain an exception verification result of the first business data. According to the method and the device, the data exception between different field values in the business data can be accurately determined, and the accuracy of the business data is improved.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a data verification method, device, electronic device, storage medium and program product. Background Art

[0002] Business data analysis is crucial for improving operational efficiency, optimizing decision-making processes, and enhancing market competitiveness. In practice, ensuring data consistency and self-consistency across different fields in business data becomes a crucial issue.

[0003] In the related art, the consistency and accuracy of data are ensured only by setting verification rules such as mathematical operations, which results in low accuracy. Summary of the Invention

[0004] The embodiments of the present application provide a data verification method, device, electronic device, storage medium and program product, which can improve the accuracy of business data.

[0005] This embodiment of the present application provides a data verification method, the method comprising:

[0006] Determining a first correlation coefficient between the first field and the second field based on a field value of the first field and a field value of the second field in the first business data;

[0007] When the first correlation coefficient is greater than or equal to a first threshold, combining the field value of the first field and the field value of the second field to obtain a field value combination;

[0008] Determine second business data corresponding to the field value combination from the first business data;

[0009] An abnormality check is performed on the field value combination based on the first quantity of the second business data to obtain an abnormality check result of the first business data.

[0010] The present invention provides a data verification device, including:

[0011] a correlation determination module, configured to determine a first correlation coefficient between the first field and the second field based on a field value of the first field and a field value of the second field in the first business data;

[0012] a field value combining module, configured to combine the field value of the first field and the field value of the second field to obtain a field value combination when the first correlation coefficient is greater than or equal to a first threshold;

[0013] A second business data determination module, configured to determine second business data corresponding to the field value combination from the first business data;

[0014] A verification module is used to perform an abnormality check on the field value combination based on the first quantity of the second business data to obtain an abnormality check result of the first business data.

[0015] An embodiment of the present application provides an electronic device, comprising:

[0016] a memory for storing computer-executable instructions or computer programs;

[0017] The processor is used to implement the data verification method provided in the embodiment of the present application when executing the computer executable instructions or computer program stored in the memory.

[0018] An embodiment of the present application provides a computer-readable storage medium storing a computer program or computer-executable instructions for implementing the data verification method provided in the embodiment of the present application when executed by a processor.

[0019] An embodiment of the present application provides a computer program product, including a computer program or computer-executable instructions. When the computer program or computer-executable instructions are executed by a processor, the data verification method provided in the embodiment of the present application is implemented.

[0020] The embodiments of the present application have the following beneficial effects:

[0021] A first correlation coefficient is calculated between the first field and the second field. Since the first correlation coefficient is used to characterize the degree of association between the first field and the second field, when the first correlation coefficient is greater than or equal to a first threshold, field value combinations are performed, thereby accurately identifying associated fields in the first business data. For a field value combination consisting of two associated fields, the second business data corresponding to the field value combination in the first business data is determined, and an anomaly check is performed on the field value combination based on a first quantity of the second business data to obtain an anomaly check result for the first business data. This effectively identifies business data that may have anomalies, thereby improving the overall accuracy of the business data. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 This is a schematic diagram of the architecture of the data verification system provided in an embodiment of the present application;

[0023] Figure 2 This is a schematic diagram of the structure of an electronic device provided by an embodiment of the present application;

[0024] Figure 3 This is a flow diagram of the data verification method provided in the embodiment of the present application. Figure 1 ;

[0025] Figure 4This is a flow diagram of the data verification method provided in the embodiment of the present application. Figure 2 ;

[0026] Figure 5 This is a flow diagram of the data verification method provided in the embodiment of the present application. Figure 3 ;

[0027] Figure 6 This is a flow diagram of the data verification method provided in the embodiment of the present application. Figure 4 ;

[0028] Figure 7 This is a flow diagram of the data verification method provided in the embodiment of the present application. Figure 5 ;

[0029] Figure 8 This is another optional flow chart of the data verification method provided in the embodiment of the present application;

[0030] Figure 9 This is a flowchart of the data verification method provided in the embodiment of the present application;

[0031] Figure 10 This is a flowchart of the correlation detection process provided by the embodiment of the present application;

[0032] Figure 11 This is a process framework diagram for abnormal data screening provided by an embodiment of the present application;

[0033] Figure 12 This is a flowchart of the abnormality check process provided in the embodiment of the present application. DETAILED DESCRIPTION

[0034] In order to make the purpose, technical solutions and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings. The described embodiments should not be regarded as limiting this application. All other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0035] In the following description, reference is made to “some embodiments”, which describes a subset of all possible embodiments, but it will be understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

[0036] In the following description, the terms "first\second\third" involved are merely used to distinguish similar objects and do not represent a specific ordering of the objects. It can be understood that "first\second\third" can be interchanged with a specific order or sequence where permitted, so that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described herein.

[0037] In the embodiments of the present application, the term "module" or "unit" refers to a computer program or a part of a computer program that has a predetermined function and works together with other related parts to achieve a predetermined goal, and can be implemented in whole or in part by using software, hardware (such as processing circuits or memories) or a combination thereof. Similarly, a processor (or multiple processors or memories) can be used to implement one or more modules or units. In addition, each module or unit can be part of an overall module or unit that includes the function of the module or unit.

[0038] Unless otherwise defined, all technical and scientific terms used in the embodiments of the present application have the same meanings as those commonly understood by those skilled in the art. The terms used in the embodiments of the present application are only for the purpose of describing the embodiments of the present application and are not intended to limit the present application.

[0039] The relevant data collection and processing in the embodiments of this application should be strictly in accordance with the requirements of relevant laws and regulations when applied in examples, and the informed consent or separate consent of the personal information subject should be obtained. Subsequent data use and processing should be carried out within the scope of authorization of laws and regulations and the personal information subject.

[0040] Before further describing the embodiments of the present application in detail, the nouns and terms involved in the embodiments of the present application are explained. The nouns and terms involved in the embodiments of the present application are subject to the following interpretations.

[0041] 1) Correlation test: Correlation test is a statistical test to determine whether variables are correlated and the degree of correlation.

[0042] 2) Continuous variable: Also known as a continuity indicator or continuity field, it is a variable that describes the numerical characteristics of things. Its value is numerical data and can be used for numerical calculations (addition, subtraction, multiplication, division, etc.).

[0043] 3) Discrete variables: Also known as discrete indicators or discrete fields, these variables represent data attributes, characteristics, or states, and are typically character or text types. They may also be coded numerically, such as a gender field where 1 represents male and 0 represents female. Numerical calculations cannot be performed on discrete variables.

[0044] 4) Five-Tier Resource Borrowing Classification: The five-tier resource borrowing classification system primarily assesses the borrower's repayment capacity, specifically their ability to ultimately repay the principal and interest. Commercial banks categorize resource borrowing into five risk levels: normal, special mention, substandard, doubtful, and loss.

[0045] 5) Correlation coefficient: A statistically used indicator coefficient used to measure the linear correlation between two variables. Commonly used ones include the Pearson correlation coefficient, the Spearman correlation coefficient, and the Kendall correlation coefficient.

[0046] 6) Self-information: used to measure the amount of information contained in a single event.

[0047] 7) Joint self-information: used to measure the amount of information when two random variables appear together.

[0048] 8) Conditional self-information: used to measure the uncertainty of one variable when the value of another variable is known.

[0049] 9) Joint entropy: used to measure the uncertainty of two or more random variables as a whole.

[0050] 10) Conditional entropy: A measure of the uncertainty of a random variable given the information of another random variable.

[0051] In the related art, business personnel usually configure basic checks for discrete fields based on their own understanding of the business. For example, for discrete field A, the configured basic checks usually only determine whether the field value is empty, whether the value range exceeds the definition domain, etc., and it is difficult to link it to the relationship with the related discrete field B. As mentioned above, the configuration of discrete fields mostly focuses on whether the field value is empty, whether the value range exceeds the definition domain, etc., lacks a more comprehensive and strict check for business logic, and the scenario coverage of the configuration rules is low. Due to the particularity of discrete fields, it is impossible to directly use clear rule check values ​​like continuous fields. Therefore, the judgment of the relationship between discrete fields depends on the experience of business personnel. In this case, some discrete fields that should be checked but are not checked are often omitted, and the correlation determination rate of related fields is low. In summary, the data verification accuracy for discrete fields in the related art is low, resulting in low accuracy of business data.

[0052] Based on the problems existing in the related technologies, the embodiments of the present application provide a data verification method, device, electronic device, computer-readable storage medium and computer program product, which can accurately determine data anomalies between different field values ​​in business data and improve the accuracy of business data.

[0053] The following describes an exemplary application of the data verification device provided in the embodiment of the present application. The data verification device provided in the embodiment of the present application is an electronic device for implementing a data verification method. The electronic device provided in the embodiment of the present application can be implemented as various types of terminals such as laptops, tablet computers, desktop computers, set-top boxes, smart phones, smart speakers, smart watches, smart TVs, and vehicle-mounted terminals, and can also be implemented as servers. Among them, the server can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers. The terminal and the server can be directly or indirectly connected via wired or wireless communication, which is not limited in the embodiment of the present application. Below, an exemplary application of the data verification device when it is implemented as a server or a terminal will be described.

[0054] See also Figure 1 , Figure 1 : This is a schematic diagram of the architecture of the data verification system provided in an embodiment of the present application. To support a data verification application, the data verification system 100 provided in an embodiment of the present application includes at least a terminal 400, a network 300, and a server 200, wherein the server 200 is the server of the data verification application. The server 200 can constitute the data verification device of the embodiment of the present application, that is, the data verification method of the embodiment of the present application is implemented through the server 200. The terminal is connected to the server 200 via the network 300, which can be a wide area network or a local area network, or a combination of the two.

[0055] See also Figure 1 , the user can perform interactive operations on the client side of the data verification application through the terminal 400, and the interactive operations may be, for example, inputting business data operations, clicking to start verification operations, etc. After receiving the interactive operations of the user, the client side may encapsulate the first business data into a data verification request through the terminal, and send the data verification request to the server 200 through the network 300. After receiving the data verification request, the server 200 responds to the data verification request and determines the first correlation coefficient between the first field and the second field based on the field value of the first field and the field value of the second field in the first business data; when the first correlation coefficient is greater than or equal to the first threshold, the server 200 combines the field value of the first field and the field value of the second field to obtain a field value combination; the server 200 determines the second business data corresponding to the field value combination from the first business data; the server 200 performs an anomaly check on the field value combination based on the first quantity of the second business data to obtain an anomaly check result of the first business data. After generating the anomaly check result, the server 200 may also send the anomaly check result to the terminal 400 to display the anomaly check result to the user.

[0056] In some embodiments, the data verification method of the embodiment of the present application can also be performed by the terminal 400 itself, that is, after the terminal 400 receives the interactive operation input by the user through the client, the terminal 400 determines the first correlation coefficient between the first field and the second field based on the field value of the first field and the field value of the second field in the first business data; when the first correlation coefficient is greater than or equal to the first threshold, the terminal 400 combines the field value of the first field and the field value of the second field to obtain a field value combination; the terminal 400 determines the second business data corresponding to the field value combination from the first business data; the terminal 400 performs an anomaly check on the field value combination based on the first quantity of the second business data to obtain an anomaly check result for the first business data. After the anomaly check result is generated, the anomaly check result is displayed on the client interface of the terminal 400.

[0057] When financial companies report financial data to regulatory agencies, inter-field data verification is required to ensure consistency and self-consistency between financial data reported to different regulatory agencies. A user can input financial data into a data verification application client via terminal 400. The client encapsulates the financial data into a data verification request and sends the data verification request to server 200 via network 300. After receiving the data verification request, the server 200 responds to the data verification request and determines a first correlation coefficient between the first field and the second field based on the field value of the first field "Five-level classification of resource borrowing" and the field value of the second field "Resource borrowing status" in the financial data; when the first correlation coefficient is greater than or equal to the first threshold, the first field "Five-level classification of resource borrowing" and the second field "Resource borrowing status" are associated fields and data verification between the fields is required. The server 200 combines the field value of the first field and the field value of the second field to obtain a field value combination; for example, the field value of the combination of the field value "FQ2 (overdue days (1, 30])" of the first field and the field value "LS01 (normal)" of the second field is combination, the server 200 determines the second business data corresponding to the field value combination from the first business data; the server 200 performs an anomaly check on the field value combination based on the first quantity of the second business data, and obtains an anomaly check result for the first business data. If the anomaly check result of the first business data is that the field value combination is a potential anomaly combination, then the "FQ2, LS01" field value combination is more likely to be abnormal data and can be transferred to manual verification. If the manual verification result is that the "FQ2, LS01" field value combination is abnormal, all data with the "FQ2, LS01" field value combination in the financial data will be filtered out as abnormal data, and the abnormal data will be displayed on the client interface of the terminal to facilitate users to delete or modify the abnormal data, thereby ensuring the correctness of the financial data reported to the regulatory agency.

[0058] In an e-commerce analysis scenario, order data anomaly detection is required. After receiving a data verification request, server 200 begins analyzing two fields in the order data: payment method (e.g., credit card, payment application, etc.) and shipping region (e.g., City A, City B, etc.). Server 200 calculates a first correlation coefficient between the first field "payment method" and the second field "shipping region" based on their field values. If the first correlation coefficient is greater than or equal to a first threshold, the first field "payment method" and the second field "shipping region" are related fields, requiring data verification between the fields. Server 200 combines the field values ​​of the first field with the field values ​​of the second field to obtain a field value combination, such as a combination of "credit card - city A." Server 200 then filters all second business data corresponding to these field value combinations from the order data. Server 200 then performs an anomaly check on these field value combinations based on the first quantity of the second business data, obtaining an anomaly verification result for the first business data. For example, if the first quantity of the second business data for the field value combination of "credit card - city A" is much higher than expected, or increases suddenly, this may indicate an anomaly. After obtaining the abnormal field value combination, server 200 filters out all data with the abnormal field value combination as abnormal data, and displays the abnormal data on the client interface of the terminal, so that users can quickly identify potential problems in the order data and take corresponding measures to deal with them.

[0059] See also Figure 2 , Figure 2 is a structural diagram of an electronic device provided in an embodiment of the present application, Figure 2 The electronic device shown includes: at least one processor 410, a memory 450, at least one network interface 420 and a user interface 430. The various components in the terminal 400 are coupled together via a bus system 440. It is understood that the bus system 440 is used to achieve connection and communication between these components. In addition to including a data bus, the bus system 440 also includes a power bus, a control bus and a status signal bus. However, for the sake of clarity, the bus system 440 is not shown in FIG. Figure 2 Various buses are labeled as bus system 440 .

[0060] The processor 410 can be an integrated circuit chip with signal processing capabilities, such as a general-purpose processor, a digital signal processor (DSP), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc., where the general-purpose processor can be a microprocessor or any conventional processor, etc.

[0061] The user interface 430 includes one or more output devices 431 that enable presentation of media content, including one or more speakers and / or one or more visual display screens. The user interface 430 also includes one or more input devices 432, including user interface components that facilitate user input, such as a keyboard, mouse, microphone, touch screen display, camera, other input buttons and controls.

[0062] The memory 450 may be removable, non-removable, or a combination thereof. Exemplary hardware devices include solid-state memory, hard drives, optical drives, etc. The memory 450 may optionally include one or more storage devices that are physically remote from the processor 410.

[0063] The memory 450 includes volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. The non-volatile memory may be a read-only memory (ROM), and the volatile memory may be a random access memory (RAM). The memory 450 described in the embodiments of the present application is intended to include any suitable type of memory.

[0064] In some embodiments, the memory 450 can store data to support various operations, examples of which include programs, modules, and data structures, or a subset or superset thereof, as exemplified below.

[0065] Operating system 451, including system programs for processing various basic system services and performing hardware-related tasks, such as the framework layer, core library layer, and driver layer, which are used to implement various basic services and process hardware-based tasks;

[0066] A network communication module 452 is used to reach other electronic devices via one or more (wired or wireless) network interfaces 420. Exemplary network interfaces 420 include Bluetooth, Wi-Fi, and Universal Serial Bus (USB);

[0067] a presentation module 453 for enabling presentation of information via one or more output devices 431 (e.g., a display screen, a speaker, etc.) associated with the user interface 430 (e.g., a user interface for operating peripheral devices and displaying content and information);

[0068] The input processing module 454 is configured to detect one or more user inputs or interactions from one of the one or more input devices 432 and to translate the detected inputs or interactions.

[0069] In some embodiments, the apparatus provided in the embodiments of the present application may be implemented in software. Figure 2 The data verification device 455 stored in the memory 450 is shown. This device can be software in the form of a program or plug-in, and includes the following software modules: a correlation determination module 4551, a field value combination module 4552, a second business data determination module 4553, and a verification module 4554. These modules are logical and can be arbitrarily combined or further separated according to the functions implemented. The functions of each module will be described below.

[0070] In other embodiments, the data verification device provided in the embodiments of the present application can be implemented in hardware. As an example, the device provided in the embodiments of the present application can be a processor in the form of a hardware decoding processor, which is programmed to execute the data verification method provided in the embodiments of the present application. For example, the processor in the form of a hardware decoding processor can adopt one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), programmable logic devices (PLDs), complex programmable logic devices (CPLDs), field-programmable gate arrays (FPGAs) or other electronic components.

[0071] It should be noted that, based on their understanding of the following, those skilled in the art can apply the data verification method provided in the embodiments of this application to any scenario requiring data consistency or self-consistency verification, such as financial scenarios, medical and health scenarios, education and training scenarios, smart city scenarios, etc. The following description will be made using the financial scenario as an example.

[0072] The following describes the data verification method provided by the embodiment of the present application. As mentioned above, the electronic device that implements the data verification method of the embodiment of the present application can be a terminal, a server, or a combination of the two. Therefore, the execution entity of each step will not be repeated below.

[0073] Figure 3 This is a flow diagram of the data verification method provided in the embodiment of the present application. Figure 1 , the following will be combined Figure 3 The steps shown are explained as Figure 3 As shown, the data verification method is described as an example in which the execution subject is a server. The method includes the following steps 101 to 104:

[0074] In step 101 , a first correlation coefficient between a first field and a second field is determined based on a field value of a first field and a field value of the second field in first business data.

[0075] Here, the first business data is data related to the business scenario. For example, in the field of e-commerce, the first business data may include data such as user purchase records, product information, and user reviews; in the field of finance, the first business data may include data such as the number of resource borrowings, the five-level classification of resource borrowings, and the resource borrowing status. The first business data may contain various types of data fields, such as continuous fields (such as price) and discrete fields (such as gender). The number of first business data may be one or more. The number of business data is the unit of measurement for business data, such as the number of records, the number of data items, or the size of a data set. In the embodiments of the present application, the number of all subsequent business data can be counted using the number of records. For example, the second number of first business data is the number of first business data in the data table. For example, in a financial report, there are 500 rows of first business data, then the second number of first business data is 500, and each first business data has fields such as the number of resource borrowings, the five-level classification of resource borrowings, and the resource borrowing status, as well as the values ​​of the fields. The first field and the second field may be discrete fields or continuous fields, and the embodiments of the present application do not limit this. Taking the example where both the first field and the second field are discrete fields, a discrete field refers to an attribute or feature in the first business data whose value is a discrete value. Discrete values ​​generally refer to a finite number of discontinuous, clearly distinguishable values. For example, the resource borrowing status field can be a typical discrete field, and its values ​​may be normal, extended, overdue, and shortened. The field value is the specific value of the discrete field. The first field and the second field are any two different discrete fields in the first business data. The first field can be a five-level classification of resource borrowing, and the second field can be the resource borrowing status. The first correlation coefficient is used to measure the closeness of the relationship between the two fields, that is, to evaluate the correlation between the first field and the second field.

[0076] In some embodiments, see Figure 4 , Figure 4 It is shown that in step 101, based on the field value of the first field and the field value of the second field in the first business data, determining the first correlation coefficient between the first field and the second field can be achieved by the following steps 1011 to 1012:

[0077] In step 1011 , a second correlation coefficient between the first field and the second field is determined based on the field value of the first field and the field value of the second field in a plurality of first business data.

[0078] Among them, one first business data corresponds to a first field and a second field.

[0079] Here, the second correlation coefficient is used to measure the closeness of the relationship between the two fields, that is, to evaluate the correlation between the first field and the second field. The number of second correlation coefficients can be 1 or more. When the number of second correlation coefficients is multiple, the calculation method of each second correlation coefficient is different, and the embodiment of the present application does not limit the specific correlation coefficient calculation method. For example, multiple second correlation coefficients may include the Spearman correlation coefficient and the Kendall correlation coefficient. For each first business data, the first business data includes field values ​​of multiple fields, each field corresponds to a field value, and any two fields are obtained from the multiple fields as the first field and the second field. The field value of the first field and the field value of the second field in the first business data are obtained. Based on the field value of the first field and the field value of the second field in the multiple first business data, the Spearman correlation coefficient and the Kendall correlation coefficient are determined respectively.

[0080] In some embodiments, determining the second correlation coefficient between the first field and the second field based on the field value of the first field and the field value of the second field in multiple first business data can be achieved in the following way: first, based on the field value of the first field, determine a first sequence containing multiple first business data, based on the field value of the second field, determine a second sequence containing multiple first business data, and determine the first position information of each first business data in the first sequence, and the second position information of each first business data in the second sequence; then, based on the difference between the first position information and the second position information of each first business data, and the second number of first business data, determine the second correlation coefficient between the first field and the second field.

[0081] Here, the second correlation coefficient may be the Spearman correlation coefficient. Multiple pieces of first business data may be sorted based on the field value of the first field in each piece of first business data to obtain a first sequence. Multiple pieces of first business data may be sorted based on the field value of the second field in each piece of first business data to obtain a second sequence. However, field values ​​cannot be directly compared for sorting, so the field values ​​are first encoded to obtain code values. The field value of the first field is numerically encoded to obtain the code value of the first field. The field value of the second field is numerically encoded to obtain the code value of the second field. For example, the first field represents a five-level classification of resource borrowing, and the field values ​​may include FQ1 (normal and not overdue), FQ2 (overdue between 1 and 30 days), FQ3 (overdue between 31 and 60 days), FQ4 (overdue between 16 and 90 days), and FQ5 (overdue over 90 days). The resulting code values ​​after encoding are 0, 1, 2, 3, and 4. The second field represents the resource borrowing status, and its possible values ​​include LS01 (normal), LS02 (extended), LS03 (overdue), and LS04 (reduced). The encoded code values ​​are 0, 1, 2, and 3. The field values ​​of the first fields of the plurality of first business data in the first sequence are arranged in ascending order according to the corresponding code values, and the field values ​​of the second fields of the plurality of first business data in the second sequence are arranged in ascending order according to the corresponding code values.

[0082] The first position information is the relative position or index of each piece of first business data in the first sequence. Specifically, for each piece of first business data, the first position information of the first business data in the first sequence is the sequence number of the first business data in the first sequence. For example, in a first sequence consisting of 10 pieces of first business data, if the code value of the first field "Five-level Resource Borrowing Classification" is used as the sorting basis, if the first business data is ranked 3rd, then the first position information of the first business data is 3. The same applies to the second position information and is not further described.

[0083] After obtaining the first location information and the second location information of each first business data, for each first business data, determine the difference between the first location information and the second location information of the first business data. Add the squares of the differences between the first location information and the second location information of the plurality of first business data to obtain a sum of the squared differences. Determine a Spearman correlation coefficient based on the sum of the squared differences and the second number of first business data. The Spearman correlation coefficient may satisfy the following formula (1).

[0084]

[0085] Where ρ is the Spearman correlation coefficient, n is the second number of the first business data, d iis the difference between the first location information and the second location information of the i-th first service data, is the sum of squared differences, i is an integer between 1 and n, and n is a positive integer.

[0086] In some embodiments, determining the second correlation coefficient between the first field and the second field based on the field value of the first field and the field value of the second field in multiple first business data can be achieved in the following way: first, determining the first difference in the field value of the first field of two first business data in the multiple first business data, and the second difference in the field value of the second field of the two first business data; then, when the first difference and the second difference have the same integer sign, determining the two first business data as a first business data pair; when the first difference and the second difference have different integer signs, determining the two first business data as a second business data pair; finally, determining the second correlation coefficient between the first field and the second field based on the difference between the number of first business data pairs and the number of second business data pairs, and the second number of first business data.

[0087] Here, the second correlation coefficient may be a Kendall correlation coefficient. For any two first business data among a plurality of first business data, a first difference in the field values ​​of the first fields of the two first business data and a second difference in the field values ​​of the second fields may be determined. When the product of the first difference and the second difference is greater than or equal to 0, the first difference and the second difference have the same integer sign, and the two first business data are determined to be a first business data pair. When the product of the first difference and the second difference is less than 0, the first difference and the second difference have different integer signs, and the two first business data are determined to be a second business data pair. The integer sign is the positive and negative attribute of an integer, that is, whether an integer is positive, negative or zero. It should be noted that since the field values ​​of discrete fields cannot be directly subjected to mathematical operations, they can be encoded to obtain code values ​​and then operated on. The differences in the field values ​​in the embodiments of the present application are all differences in the encoded code values.

[0088] For example, in a first business data pair, the first field X of one first business data item A has a field value of FQ1 and a code value of 0, while the first field X of the other first business data item B has a field value of FQ2 and a code value of 1, i.e., the first difference is 1-0=1; the second field Y of the first business data item A has a field value of LS01 and a code value of 0, while the second field Y of the other first business data item B has a field value of LS03 and a code value of 2, i.e., the second difference is 2-0=2. The integer signs of the first difference and the second difference are both positive, and the first difference and the second difference have the same integer sign. In a second business data pair, the first field X of one first business data item C has a field value of FQ5 and a code value of 4, while the first field X of the other first business data item D has a field value of FQ2 and a code value of 1, i.e., the first difference is 1-4=-3; the second field Y of the first business data item C has a field value of LS02 and a code value of 1, while the second field Y of the other first business data item D has a field value of LS03 and a code value of 2, i.e., the second difference is 2-1=1. The integer sign of the first difference is negative, the integer sign of the second difference is positive, and the first difference and the second difference have different integer signs.

[0089] The difference in the number of first business data pairs and second business data pairs in the plurality of first business data is determined, and a Kendall correlation coefficient is determined based on the difference in the number and the second number of the first business data. The Kendall correlation coefficient may satisfy the following formula (2).

[0090]

[0091] Wherein, τ is the Kendall correlation coefficient, c is the number of first business data pairs, d is the number of second business data pairs, n is the second number of first business data, and cd is the difference between the number of first business data pairs and the number of second business data pairs.

[0092] In step 1012 , a first correlation coefficient between the first field and the second field is determined from the second correlation coefficients.

[0093] Here, if the number of second correlation coefficients is greater than 1, after determining multiple second correlation coefficients, the maximum absolute value of the multiple second correlation coefficients can be determined as the first correlation coefficient between the first field and the second field. Alternatively, a weighted average can be performed on the multiple second correlation coefficients, and the absolute value of the obtained value can be determined as the first correlation coefficient between the first field and the second field. If the number of second correlation coefficients is 1, the second correlation coefficient can be directly determined as the first correlation coefficient between the first field and the second field.

[0094] The embodiment of the present application determines the second correlation coefficient and determines the first correlation coefficient between the first field and the second field from the second correlation coefficient, thereby improving the accuracy of the correlation coefficient and further accurately identifying fields with an associated relationship.

[0095] In step 102, when the first correlation coefficient is greater than or equal to the first threshold, the field value of the first field and the field value of the second field are combined to obtain a field value combination.

[0096] Here, the first threshold is a predefined numerical standard used to determine whether the correlation between two fields is strong enough. The embodiment of the present application does not specifically limit the value of the first threshold, and it can be set based on actual needs. For example, the first threshold can be 0.7. When the first correlation coefficient is greater than or equal to 0.7, the first field and the second field are associated fields. Select any field value from all possible values ​​of the first field, and select any field value from all possible values ​​of the second field, and combine them to obtain a field value combination. For example, all possible values ​​of the first field X are a and b, and all possible values ​​of the second field Y are e, then there are two field value combinations: a and e, and b and e.

[0097] In step 103, second business data corresponding to the field value combination is determined from the first business data.

[0098] Here, the field value of the first field and the field value of the second field of the second business data corresponding to the field value combination are the same as the field value of the field value combination. For each field value combination, the field value combination can be used as a filtering condition to filter out at least one second business data that meets the filtering condition from multiple first business data. The filtering condition can be written using a Structured Query Language (SQL) statement or other data processing logic.

[0099] In step 104, an abnormality check is performed on the field value combination based on the first quantity of the second business data to obtain an abnormality check result of the first business data.

[0100] Here, the anomaly check result of the first business data may include a field value combination in the first business data being a potentially abnormal combination, or each field value combination in the first business data being a normal combination. A potentially abnormal combination refers to a field value combination that has a high probability of being abnormal, and generally refers to a field value combination that occurs infrequently.

[0101] In some embodiments, see Figure 5 , Figure 5It is shown that in step 104, an abnormality check is performed on the field value combination based on the first quantity of the second business data to obtain an abnormality check result of the first business data, which can be achieved by the following steps 1041A to 1042A:

[0102] In step 1041A, a joint probability of the field value combination is determined based on the first number and the second number.

[0103] Here, the second quantity is the total number of first business data. The joint probability of a field value combination is the probability that the field value of the first field and the field value of the second field in the field value combination occur simultaneously. The joint probability of a field value combination is the ratio of the first quantity of second business data corresponding to the field value combination to the second quantity of first business data.

[0104] In step 1042A, when the joint probability is less than the second threshold, the first anomaly check result is determined as the anomaly check result of the first business data, wherein the first anomaly check result is used to characterize that the field value combination is a potential anomaly combination.

[0105] Here, the second threshold is a key parameter used to determine whether a field value combination is a potential abnormal combination. The present embodiment does not limit the value of the second threshold and can be set based on actual circumstances. For example, the second threshold can be 10%. When the joint probability is less than 10%, the field value combination is determined to be a potential abnormal combination, that is, a first abnormality check result is obtained that indicates that the field value combination may be a potential abnormal combination.

[0106] By setting a reasonable second threshold, the embodiment of the present application can effectively screen out field value combinations that may actually pose risks, avoid a large number of false positives and missed negatives, and improve the accuracy and efficiency of data anomaly detection.

[0107] In some embodiments, see Figure 6 , Figure 6 It is shown that in step 104, an abnormality check is performed on the field value combination based on the first quantity of the second business data to obtain an abnormality check result of the first business data, which can be achieved by the following steps 1041B to 1042B:

[0108] In step 1041B, based on the first quantity, the joint self-information, the first conditional self-information, the second conditional self-information, the joint entropy, the first conditional entropy, and the second conditional entropy of the field value combination are determined.

[0109] Among them, the field value known in the first conditional self-information is different from the field value known in the second conditional self-information, and the field value known in the first conditional entropy is different from the field value known in the second conditional entropy.

[0110] Here, assume that the first field in a field value combination is X, with a field value of a, and the second field is Y, with a field value of b. The joint self-information I(X, Y) of this field value combination is the amount of information when field values ​​a and b appear together. The first conditional self-information I(Y|X) is the self-information of the field value b of the second field Y, given that the field value of the first field is a. The second self-information I(X|Y) is the self-information of the field value a of the first field X, given that the field value of the second field is b. The joint entropy H(X, Y) represents the uncertainty shared by the first field X and the second field Y. The first conditional entropy H(Y|X) is the entropy of the second field Y, given that the field value of the first field X is a. The second conditional entropy H(X|Y) is the entropy of the first field X, given that the field value of the second field Y is b.

[0111] In an embodiment of the present application, determining the joint self-information, first conditional self-information, second conditional self-information, joint entropy, first conditional entropy and second conditional entropy of the field value combination based on the first quantity can be achieved in the following manner: first, determining the joint probability of the field value combination based on the first quantity and the second quantity of the first business data; then, determining the first conditional probability of the field value combination based on the first quantity and the third quantity of the third business data, wherein the field value of the first field in the third business data is the same as the field value of the first field in the field value combination; and determining the second conditional probability of the field value combination based on the first quantity and the fourth quantity of the fourth business data, wherein the field value of the second field in the fourth business data is the same as the field value of the second field in the field value combination; finally, determining the joint self-information, first conditional self-information, second conditional self-information, joint entropy, first conditional entropy and second conditional entropy of the field value combination based on the joint probability, the first conditional probability and the second conditional probability.

[0112] Here, for each field value combination, the joint probability of the field value combination is the ratio of the first number of second business data items having the field value combination to the second number of first business data items. Using the field value of the first field X in the field value combination as a screening condition, at least one third business data item having the field value of the first field in the field value combination is screened from multiple first business data items. The ratio of the first number to the third number of the third business data items is determined as the first conditional probability P(Y|X) of the field value combination. Using the field value of the second field in the field value combination as a screening condition, at least one fourth business data item having the field value of the second discrete field Y in the field value combination is screened from multiple first business data items. The ratio of the first number to the fourth number of the fourth business data items is determined as the second conditional probability P(X|Y) of the field value combination. Based on the joint probability P(X, Y), the first conditional probability P(Y|X), and the second conditional probability P(X|Y), the joint self-information, first conditional self-information, second conditional self-information, joint entropy, first conditional entropy, and second conditional entropy of the field value combination are determined.

[0113] In an embodiment of the present application, there are multiple field value combinations; based on the joint probability, the first conditional probability and the second conditional probability, determining the joint self-information, the first conditional self-information, the second conditional self-information, the joint entropy, the first conditional entropy and the second conditional entropy can be achieved in the following manner: first, based on the joint probability of each field value combination, the joint self-information is determined, and the sum of the products of the multiple joint self-information and the joint probability is determined as the joint entropy; then, based on the first conditional probability of each field value combination, the first conditional self-information is determined, and based on the second conditional probability of each field value combination, the second conditional self-information is determined; then, based on the third number and the second number, the first probability of the field value combination is determined, and the sum of the products of the first probability, the first conditional probability and the first conditional self-information of the multiple field value combinations is determined as the first conditional entropy; finally, based on the fourth number and the second number, the second probability of the field value combination is determined, and the sum of the products of the second probability, the second conditional probability and the second conditional self-information of the multiple field value combinations is determined as the second conditional entropy.

[0114] Here, the negative value of the logarithm of the joint probability can be determined as the joint self-information. The joint self-information I(X, Y) can satisfy the following formula (3).

[0115] I(X,Y)=-logP(X,Y) Formula (3)

[0116] Multiply the joint self-information of each field value combination by the joint probability to obtain the product corresponding to each field value combination. Sum the products corresponding to multiple field value combinations to obtain the joint entropy H(X, Y). The joint entropy H(X, Y) can satisfy the following formula (4).

[0117] H(X,Y) =-∑P(X,Y)logP(X,Y) Formula (4)

[0118] For each field value combination, the negative value of the logarithm of the first conditional probability P(Y|X) of the field value combination is determined as the first conditional self-information I(Y|X). The negative value of the logarithm of the second conditional probability P(X|Y) of the field value combination is determined as the second conditional self-information I(X|Y). The first conditional self-information I(Y|X) satisfies the following formula (5), and the second conditional self-information I(X|Y) satisfies the following formula (6).

[0119] I(Y|X)=-logP(Y|X) Formula (5)

[0120] I(X|Y)=-logP(X|Y) Formula (6)

[0121] The ratio of the third quantity to the second quantity is determined as the first probability P(X) of the field value combination, and the sum of the first probabilities, first conditional probabilities, and first conditional self-information of multiple field value combinations is determined as the first conditional entropy H(Y|X). The ratio of the fourth quantity to the second quantity is determined as the second probability P(Y) of the field value combination, and the sum of the second probabilities, second conditional probabilities, and second conditional self-information of multiple field value combinations is determined as the second conditional entropy H(X|Y). The first conditional entropy H(Y|X) satisfies the following formula (7), and the second conditional entropy H(X|Y) satisfies the following formula (8).

[0122] H(Y|X)=-∑P(X)P(X|Y)logP(X|Y) Formula (7)

[0123] H(X|Y)=-∑P(Y)P(Y|X)logP(Y|X) Formula (8)

[0124] In step 1042B, when the joint self-information is greater than the joint entropy, the first conditional self-information is greater than the first conditional entropy, and the second conditional self-information is greater than the second conditional entropy, the second anomaly check result is determined as the anomaly check result of the first business data, wherein the second anomaly check result is used to characterize that the field value combination is a potential anomaly combination.

[0125] Here, for each field value combination, when the joint self-information of the field value combination is greater than the joint entropy, the first conditional self-information is greater than the first conditional entropy, and the second conditional self-information is greater than the second conditional entropy, the field value combination is determined to be a potential abnormal combination, that is, the second abnormality check result is determined to characterize the field value combination as a potential abnormal combination. When the joint self-information of the field value combination is less than or equal to the joint entropy, or the first conditional self-information is less than or equal to the first conditional entropy, or the second conditional self-information is less than or equal to the second conditional entropy, the field value combination is a normal combination. If each field value combination is a normal combination, an abnormality check result is obtained to characterize that the first business data is normal.

[0126] In the embodiment of the present application, by comparing the size of self-information and entropy, field value combinations whose self-information is greater than entropy can be screened out, which means that the amount of information of this field value combination is greater than the average value, that is, the probability of occurrence of this field value combination is low, and it is a potential abnormal combination, thereby improving the accuracy and efficiency of data verification.

[0127] The embodiment of the present application calculates a first correlation coefficient between a first field and a second field, and performs a field value combination when the first correlation coefficient is greater than or equal to a first threshold value, thereby accurately identifying associated fields in the first business data. For a field value combination consisting of two associated fields, the second business data corresponding to the field value combination in the first business data is determined, and an anomaly check is performed on the field value combination based on a first quantity of the second business data to obtain an anomaly check result for the first business data. This effectively determines business data that may have anomalies, thereby improving the overall accuracy of the business data.

[0128] In some embodiments, as Figure 7 As shown, when the abnormality check result is used to characterize that the field value combination is a potential abnormal combination, the data verification method provided in the embodiment of the present application further includes the following steps 201 to 202:

[0129] In step 201, the abnormal configuration table is queried based on the field value combination.

[0130] The exception configuration table includes abnormal field value combinations.

[0131] Here, the exception configuration table is a data structure or database table for defining and managing abnormal situations. The exception configuration table may include a predefined abnormal field value combination. The exception configuration table may include the field names of the two fields in the abnormal field value combination, the field values ​​of the two fields, and the cause of the abnormality. After determining the field value combination as a potential abnormal combination, it is possible to query whether the field value combination exists in the exception configuration table, that is, whether there is an abnormal field value combination, in which the field names of the two fields and the field values ​​of the two fields are the same as the field names and field values ​​of the first field and the second field in the field value combination.

[0132] In step 202, when a field value combination is found in the abnormal configuration table, fifth business data corresponding to the field value combination in the first business data is determined as abnormal data.

[0133] Here, when the field value combination is found in the abnormal configuration table, the fifth business data corresponding to the field value combination in the plurality of first business data is determined as abnormal data. If the field value combination is not found in the abnormal configuration table, manual verification can be performed to determine whether the field value combination is abnormal. If the manual verification result is abnormal, the field value combination determined as abnormal is stored in the abnormal configuration table.

[0134] The embodiment of the present application queries the field value combinations from the abnormal configuration table and determines the fifth business data in the first business data containing these combinations as abnormal data accordingly, so as to more accurately locate the abnormal data and improve the efficiency and accuracy of data verification.

[0135] In some embodiments, after step 201, the data verification method provided by the embodiments of the present application further includes the following steps: first, when no field value combination is queried from the abnormal configuration table, the first normal configuration table is queried based on the field value combination, wherein the first normal configuration table includes a normal field value combination; then, when a field value combination is queried from the first normal configuration table, the fifth business data corresponding to the field value combination in the first business data is determined as normal data.

[0136] Here, the first normal configuration table is a database table that records all known, considered normal field value combinations. These normal field value combinations can be based on business rules, common patterns in historical data, or any other verified correct field value combinations. Each row in the first normal configuration table represents a specific field value combination. For example, a normal field value combination in the first normal configuration table is a five-level classification: FQ1, and a resource borrowing status: LS02. If a potentially abnormal field value combination has a first field with a five-level split and a field value of FQ1, and a second field with a resource borrowing status and a field value of LS02, this field value combination can be found in the first normal configuration table, and the fifth business data corresponding to this field value combination in the first business data can be determined as normal data. If this field value combination is not found in the first normal configuration table, manual verification can be performed to determine whether the field value combination is abnormal. If the manual verification result is abnormal, the field value combination determined to be abnormal is stored in the abnormal configuration table. If the manual verification result is normal, the field value combination determined to be normal is stored in the first normal configuration table.

[0137] In the embodiment of the present application, by comparing with known normal field value combinations, data that does not meet expectations can be effectively identified, thereby improving the overall data quality. In the case that the abnormal configuration table fails to identify, further verification using the first normal configuration table can more comprehensively cover various situations and reduce false positives and false negatives. Storing common normal field value combinations in a configuration table makes the subsequent data processing logic more concise and clear, reducing complexity. And as the business develops and changes, the first normal configuration table can be updated to adapt to new data patterns, ensuring the flexibility and scalability of data verification.

[0138] In some embodiments, after step 201, the data verification method provided by the embodiments of the present application also includes the following steps: first, when no field value combination is queried from the abnormal configuration table, the field value of the business scenario field is determined from the fifth business data; then, based on the field value combination and the field value of the business scenario field, the second normal configuration table is queried, wherein the second normal configuration table includes normal pairings between different field value combinations and field values ​​of different business scenario fields; finally, when a normal pairing of the field value combination and the field value of the business scenario field is queried from the second normal configuration table, the fifth business data is determined as normal data.

[0139] Here, a business scenario field refers to a field that reflects key attributes or statuses of business activities, such as resource borrowing extension or resource borrowing overdue. The second normal configuration table is a database table that records all known, considered normal, pairings between different field value combinations and field values ​​of different business scenario fields. A normal pairing between different field value combinations and field values ​​of different business scenario fields means that the field value combination is reasonable or expected within the business scenario represented by the field value of a business scenario field. For example, a normal pairing in the second normal configuration table is: five-level classification: FQ1, resource borrowing status: LS02, business scenario field: resource borrowing extension. If the five-level resource borrowing classification is FQ1 (normal), then the resource borrowing status is LS02 (extended), which is reasonable. SQL can be used to create a filter statement combining the field value combination and the field value of the business scenario field, and then query the second normal configuration table using this filter statement. If a normal pairing of the field value combination and the field value of the business scenario field is found, the fifth business data is determined to be normal data.

[0140] The embodiment of the present application judges abnormal situations of field value combinations in combination with business scenarios, which can be more in line with actual business situations and improve the accuracy of business data verification.

[0141] Figure 8 This is another optional flow chart of the data verification method provided in the embodiment of the present application, such as Figure 8 As shown, the method includes the following steps 301 to 309:

[0142] Step 301: The terminal receives an interactive operation from a user.

[0143] Here, the user's interactive operation may be inputting business data, clicking to start verification, and the like.

[0144] Step 302: The terminal generates a data verification request in response to the interactive operation.

[0145] Here, after receiving the user's interactive operation, the terminal may encapsulate the first service data into a data verification request.

[0146] Step 303: The terminal sends a data verification request to the server.

[0147] Step 304 : In response to the data verification request, the server determines a first correlation coefficient between the first field and the second field based on the field value of the first field and the field value of the second field in the first business data.

[0148] Here, the specific process of determining the first correlation coefficient between the first field and the second field based on the field value of the first field and the field value of the second field in the first business data can refer to step 101 in the above embodiment and will not be repeated here.

[0149] Step 305: When the first correlation coefficient is greater than or equal to the first threshold, the server combines the field value of the first field and the field value of the second field to obtain a field value combination.

[0150] Here, the specific process of combining the field value of the first field and the field value of the second field to obtain the foreground word of the field value combination can refer to step 102 in the above embodiment, and will not be repeated here.

[0151] Step 306: The server determines second business data corresponding to the field value combination from the first business data.

[0152] Here, the specific process of determining the second business data corresponding to the field value combination from the first business data can refer to step 103 in the above embodiment, and will not be repeated here.

[0153] Step 307: The server performs an anomaly check on the field value combination based on the first quantity of the second business data to obtain an anomaly check result of the first business data.

[0154] Here, the specific process of performing anomaly check on the field value combination based on the first quantity of the second business data and obtaining the anomaly check result of the first business data can refer to step 104 in the above embodiment and will not be repeated here.

[0155] Step 308: The server sends the abnormality check result to the terminal.

[0156] Step 309: The terminal displays the abnormality check result on the current interface.

[0157] In an embodiment of the present application, a data verification request is sent to a server through a terminal, so that the server calculates a first correlation coefficient between a first field and a second field, and performs a field value combination when the first correlation coefficient is greater than or equal to a first threshold value, thereby accurately identifying the associated fields in the first business data, and then performing an abnormality check on the field value combination consisting of the two associated fields. For the obtained field value combination, the server determines the second business data corresponding to the field value combination in the first business data, performs an abnormality check on the field value combination based on the first quantity of the second business data, and obtains an abnormality check result for the first business data, which can effectively discover business data that may have abnormalities, thereby improving the overall accuracy of the business data.

[0158] The following describes an exemplary application of the embodiments of the present application in a practical application scenario.

[0159] In the financial sector, different regulatory agencies have varying supervisory and reporting requirements for financial companies. For example, the People's Bank of China requires financial companies to regularly submit information such as personal credit reports, interest rate filings, and basic data from the People's Bank of China. The Financial Regulatory Commission (FSRC) requires financial companies to submit data such as the Standardized Supervisory Data (EAST) and off-site supervisory reports (1104 reports). These different reporting requirements have both similarities and differences. Ensuring data consistency and self-consistency across different reports and across different fields within the same report is a major challenge in the regulatory reporting process. Compared to other application data, regulatory reporting data has a lower tolerance for errors and anomalies. Reporting erroneous data to regulatory agencies can result in varying degrees of notification and penalties, depending on the scale and impact of the error. Therefore, financial companies should do their utmost to ensure the accuracy and self-consistency of reported data. Reported financial data includes both continuous and discrete indicators. For example, continuous field indicators include information such as credit limit, principal balance, and repayment amount, while discrete indicators include resource borrowing status, five-level classification, and resource borrowing quantity. Because continuous fields carry clear business meaning and cross-references (the relationship between related figures in ledgers and accounting reports that allows for cross-verification and verification), they are generally easier for reporting personnel to prioritize and design corresponding validation rules to ensure data quality. For example, the rule: (Loan Amount = Repayment Amount + Principal Balance) can simultaneously validate three continuous fields. Discrete fields, on the other hand, represent information such as status, type, and attributes, so rules cannot be simply designed using numerical calculations (addition, subtraction, multiplication, division, etc.) or Boolean algebra (AND, OR, NOT, greater than, less than, equal to, etc.). Therefore, it is often difficult for reporting personnel to proactively discover and design appropriate validation rules. Therefore, a new approach is required to discover relevant fields and configure validation rules for discrete fields.

[0160] In related technologies, basic validation is usually configured based on the reporting personnel's own understanding of the business. For example, for discrete field A, the configured validation is usually only to determine whether the field value is empty, whether the value range exceeds the domain, etc., and it is difficult to link it to the related discrete field B. For example, the correlation between the five-level classification field and the overdue start date field. The five-level classification field is used to measure the user's overdue status and degree of overdue. If the five-level classification field is a normal class, then based on business logic, the overdue start date field should be a null value or an empty string, that is, it is in a non-overdue state. If the overdue start date exists, it means that the five-level classification field and the overdue date field are inconsistent, which requires attention and processing.

[0161] The data verification method provided in the embodiment of the present application is a discrete field consistency verification configuration method for financial big data. Figure 9 This is a flow chart of the data verification method provided in the embodiment of the present application. Figure 9 Step 401: Discrete field screening. To fully cover the range of fields to be analyzed, discrete fields here include not only fields whose value types are strings or text types, but also fields whose values ​​are integers but enumerated types (for example, a gender field where 0 represents female and 1 represents male). Step 402: Correlation detection. Detect the correlation of discrete variables. Figure 10 This is a flowchart of the correlation detection process provided by the embodiment of the present application. Figure 10 , step 501, numerical encoding. Since most discrete fields are text types, non-numeric data needs to be encoded as numerical values. Correlation detection needs to be calculated for numerical values. As shown in Table 1, the five-level classification of resource borrowing is a discrete field, and the field value (lvl_grade) is a character type, including five values ​​of FQ1, FQ2, FQ3, FQ4 and FQ5, which are encoded as integer types 0, 1, 2, 3, and 4. As shown in Table 2, the resource borrowing status is a discrete field, and the field value (loan_status) is a character type, which is encoded as an integer type.

[0162] Table 1 Five-level classification coding results of discrete field resource borrowing

[0163]

[0164] Table 2 Discrete field resource borrowing status coding results

[0165] Field Value Resource borrowing status Numerical encoding LS01 normal 0 LS02 Extension 1 LS03 Overdue 2 LS04 contraction 3

[0166] Step 502: Calculate the correlation coefficient. A statistical method is used to calculate the correlation between two fields (corresponding to the first and second fields in the above embodiment). The Spearman correlation coefficient and the Kendall correlation coefficient are used to calculate the correlation between each pair of discrete fields. The result of the correlation coefficient test is between [-1, 1]. This coefficient is used to measure the linear correlation between two fields. When the test result coefficient approaches 1 (can be equal to 1), it indicates that the two fields have a strong positive linear correlation; if the test result approaches -1 (can be equal to -1), it indicates that the two fields have a strong negative linear correlation; if the test result approaches 0, the two fields have no obvious linear correlation. Generally, the correlation coefficient can be divided into three categories based on its level: strong correlation, medium correlation, and weak correlation. The absolute value of a strong correlation coefficient is generally between 0.9 and 1, the absolute value of a medium correlation coefficient is generally between 0.7 and 0.9, and the absolute value of a weak correlation coefficient is generally between 0 and 0.7. The correlation coefficient is credible only if its absolute value is greater than or equal to 0.7. Step 503: Filter relevant discrete fields. After calculating the Spearman correlation coefficient and Kendall correlation coefficient of a group of discrete fields (corresponding to the multiple second correlation coefficients in the above embodiment), take the maximum value of the absolute values ​​of the two coefficients (corresponding to the first correlation coefficient in the above embodiment), and use this maximum value as a benchmark to screen discrete fields with an absolute value of the correlation coefficient greater than or equal to 0.7 (corresponding to the first threshold in the above embodiment), and determine that the two discrete fields have a strong correlation. It should be noted that the correlation test acts on two fields. If there are multiple discrete fields to be tested, they need to be tested separately between each two fields, and finally the fields with an absolute value of the correlation coefficient greater than or equal to 0.7 are screened.

[0167] Step 403: Data distribution statistics and suspected abnormal data screening. The discrete fields with correlation have been screened out in step 402. For the values ​​of the discrete fields, the statistical distribution is analyzed. Figure 11 This is a process framework diagram for abnormal data screening provided by the embodiment of this application. Figure 11 , step 601, data distribution statistics. Distribution statistics are performed on the values ​​of the two discrete fields to obtain the statistical magnitude (corresponding to the first quantity in the above embodiment) of each code value pair (corresponding to the field value combination in the above embodiment). Here, the five-level classification is defined as variable X, and the resource borrowing status is defined as Y. The joint probability P(X, Y), conditional probability P(X|Y) (corresponding to the second conditional probability in the above embodiment) and P(Y|X) (corresponding to the first conditional probability in the above embodiment) are respectively counted, corresponding to the three columns M\N\O in Table 3. The joint probability is the probability that two discrete fields X and Y take specific field values ​​at the same time. Table 3 is an example of data distribution statistics results.

[0168] Table 3 Data distribution statistics

[0169]

[0170]

[0171] Step 602: Calculate the joint self-information and joint entropy. Joint self-information, for two discrete fields X and Y, represents the reduction in the amount of information brought about by knowing the field values ​​of both X and Y. Joint self-information can be calculated using joint probability. Joint self-information I(X, Y) corresponds to column P in Table 3. For two discrete fields X and Y, joint entropy H(X, Y) represents the uncertainty shared by these two discrete fields. The probability-weighted average of the self-information of each pair of discrete fields is defined as the joint entropy. Joint entropy H(X, Y) corresponds to column S in Table 3.

[0172] Step 603: Calculate the conditional self-information I(X|Y) and conditional entropy H(X|Y). Conditional self-information I(X|Y) is the self-information of a discrete field X given a known discrete field Y. Conditional self-information I(X|Y) can be calculated based on the conditional probability P(X|Y). Conditional self-information I(X|Y) corresponds to column Q in Table 3. Conditional entropy H(X|Y) is the entropy of another discrete field X given a known discrete field Y. Conditional entropy H(X|Y) corresponds to column T in Table 3.

[0173] Step 604: Calculate the conditional self-information I(Y|X) and conditional entropy H(Y|X). Conditional self-information I(Y|X) is the self-information of discrete field Y given a known discrete field X. Conditional self-information I(Y|X) can be calculated based on the conditional probability P(Y|X). Conditional self-information I(Y|X) corresponds to column R in Table 3. Conditional entropy H(Y|X) is the entropy of discrete field Y given another discrete field X. Conditional entropy H(Y|X) corresponds to column U in Table 3.

[0174] Step 605: filter out the paired combinations whose self-information is greater than the entropy. Filter out the data whose joint self-information I(X, Y) is greater than the joint entropy H(X, Y), i.e. Figure 5 The data whose value in column P is greater than the value in column S. Filter out the data whose conditional self-information I(X|Y) is greater than the conditional entropy H(X|Y), that is, Figure 5 The data whose value in column Q is greater than the value in column T. Filter out the data whose conditional self-information I(Y|X) is greater than the conditional entropy H(Y|X), that is, Figure 5 The value of column R is greater than the value of column U.

[0175] Step 606, find the intersection. Find the code value pair whose joint self-information is greater than the joint entropy and whose conditional self-information is greater than the corresponding conditional entropy, as the potential abnormal code value (corresponding to the potential abnormal combination in the above embodiment). Entropy is used to describe the degree of chaos of a system in thermodynamics. In informatics, entropy is used to describe the degree of uncertainty of a variable information. The larger the entropy value, the greater the information uncertainty. For example, if a field x always takes a fixed value of 1, then the uncertainty of the field is 0. If the value range of field x is between 0 and 10000, then the uncertainty of the field is greater. In the embodiment of the present application, the joint self-information I(X, Y) represents the amount of information carried when a set of variables (X=a, Y=b) appear at the same time. The joint entropy H(X, Y) is the expectation of the joint self-information I(X, Y). The entropy H can be simply understood as the weighted average of the self-information I, and the weight is the probability of the self-information I appearing. If the variable self-information I(X=a, Y=b) of a particular combination is greater than the entropy H(X, Y), this means that the information content of this combination is greater than the average. This means that the probability of this combination occurring is lower, making it more likely to be an abnormal code value pair. Similarly, the comparison between conditional self-information and conditional entropy works the same way.

[0176] Step 404: Configure verification rules specifically. Figure 12 This is a flowchart of the abnormality check process provided by the embodiment of this application. Figure 12 , step 701, determine whether the potential abnormal code value combination is abnormal. Manual intervention verification is performed on the newly added potential abnormal code value combination detected in step 403, and the verification results are of two types. If the data is normal, go to step 702; or, if the data is abnormal, go to step 703. Step 702, include it in the normal state configuration table. If the filtered data belongs to a normal business scenario, then the situation is configured in the normal state configuration table normal_tbl. As shown in Table 4, the normal state configuration table is mainly used to record the business scenarios in which the filtered low-frequency code value combinations appear, as well as the screening logic of the business scenarios. When the next test is performed, this part of the data can be marked as normal by associating it with this normal state configuration table, without the need for manual repeated verification. Logic refers to the Structured Query Language (SQL) statement that filters out this part of the data. Specifically, it is to filter out data that meets the three conditions of check code value a, check code value b, and business scenarios. For example, it is determined that data that satisfies the five-level classification of resource borrowing, field value lvl_grade=FQ1, field value loan_status=LS02, and business scenario is resource borrowing extension, is normal data.

[0177] Table 4 Normal state configuration table

[0178]

[0179] Step 703: Add to the abnormal status configuration table. Similarly, data verified as abnormal must also be configured in the corresponding abnormal status configuration table (error_tbl), as shown in Table 5. Unlike the normal status configuration table, the abnormal status configuration table records the abnormal code value combination of two fields. In other words, the presence of this code value combination indicates that the data is abnormal. The abnormal status configuration table should be a regular data quality inspection item and checked daily. Once abnormal data is found, it should be promptly disclosed and handled.

[0180] Table 5 Abnormal status configuration table

[0181]

[0182] The following describes a specific embodiment of the above method using the reported financial data as an example. Table 6 shows the sample details of the financial data.

[0183] Table 6 Sample Detailed Data

[0184]

[0185] In this example, the sample detailed data includes six fields. The loan number is the primary key and is not included in the correlation analysis. The remaining five-level classification, resource borrowing status, resource borrowing period, and institution number are discrete fields, while the resource borrowing balance is a continuous field. Note that Table 6 only shows five sample details. In practice, more financial data can be selected to ensure meaningful statistical results (with a sufficient data volume).

[0186] First, filter out the discrete fields. The five-level classification, resource borrowing status, resource borrowing period, and institution number. The five-level classification and resource borrowing status have been described in the previous article and will not be repeated here. The resource borrowing period indicates the number of installments when the customer borrows resources. The value range is usually between 1 and 60 periods, depending on the customer's own operation choice. The institution number represents the code of the reporting institution and is a fixed field (the value is a constant fixed value). Then, the four discrete fields that have been filtered out are numerically encoded respectively to obtain Table 7. It should be noted that the code usually starts counting from 0.

[0187] Table 7 Discrete field encoding results

[0188]

[0189]

[0190] For the four coded discrete fields, the Spearman and Kendall correlation coefficients were calculated for each pair, resulting in the matrix shown in Table 8. This matrix is ​​symmetric and represents the linear correlation between each field and other fields (including itself). Diagonal values ​​of 1 indicate a self-correlation of 1. It should be noted that Table 8 is only an example and is not related to the data in Table 7.

[0191] Table 8 Correlation coefficient matrix

[0192]

[0193] The test results show that only one pair of discrete fields, [five-class classification - resource borrowing status], has a strong linear correlation with the absolute value of the Kendall correlation coefficient and the Spearman correlation coefficient greater than 0.7. The distribution of the code values ​​for the five-class classification X and the resource borrowing status Y was statistically analyzed, and the corresponding self-information and entropy were calculated. The statistical and calculation results can be seen in Table 3.

[0194] Calculate the joint probability distribution P(X, Y), which is the percentage of the code value associated with this combination in the total. For example, if the FQ1-LS01 statistic is 100, the joint probability P(X = FQ1, Y = LS01) = 100 / 750 = 13.33%. Other values ​​are calculated in the same way. Calculate the conditional probability distributions P(Y|X) and P(X|Y). For example, when X = FQ01, the probability of Y = LS01 is P(Y = LS01 | X = FQ1) = 100 / (100 + 100 + 100) = 33.33%. Other conditional probabilities are calculated in the same way. Calculate the self-information and conditional self-information. Self-information is calculated as the base-2 logarithm of the probability of the code value occurring. For example, I(X = FQ1, Y = LS01) = -log2(13.33%) = 2.9069. Other self-information is calculated in the same way. Calculate the joint entropy H(X, Y) and the conditional entropies H(X|Y) and H(Y|X). The three entropy values ​​correspond to the mathematical expectation of their respective self-information. The result in column S of Table 3 is the joint entropy H(X, Y), which is the mathematical expectation of column P I(X, Y), calculated by multiplying each row of column P by column M and summing them. Compare the three self-information columns P\Q\R with the three entropy columns S\T\U in Table 3 to filter out data with self-information greater than the entropy. The intersection of the screening results, FQ2-LS01 code value pairs, is considered potential anomaly data.

[0195] Next, we analyze low-frequency code value pairs and configure targeted verification rules. The analysis reveals that the code value combination FQ02-LS01 is abnormal data, and we add this abnormal data to the abnormal verification configuration table.

[0196] The embodiment of the present application mainly solves the data quality in the regulatory reporting process of financial institutions, and is aimed at the discovery and verification of data anomalies between different discrete field values. In conventional solutions, the configuration of problem discovery and verification relies heavily on the industry experience of developers or business personnel, and the risk of incorrect processing and missing necessary verification is very high. It is necessary to introduce new methods to assist in reducing the risk of incorrect processing. The embodiment of the present application introduces entropy and self-information in information theory, and comprehensively determines whether the data belongs to a potential abnormal object by comparing their values. For the detected potential abnormal data, normal and abnormal data are further distinguished by the normal configuration table and the abnormal configuration table. On the one hand, the normal configuration table can provide the screening logic for this type of data, and the data can be directly marked as normal in the next verification, without the need for manpower to repeat the verification. On the other hand, the abnormal configuration table is identified as problem data and needs to be strictly verified every day. The person in charge discovers the problem and intervenes in time.

[0197] It is understandable that in the embodiments of the present application, when user information and other related data are involved, when the embodiments of the present application are applied to specific products or technologies, user permission or consent must be obtained, and the collection, use and processing of relevant data must comply with relevant laws, regulations and standards.

[0198] The following continues to describe the exemplary structure of the data verification device 455 provided in the embodiment of the present application implemented as a software module. In some embodiments, such as Figure 2 As shown, the software modules stored in the data verification device 455 of the memory 450 may include: a correlation determination module 4551, used to determine a first correlation coefficient between the first field and the second field based on the field value of the first field and the field value of the second field in the first business data; a field value combination module 4552, used to combine the field value of the first field and the field value of the second field to obtain a field value combination when the first correlation coefficient is greater than or equal to a first threshold; a second business data determination module 4553, used to determine the second business data corresponding to the field value combination from the first business data; and a verification module 4554, used to perform anomaly verification on the field value combination based on the first quantity of the second business data to obtain an anomaly verification result of the first business data.

[0199] In some embodiments, the verification module 4554 is also used to determine the joint probability of the field value combination based on the first quantity and the second quantity, where the second quantity is the total number of the first business data; when the joint probability is less than the second threshold, determine the first anomaly verification result of the first business data, wherein the first anomaly verification result is used to characterize that the field value combination is a potential anomaly combination.

[0200] In some embodiments, the verification module 4554 is also used to determine the joint self-information, first conditional self-information, second conditional self-information, joint entropy, first conditional entropy and second conditional entropy of the field value combination based on the first quantity, wherein the field value known in the first conditional self-information is different from the field value known in the second conditional self-information, and the field value known in the first conditional entropy is different from the field value known in the second conditional entropy; when the joint self-information is greater than the joint entropy, the first conditional self-information is greater than the first conditional entropy, and the second conditional self-information is greater than the second conditional entropy, determine the second abnormality verification result of the first business data, wherein the second abnormality verification result is used to characterize that the field value combination is a potential abnormal combination.

[0201] In some embodiments, the verification module 4554 is also used to determine the joint probability of the field value combination based on the first quantity and the second quantity of the first business data; determine the first conditional probability of the field value combination based on the first quantity and the third quantity of the third business data, wherein the field value of the first field in the third business data is the same as the field value of the first field in the field value combination; determine the second conditional probability of the field value combination based on the first quantity and the fourth quantity of the fourth business data, wherein the field value of the second field in the fourth business data is the same as the field value of the second field in the field value combination; determine the joint self-information, first conditional self-information, second conditional self-information, joint entropy, first conditional entropy and second conditional entropy of the field value combination based on the joint probability, the first conditional probability and the second conditional probability.

[0202] In some embodiments, there are multiple field value combinations; the verification module 4554 is also used to determine the joint self-information based on the joint probability of each field value combination, and determine the joint entropy by adding the products of the multiple joint self-information and the joint probability; determine the first conditional self-information based on the first conditional probability of each field value combination, and determine the second conditional self-information based on the second conditional probability of each field value combination; determine the first probability of the field value combination based on the third number and the second number, and determine the first conditional entropy by adding the products of the first probability, the first conditional probability and the first conditional self-information of the multiple field value combinations; determine the second probability of the field value combination based on the fourth number and the second number, and determine the second conditional entropy by adding the products of the second probability, the second conditional probability and the second conditional self-information of the multiple field value combinations.

[0203] In some embodiments, the correlation determination module 4551 is also used to determine multiple second correlation coefficients based on the field value of the first field and the field value of the second field of each first business data in multiple first business data; and determine the first correlation coefficient between the first field and the second field from the multiple second correlation coefficients.

[0204] In some embodiments, the correlation determination module 4551 is also used to determine a first sequence containing multiple first business data based on the field value of the first field of each first business data, determine a second sequence containing multiple first business data based on the field value of the second field of each first business data, and determine the first position information of each first business data in the first sequence, and the second position information of each first business data in the second sequence; determine the second correlation coefficient between the first field and the second field based on the difference between the first position information and the second position information of each first business data, and the second quantity of the first business data.

[0205] In some embodiments, the correlation determination module 4551 is also used to determine a first difference in field values ​​of the first fields of two first business data in a plurality of first business data, and a second difference in field values ​​of the second fields of the two first business data; when the first difference and the second difference have the same integer sign, the two first business data are determined as a first business data pair; when the first difference and the second difference have different integer signs, the two first business data are determined as a second business data pair; based on the difference between the number of first business data pairs and the number of second business data pairs, and the second number of first business data, the second correlation coefficient between the first field and the second field is determined.

[0206] In some embodiments, the data verification device 455 also includes a query module for querying an abnormal configuration table based on the field value combination when the abnormal verification result is used to characterize that the field value combination is a potential abnormal combination, wherein the abnormal configuration table includes abnormal field value combinations; when the field value combination is queried from the abnormal configuration table, the fifth business data corresponding to the field value combination in the first business data is determined as abnormal data.

[0207] An embodiment of the present application provides a computer program product, which includes a computer program or computer-executable instructions stored in a computer-readable storage medium. A processor of an electronic device reads the computer-executable instructions from the computer-readable storage medium and executes the computer-executable instructions, causing the electronic device to perform the data verification method described above in the embodiment of the present application.

[0208] The embodiment of the present application provides a computer-readable storage medium in which computer-executable instructions or computer programs are stored. When the computer-executable instructions or computer programs are executed by a processor, the processor will execute the data verification method provided in the embodiment of the present application, for example, Figure 3 The data verification method shown.

[0209] In some embodiments, the computer-readable storage medium may be a memory such as RAM, ROM, flash memory, magnetic surface memory, optical disk, or CD-ROM; or may be various devices including one or any combination of the above memories.

[0210] In some embodiments, computer-executable instructions may be in the form of a program, software, software module, script, or code, written in any form of programming language (including compiled or interpreted languages, or declarative or procedural languages), and may be deployed in any form, including as a stand-alone program or as a module, component, subroutine, or other unit suitable for use in a computing environment.

[0211] As an example, computer-executable instructions may, but need not, correspond to a file in a file system, may be stored as part of a file that stores other programs or data, such as in one or more scripts in a HyperText Markup Language (HTML) document, in a single file dedicated to the program in question, or in multiple coordinating files (e.g., files storing one or more modules, subroutines, or code portions).

[0212] By way of example, computer-executable instructions may be deployed to be executed on one electronic device, or on multiple electronic devices located at one site, or on multiple electronic devices distributed across multiple sites and interconnected by a communication network.

[0213] In summary, the embodiments of the present application focus on solving the problems of correlation detection and abnormal code value pairing verification of discrete variables. Through the correlation detection module, the problem of identifying variables with linear correlation between two of many discrete variables is solved; through the data distribution analysis module, combined with the relationship between entropy and self-information, the problem of identifying abnormal code value pairs in the field is solved; through the verification configuration module, the normal state configuration table and the abnormal verification configuration table are configured in a targeted manner to ensure that such problems will not occur in future data.

[0214] The above description is merely an embodiment of the present application and is not intended to limit the scope of protection of the present application. Any modifications, equivalent replacements, and improvements made within the spirit and scope of the present application are included in the scope of protection of the present application.

Claims

1. A data verification method, characterized in that: The method comprises: Determine a first correlation coefficient between the first field and the second field based on a field value of a first field and a field value of a second field in the first business data; When the first correlation coefficient is greater than or equal to a first threshold, combining the field value of the first field and the field value of the second field to obtain a field value combination; Determine second business data corresponding to the field value combination from the first business data; An abnormality check is performed on the field value combination based on the first quantity of the second business data to obtain an abnormality check result of the first business data.

2. The method according to claim 1, characterized in that: The performing anomaly check on the field value combination based on the first quantity of the second business data to obtain an anomaly check result of the first business data includes: Determine a joint probability of the field value combination based on a first number and a second number, where the second number is a total number of the first business data; When the joint probability is less than the second threshold, a first anomaly check result is determined as an anomaly check result of the first business data, wherein the first anomaly check result is used to characterize that the field value combination is a potential anomaly combination.

3. The method according to claim 1, characterized in that The performing anomaly check on the field value combination based on the first quantity of the second business data to obtain an anomaly check result of the first business data includes: Based on the first number, determining joint self-information, first conditional self-information, second conditional self-information, joint entropy, first conditional entropy, and second conditional entropy of the field value combination, wherein the field value known in the first conditional self-information is different from the field value known in the second conditional self-information, and the field value known in the first conditional entropy is different from the field value known in the second conditional entropy; When the joint self-information is greater than the joint entropy, the first conditional self-information is greater than the first conditional entropy, and the second conditional self-information is greater than the second conditional entropy, the second anomaly check result is determined as the anomaly check result of the first business data, wherein the second anomaly check result is used to characterize that the field value combination is a potential anomaly combination.

4. The method according to claim 3, characterized in that The determining, based on the first quantity, joint self-information, first conditional self-information, second conditional self-information, joint entropy, first conditional entropy, and second conditional entropy of the field value combination includes: Determining a joint probability of the field value combination based on the first number and a second number of the first business data; Determine a first conditional probability of the field value combination based on the first quantity and a third quantity of the third business data, wherein the field value of the first field in the third business data is the same as the field value of the first field in the field value combination; Determining a second conditional probability of the field value combination based on the first quantity and a fourth quantity of fourth business data, wherein a field value of a second field in the fourth business data is the same as a field value of the second field in the field value combination; Based on the joint probability, the first conditional probability and the second conditional probability, the joint self-information, the first conditional self-information, the second conditional self-information, the joint entropy, the first conditional entropy and the second conditional entropy of the field value combination are determined.

5. The method according to claim 1, characterized in that The determining, based on the field value of the first field and the field value of the second field in the first business data, a first correlation coefficient between the first field and the second field includes: Determine a second correlation coefficient between the first field and the second field based on a field value of the first field and a field value of the second field in a plurality of first business data, wherein one first business data corresponds to one first field and one second field; A first correlation coefficient between the first field and the second field is determined from the second correlation coefficient.

6. The method according to claim 5, characterized in that The determining, based on the field value of the first field and the field value of the second field in the plurality of first business data, a second correlation coefficient between the first field and the second field comprises: Determine a first sequence including the plurality of first business data based on the field value of the first field, determine a second sequence including the plurality of first business data based on the field value of the second field, and determine first position information of each first business data in the first sequence, and second position information of each first business data in the second sequence; The second correlation coefficient between the first field and the second field is determined based on a difference between the first position information and the second position information of each of the first business data and a second amount of the first business data.

7. The method according to claim 5, characterized in that The determining, based on the field value of the first field and the field value of the second field in the plurality of first business data, a second correlation coefficient between the first field and the second field comprises: Determine a first difference value between field values ​​of first fields of two first business data among the plurality of first business data, and a second difference value between field values ​​of second fields of two first business data; In a case where the first difference value and the second difference value have the same integer sign, determining the two first service data as a first service data pair; In a case where the first difference value and the second difference value have different integer signs, determining the two first service data as a second service data pair; The second correlation coefficient between the first field and the second field is determined based on a difference between the number of the first business data pairs and the number of the second business data pairs, and a second number of the first business data.

8. The method according to any one of claims 1 to 7, characterized in that In the case where the abnormality check result is used to characterize that the field value combination is a potential abnormal combination, the method further includes: querying an abnormal configuration table based on the field value combination, wherein the abnormal configuration table includes abnormal field value combinations; In the case where the field value combination is found from the abnormal configuration table, fifth business data corresponding to the field value combination in the first business data is determined as abnormal data.

9. A data verification device, characterized in that: The device comprises: A correlation determination module, configured to determine a first correlation coefficient between the first field and the second field based on a field value of the first field and a field value of the second field in the first business data; a field value combination module, configured to combine the field value of the first field and the field value of the second field to obtain a field value combination when the first correlation coefficient is greater than or equal to a first threshold; A second business data determination module, used to determine second business data corresponding to the field value combination from the first business data; A verification module is used to perform an abnormality check on the field value combination based on the first quantity of the second business data to obtain an abnormality check result of the first business data.

10. An electronic device, characterized in that: The electronic device comprises: A memory for storing computer executable instructions or computer programs; A processor, used to implement the data verification method according to any one of claims 1 to 8 when executing the computer executable instructions or computer programs stored in the memory.

11. A computer-readable storage medium storing computer-executable instructions or a computer program, characterized in that: When the computer executable instructions or computer program are executed by a processor, the data verification method according to any one of claims 1 to 8 is implemented.

12. A computer program product comprising computer executable instructions or a computer program, characterized in that: When the computer executable instructions or computer program are executed by a processor, the data verification method according to any one of claims 1 to 8 is implemented.

Images